chapter 1 - Bentham Science
chapter 1 - Bentham Science
chapter 1 - Bentham Science
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
FIBER BRAGG GRATING SENSORS: RECENT ADVANCEMENTS,<br />
INDUSTRIAL APPLICATIONS AND MARKET EXPLOITATION<br />
Editors:<br />
Andrea Cusano, Antonello Cutolo and Jacques Albert
eBooks End User License Agreement<br />
Please read this license agreement carefully before using this eBook. Your use of this eBook/<strong>chapter</strong> constitutes your agreement<br />
to the terms and conditions set forth in this License Agreement. <strong>Bentham</strong> <strong>Science</strong> Publishers agrees to grant the user of this<br />
eBook/<strong>chapter</strong>, a non-exclusive, nontransferable license to download and use this eBook/<strong>chapter</strong> under the following terms and<br />
conditions:<br />
1. This eBook/<strong>chapter</strong> may be downloaded and used by one user on one computer. The user may make one back-up copy of this<br />
publication to avoid losing it. The user may not give copies of this publication to others, or make it available for others to copy or<br />
download. For a multi-user license contact permission@bentham.org<br />
2. All rights reserved: All content in this publication is copyrighted and <strong>Bentham</strong> <strong>Science</strong> Publishers own the copyright. You may<br />
not copy, reproduce, modify, remove, delete, augment, add to, publish, transmit, sell, resell, create derivative works from, or in<br />
any way exploit any of this publication’s content, in any form by any means, in whole or in part, without the prior written<br />
permission from <strong>Bentham</strong> <strong>Science</strong> Publishers.<br />
3. The user may print one or more copies/pages of this eBook/<strong>chapter</strong> for their personal use. The user may not print pages from<br />
this eBook/<strong>chapter</strong> or the entire printed eBook/<strong>chapter</strong> for general distribution, for promotion, for creating new works, or for<br />
resale. Specific permission must be obtained from the publisher for such requirements. Requests must be sent to the permissions<br />
department at E-mail: permission@bentham.org<br />
4. The unauthorized use or distribution of copyrighted or other proprietary content is illegal and could subject the purchaser to<br />
substantial money damages. The purchaser will be liable for any damage resulting from misuse of this publication or any<br />
violation of this License Agreement, including any infringement of copyrights or proprietary rights.<br />
Warranty Disclaimer: The publisher does not guarantee that the information in this publication is error-free, or warrants that it<br />
will meet the users’ requirements or that the operation of the publication will be uninterrupted or error-free. This publication is<br />
provided "as is" without warranty of any kind, either express or implied or statutory, including, without limitation, implied<br />
warranties of merchantability and fitness for a particular purpose. The entire risk as to the results and performance of this<br />
publication is assumed by the user. In no event will the publisher be liable for any damages, including, without limitation,<br />
incidental and consequential damages and damages for lost data or profits arising out of the use or inability to use the publication.<br />
The entire liability of the publisher shall be limited to the amount actually paid by the user for the eBook or eBook license<br />
agreement.<br />
Limitation of Liability: Under no circumstances shall <strong>Bentham</strong> <strong>Science</strong> Publishers, its staff, editors and authors, be liable for<br />
any special or consequential damages that result from the use of, or the inability to use, the materials in this site.<br />
eBook Product Disclaimer: No responsibility is assumed by <strong>Bentham</strong> <strong>Science</strong> Publishers, its staff or members of the editorial<br />
board for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any<br />
use or operation of any methods, products instruction, advertisements or ideas contained in the publication purchased or read by<br />
the user(s). Any dispute will be governed exclusively by the laws of the U.A.E. and will be settled exclusively by the competent<br />
Court at the city of Dubai, U.A.E.<br />
You (the user) acknowledge that you have read this Agreement, and agree to be bound by its terms and conditions.<br />
Permission for Use of Material and Reproduction<br />
Photocopying Information for Users Outside the USA: <strong>Bentham</strong> <strong>Science</strong> Publishers Ltd. grants authorization for individuals<br />
to photocopy copyright material for private research use, on the sole basis that requests for such use are referred directly to the<br />
requestor's local Reproduction Rights Organization (RRO). The copyright fee is US $25.00 per copy per article exclusive of any<br />
charge or fee levied. In order to contact your local RRO, please contact the International Federation of Reproduction Rights<br />
Organisations (IFRRO), Rue du Prince Royal 87, B-I050 Brussels, Belgium; Tel: +32 2 551 08 99; Fax: +32 2 551 08 95; E-mail:<br />
secretariat@ifrro.org; url: www.ifrro.org This authorization does not extend to any other kind of copying by any means, in any<br />
form, and for any purpose other than private research use.<br />
Photocopying Information for Users in the USA: Authorization to photocopy items for internal or personal use, or the internal<br />
or personal use of specific clients, is granted by <strong>Bentham</strong> <strong>Science</strong> Publishers Ltd. for libraries and other users registered with the<br />
Copyright Clearance Center (CCC) Transactional Reporting Services, provided that the appropriate fee of US $25.00 per copy<br />
per <strong>chapter</strong> is paid directly to Copyright Clearance Center, 222 Rosewood Drive, Danvers MA 01923, USA. Refer also to<br />
www.copyright.com
CONTENTS<br />
Foreword i<br />
Preface ii<br />
Contributors iv<br />
CHAPTERS<br />
1. Fiber Bragg Grating Sensors: A Look Back 1<br />
J. Albert<br />
2. Fiber Bragg Gratings: Advances in Fabrication Process and Tools 9<br />
K. Sugden and V. Mezentsev<br />
3. Fiber Bragg Gratings: Analysis and Synthesis Techniques 35<br />
J. Skaar<br />
4. Photonic Bandgap Engineering in FBGs by Post Processing Fabrication Techniques 53<br />
A. Cusano, D. Paladino, A. Cutolo, A. Iadicicco and S. Campopiano<br />
5. Fiber Bragg Grating Interrogation Systems 78<br />
J.L. Santos, L.A. Ferreira and F.M. Araújo<br />
6. Multiplexing Techniques for FBG Sensors 99<br />
M. López-Amo and J.M. López-Higuera<br />
7. Polarization Properties of Fiber Bragg Gratings and Their Application for Transverse Force<br />
Sensing Purposes 116<br />
C. Caucheteur, S. Bette, M.Wuilpart and P. Mégret<br />
8. Fiber Bragg Grating Sensors in Civil Engineering Applications 143<br />
J.P. Ou, Z. Zhou and G. Chen<br />
9. Fiber Bragg Grating Sensors in Aeronautics and Astronautics 171<br />
N. Takeda and Y. Okabe<br />
10. Fiber Bragg Grating Sensors in Energy Applications 185<br />
C.B. Staveley<br />
11. Fiber Bragg Grating Sensors for Railway Systems 197<br />
H.Y. Tam, S.Y. Liu, S.L. Ho and T.K. Ho<br />
12. Fiber Bragg Grating Sensors in Nuclear Environments 218<br />
F. Berghmans and A. Gusarov<br />
13. Fiber Bragg Grating Evanescent Wave Sensors for Chemical and Biological Applications 238<br />
A. Cusano, D. Paladino, A. Cutolo, A. Iadicicco and S. Campopiano<br />
14. Fiber Bragg Grating Sensors in Microstructured Optical Fibers 270<br />
T. Nasilowski<br />
15. Polymer Fiber Bragg Gratings 292<br />
D.J. Webb and K. Kalli<br />
16. Fiber Bragg Grating Sensors: Market Overview and New Perspectives 313<br />
J.W. Miller and A. Méndez<br />
Index 321
FOREWORD<br />
Optical fiber based sensor systems have fascinated the researcher and tantalized the application engineer for over<br />
forty years. The fascination lies within the countless ways through which the technically inclined optical scientist<br />
can cause light guided within an optical fiber to interact with the world outside and the myriad principles through<br />
which these modulation phenomena could be detected. The applications engineer finds the benefits which this<br />
approach offers to be many and intriguing. These include predominantly that the area or point which is to be<br />
measured needs no electrical contact and yet remains in intimate proximity to the parameter of interest. Add this to<br />
ideas like highly multiplexed all electrically passive networks, distributed sensing systems and interrogation ranges<br />
of 10 or more kilometers and a whole new spectrum of applications becomes possible.<br />
Of the multiple techniques which have been explored a few have emerged into reality. The fiber optic gyroscope<br />
now navigates spacecraft and the hydrophone array helps in mapping the seabed. Other techniques are exciting<br />
considerable interest and among these one of the principal contenders for extensive exploration is undoubtedly the<br />
Fiber Bragg grating – the subject of this book.<br />
The Bragg grating is intrinsically a simple device. A periodic structure is written into the core of an optical fiber and<br />
this periodic structure will reflect specific optical wave length dependent on the periodicity. Vary the periodicity,<br />
vary the wavelength. Since this period depends upon environmental temperature and externally applied strains and<br />
pressures we have the basis for a simple sensor which is easily and unambiguously interrogated. Furthermore long<br />
strings of these sensors each operating at a different wave length can be written into a single fiber giving an array<br />
technology linked through just a single fiber giving an array technology which can extend over kilometers.<br />
The basic concepts for the fiber grating go back more than twenty years from early exploratory work undertaken by<br />
Ken Hill and his colleagues in Canada. Since then the grating has attracted significant attention from scientists and<br />
application engineers alike and has demonstrated its applicability in areas ranging from monitoring large and<br />
complex highway bridges to measuring stresses in teeth.<br />
This book explores every nuance of this important sensing concept. The contributors are, without exception,<br />
internationally recognized experts in their field and present the diversity of techniques, applications and perceptions<br />
through which the potential offered by the Bragg grating can be appreciated. This ranges from basic fabrication<br />
processes through the design of the details of the grating structure itself into multiplexing techniques and<br />
technologies, the interrogating electro-optic system and the numerous interfaces into application. Market prospects<br />
and technology roadmaps also feature to complete the overall picture.<br />
The book promises to be an invaluable addition to the sensing engineer’s library and will provide essential<br />
background to stimulate the future prospects for this important and intriguing technology.<br />
i<br />
Brian Culshaw<br />
University of Strathclyde<br />
United Kingdom
ii<br />
PREFACE<br />
The field of fiber optics has undergone tremendous growth over the past 40 years. Initially conceived as a medium<br />
to carry light and images for medical endoscopic applications, optical fibers were later proposed in the mid 1960’s<br />
as an information-carrying medium for telecommunication applications. The outstanding success of this concept is<br />
embodied in millions of miles of telecommunications fiber that have spanned the earth, the seas, and utterly<br />
transformed our capacity to communicate and the means by which we do it. The award of the 2009 Nobel Prize in<br />
physics to C. K. Kao, who first proposed the use of optical fibers for data communication, is the crowning jewel on<br />
this fantastic story. Among the reasons why optical fibers are such an attractive communication medium are their<br />
low loss, high bandwidth, electromagnetic interference immunity, small size, light weight, safety, relatively low<br />
cost, low maintenance, etc.<br />
As optical fibers consolidated their dominant position in the telecommunications industry, the technology and<br />
commercial markets both matured, especially from the late 1980s onward. In parallel over the same period, several<br />
groups around the world have been carrying out research to exploit some of the key benefits of optical fibers for<br />
another very important application: sensing. Initially, fiber sensors were laboratory curiosities and simple proof-ofconcept<br />
demonstrations. However, more and more, optical fiber technologies are making an impact and have<br />
achieved commercial respectability and viability in industrial sensing, bio-medical laser delivery systems, aerospace<br />
and military gyroscopes, as well as automotive lighting & control – to name just a few – and span applications as<br />
diverse as oil-well downhole pressure/temperature sensors and intra-aortic catheters. This transition has taken the<br />
better part of the last 20 years and reached the point where fiber sensors enjoy widespread use for structural sensing<br />
and monitoring applications in civil engineering, aerospace, marine, oil & gas, composites, smart structures, biomedical<br />
devices, electric power industry and many others. Of course all these advances have resulted from the<br />
development of a complete understanding of the physical and chemical transducing mechanisms and of the<br />
appropriate sensor interrogation systems and technologies. For instance, there is a variety of commercial discrete<br />
sensors based on Fabry-Perot cavities and Fiber Bragg Gratings (FBGs), as well as distributed sensors based on<br />
Raman and Brillouin scattering methods, all readily available along with relevant interrogation instruments.<br />
Amongst all of these, FBG based sensors – more than any other particular sensor type – have become widely known,<br />
researched, and popular within and outside of the photonics community, leading to a significant increase in their<br />
utilization and commercial impact. Given the capability of FBGs to measure a multitude of parameters such as<br />
strain, temperature, pressure, chemical and biological agents and many others, coupled with their design flexibility<br />
for single point or multi-point sensing arrays and their relative low cost, FBGs are ideal for a multitude of sensing<br />
applications in many fields and industries.<br />
Since 2000, more than twenty companies have been active in the FBGs sensor market. In 2000, Micron Optics was<br />
able to launch in the market the first line of advanced FBG interrogators, while in 2003 LxSix (now LxDATA) and<br />
Sabeus launched the first reel-to-reel production of high reliability FBGs arrays. Finally, in May of 2007 HBM – the<br />
world’s largest supplier of strain sensing systems – began offering optical strain gages and interrogators based on<br />
FBG technology. This marked the first time that a conventional foil strain gage manufacturer adopted and embraced<br />
FBGs as an essential part of their product portfolio. A broad and intense commercial pull is expected to follow from<br />
this breakthrough.<br />
On the research side, many groups around the world are hard at work on new fabrication methods, FBG reliability,<br />
new device designs, and new data processing techniques. For instance, FBGs can now be fabricated in many types<br />
of glasses using ultrafast IR writing and new post processing techniques have been developed to customize device<br />
spectra for enhanced sensitivity in specific applications. New industrial applications have opened up in many<br />
strategic sectors, especially in the case of chemical or biological sensing. The prospects of using polymer optical<br />
fibers (POF) in sensing applications is expected to lead to the development of POF FBGs for inexpensive, simple
and low-cost disposable platforms (in the automotive industry for instance). Similarly, microstructured optical fibers<br />
are expected to have a major impact in the development of new chemical and biological systems based on<br />
optofluidics as well as active and passive microfluidics.<br />
All of these advances have created a need for a comprehensive overview of the FBG Sensor Technology. The book<br />
that we have put together covers every aspect of this technology, starting all the way back from the fortuitous<br />
discovery of Hill in 1978. The remainder of the book details the enormous efforts carried out by both the scientific<br />
and industrial communities over the last two decades to bring the field to its current level of success and its promise<br />
of further advances. The different <strong>chapter</strong>s will illustrate and describe:<br />
- new fabrication methods and advancements in photosensitivity;<br />
- new post processing techniques for spectral tailoring;<br />
- new applications in strategic industrial sectors starting from civil, aeronautic energy up to nuclear and<br />
extreme environments;<br />
- exciting new developments in the field of chemical and biological sensing;<br />
- new perspectives involving plastic and micro-structured fibers;<br />
- a clear identification of the market situation and a forecast for the next decade.<br />
In summary, our goal was to provide a complete, very exciting source of information on FBG sensors for the huge<br />
community of people now involved in the field, including both the scientific and the industrial sectors. Of special<br />
interest is the fact that the book also contains a very wide range of topics and innovative concepts that have not yet<br />
been presented as comprehensive reviews elsewhere. Finally, it is important to remark that each of the <strong>chapter</strong>s has<br />
been written by recognized experts in their fields, either from academia or the private sector according to the topic<br />
of the <strong>chapter</strong>. It has been a pleasure for the editors to work with all these authors as they have worked very hard to<br />
provide us with authoritative, clear and elaborate reviews of their specialties. We hope that the readers of the book<br />
will share our pleasure in reading these reviews, and maybe learn a thing or two along the way to further their<br />
research or business. That, surely, would be our most satisfying reward.<br />
iii<br />
Andrea Cusano & Antonello Cutolo<br />
University of Sannio, Italy<br />
Jacques Albert<br />
University of Carleton, Canada
iv<br />
CONTRIBUTORS<br />
Jacques Albert<br />
Department of Electronics, Carleton University, 1125 Colonel By Drive, Ottawa, ON, Canada K1S 5B6<br />
Francisco M. Araújo<br />
INESC Porto, Rua do Campo Alegre 687, 4169-007 Porto, Portugal<br />
FiberSensing S.A., Rua Vasconcelos Costa 277, 4470-640 Maia, Portugal<br />
Francis Berghmans<br />
Department of Applied Physics and Photonics, Vrije Universiteit Brussel, Pleinlaan 2, 1050 Brussels, Belgium<br />
SCK·CEN Belgian Nuclear Research Center, Boeretang 200, 2400 Mol, Belgium<br />
Sébastien Bette<br />
Electromagnetism and Telecommunications Department, Faculté Polytechnique, Université de Mons, Boulevard<br />
Dolez 31, 7000 Mons, Belgium<br />
Stefania Campopiano<br />
Department of Technology, University of Naples “Parthenope”, Centro Direzionale di Napoli Isola C4, 80143<br />
Napoli, Italy<br />
Christophe Caucheteur<br />
Electromagnetism and Telecommunications Department, Faculté Polytechnique, Université de Mons, Boulevard<br />
Dolez 31, 7000 Mons, Belgium<br />
Genda Chen<br />
Center for Infrastructure Engineering Studies, Missouri University of <strong>Science</strong> and Technology, Rolla, MO 65409-<br />
0710, USA<br />
Andrea Cusano<br />
Optoelectronic Division – Engineering Department, University of Sannio, Corso Garibaldi 107, 82100 Benevento,<br />
Italy<br />
Antonello Cutolo<br />
Optoelectronic Division – Engineering Department, University of Sannio, Corso Garibaldi 107, 82100 Benevento,<br />
Italy<br />
Luís A. Ferreira<br />
INESC Porto, Rua do Campo Alegre 687, 4169-007 Porto, Portugal<br />
FiberSensing S.A., Rua Vasconcelos Costa 277, 4470-640 Maia, Portugal<br />
Andrei Gusarov<br />
SCK·CEN Belgian Nuclear Research Center, Boeretang 200, 2400 Mol, Belgium<br />
Siu-lau Ho<br />
Photonics Research Centre, Department of Electrical Engineering, The Hong Kong Polytechnic University, Hung<br />
Hom, Kowloon, Hong Kong SAR, China<br />
Tin-kin Ho<br />
Photonics Research Centre, Department of Electrical Engineering, The Hong Kong Polytechnic University, Hung<br />
Hom, Kowloon, Hong Kong SAR, China<br />
Agostino Iadicicco<br />
Department of Technology, University of Naples “Parthenope”, Centro Direzionale di Napoli Isola C4, 80143<br />
Napoli, Italy
Kyriacos Kalli<br />
Nanophotonics Research Laboratory, Cyprus University of Technology, Cyprus<br />
Shun-yee Liu<br />
Photonics Research Centre, Department of Electrical Engineering, The Hong Kong Polytechnic University, Hung<br />
Hom, Kowloon, Hong Kong SAR, China<br />
Manuel López-Amo<br />
Grupo de Comunicaciones Ópticas, Universidad Pública de Navarra, Campus de Arrosdía, E-31006-Pamplona,<br />
Spain<br />
Jóse M. López-Higuera<br />
Grupo de Ingeniería Fotónica, Universidad de Cantabria, ETSII y Telecomunicación, Avda. de los Castros s/n E-<br />
39005 Santander, Spain<br />
Patrice Mégret<br />
Electromagnetism and Telecommunications Department, Faculté Polytechnique, Université de Mons, Boulevard<br />
Dolez 31, 7000 Mons, Belgium<br />
Alexis Méndez<br />
MCH Engineering LLC, 1728 Clinton Avenue, Alameda, CA 94501, USA<br />
Vladimir Mezentsev<br />
Photonics Research Group, Aston University, Aston Triangle, Birmingham B4 7ET, United Kingdom<br />
Jeff W. Miller<br />
Micron Optics Inc., 1852 Century Place, Atlanta, GA 30345, USA<br />
Tomasz Nasilowski<br />
Department of Applied Physics and Photonics, Faculty of Engineering, Vrije Universiteit Brussel, Pleinlaan 2,<br />
building F, B-1050 Brussels, Belgium<br />
Yoji Okabe<br />
Department of Mechanical and Biofunctional Systems, Institute of Industrial <strong>Science</strong>, The University of Tokyo, 4-6-<br />
1 Komaba, Meguro-ku, Tokyo 153-8505, Japan<br />
Jinping Ou<br />
School of Civil Engineering, Harbin Institute of Technology, Harbin, Heilongjiang, 150090, P.R. China<br />
School of Civil and Hydraulic Engineering, Dalian University of Technology, Dalian, 116024, P.R. China<br />
Domenico Paladino<br />
Optoelectronic Division – Engineering Department, University of Sannio, Corso Garibaldi 107, 82100 Benevento,<br />
Italy<br />
José L. Santos<br />
INESC Porto, Rua do Campo Alegre 687, 4169-007 Porto, Portugal<br />
Faculdade de Ciências da Universidade do Porto, Rua do Campo Alegre 687, 4169-007 Porto, Portugal<br />
Johannes Skaar<br />
Department of Electronics and Telecommunications, Norwegian University of <strong>Science</strong> and Technology, Trondheim,<br />
Norway; University Graduate Center, Kjeller, Norway<br />
Christopher B. Staveley<br />
Smart Fibres, Limited, 12 The Courtyard, Eastern Road, Bracknell, RG12 2XB, United Kingdom<br />
v
vi<br />
Kate Sugden<br />
Photonics Research Group, Aston University, Aston Triangle, Birmingham B4 7ET, United Kingdom<br />
Hwa-yaw Tam<br />
Photonics Research Centre, Department of Electrical Engineering, The Hong Kong Polytechnic University, Hung<br />
Hom, Kowloon, Hong Kong SAR, China<br />
Nobuo Takeda<br />
Department of Advanced Energy, Graduate School of Frontier <strong>Science</strong>s, The University of Tokyo, Mail Box 302, 5-<br />
1-5 Kashiwanoha, Kashiwa-shi, Chiba 277-8561, Japan<br />
David J. Webb<br />
Photonics Research Group, Aston University, United Kingdom<br />
Marc Wuilpart<br />
Electromagnetism and Telecommunications Department, Faculté Polytechnique, Université de Mons, Boulevard<br />
Dolez 31, 7000 Mons, Belgium<br />
Zhi Zhou<br />
School of Civil Engineering, Harbin Institute of Technology, Harbin, Heilongjiang, 150090, P.R. China<br />
Center for Infrastructure Engineering Studies, Missouri University of <strong>Science</strong> and Technology, Rolla, MO 65409-<br />
0710, USA
Fiber Bragg Grating Sensors: Recent Advancements, Industrial Applications and Market Exploitation, 2011, 1-8 1<br />
Fiber Bragg Grating Sensors: A Look Back<br />
Jacques Albert*<br />
Department of Electronics, Carleton University, 1125 Colonel by Drive, Ottawa, ON, Canada K1S 5B6<br />
Andrea Cusano, Antonello Cutolo and Jacques Albert (Eds)<br />
All rights reserved - © 2011 <strong>Bentham</strong> <strong>Science</strong> Publishers Ltd.<br />
CHAPTER 1<br />
Abstract: The development of Fiber Bragg Gratings (FBGs) is closely associated with the field optical fiber<br />
sensors. This <strong>chapter</strong> presents a personal overview of the history of FBGs, with particular emphasis on the<br />
interrelation and impact of FBGs with sensing. The major milestones for FBG-based sensing are identified, from<br />
the very discovery of fiber gratings to current developments that are described in full detail in the following<br />
<strong>chapter</strong>s of this book.<br />
THE DISCOVERY YEARS<br />
Fiber Bragg gratings (FBGs) were discovered in 1978 by Ken Hill at the Communications Research Center in<br />
Ottawa [1]. It is one of those rare discoveries that are undoubtedly associated with an individual, to the point that<br />
FBGs were called “Hill” gratings in the early days by several researchers (a term that was later reserved to gratings<br />
made using the same technique as the original discovery, i.e. internal writing) [2]. I do not intend to write a history<br />
of FBGs in this <strong>chapter</strong>, but rather to relate how this history in closely associated with optical fiber sensors in<br />
general. And this association starts from the very beginning. One of the ways that were used to measure the strength<br />
and line widths of these gratings was to measure the transmission of the grating with monochromatic light and a<br />
detector while stretching or heating the grating. The grating spectrum was then recorded as a function of strain or<br />
temperature, from which the grating period could be calculated. Therefore the first FBGs were also the first FBG<br />
sensors! This was out of necessity. In 1977 it was very difficult to couple high brightness broadband light into the<br />
core of an optical fiber and to gather enough light in transmission or reflection to measure spectra with<br />
monochromators (even single mode couplers did not exist, only multimode ones had just been invented, also in<br />
Hill’s group [3]). Furthering the problem was that writing process of the first Hill gratings was internal to the fiber<br />
core, arising from a standing wave pattern set up by the Fresnel reflection at the end of the fiber. The grating period<br />
was thus fixed to reflect only light at the writing wavelength. The fibers used were also very lossy (gratings were<br />
written directly with blue-green light at relatively low intensity because these early fibers were full of defects and<br />
color centers). These issues limited severely the applications of these gratings but it was widely recognized that if<br />
gratings of arbitrary periods could be written in “good” fibers the potential for new devices for processing and<br />
measuring guided light was enormous. In the years that followed the mechanism of the formation of FBGs was more<br />
or less elucidated as a two-photon process due to the interaction of ultraviolet (UV) light with defects states just<br />
below the bandgap of the doped silica forming the core [4].<br />
I am not quite sure what led Gerry Meltz and colleagues from the United Technologies Research Center to achieve<br />
the next breakthrough in FBG technology, i.e. a practical way to use ultraviolet light directly to write FBGs.<br />
However, I must side track for a moment to mention a related activity in those days: the generation of second<br />
harmonic light from “prepared” fibers, a discovery by Walter Margulis and Ulf Osterberg [5]. In this process, intense<br />
pulses of light propagating down a single mode fiber for a certain amount of time eventually cause a material<br />
modification that leads to the formation of a second-order nonlinearity in the glass (a normally impossible feature in a<br />
material with inversion symmetry, such as glass). Moreover, the second order nonlinearity thus formed is<br />
automatically phase matched to the light that caused it because the induced effect is periodic, i.e. a χ 2 grating is<br />
formed inside the fiber [6]. Basically, they saw green light coming out of fibers in which only infrared light was<br />
launched. This process becomes significantly more efficient if the fiber is “prepared” by sending a weak probe signal<br />
at the second harmonic alongside the pump beam. The reason I am mentioning this here is that the FBG process and<br />
the self-organized second harmonic generation (SHG) effects were long thought to arise from the same physical<br />
mechanisms and that there was a strong overlap between the research communities. It is likely that advances in the<br />
*Address correspondence to this author Professor Jacques Albert: Department of Electronics, Carleton University, 1125 Colonel By Drive,<br />
Ottawa, ON, Canada K1S 5B6; Tel: +613 520-2600; Fax: +613 520-5708; Email: Jacques_albert@carleton.ca
2 Fiber Bragg Grating Sensors Jacques Albert<br />
SHG field helped several advances in FBG science and technology. In fact, the first scientific workshops and<br />
meetings on these two topics were held jointly (more on this later).<br />
FIBER BRAGG GRATINGS BECOME A REALITY<br />
It was the 1989 breakthrough publication by Meltz, Morey and Glenn actually made FBGs practical [7-8]. Since<br />
FBGs were basically permanent holograms written in the core of an optical fiber, they decided to try holographic<br />
techniques to write FBGs from the outside of the fiber. Since it had been determined by then that UV light was the<br />
source of the FBG writing process, two questions had to be addressed: would the cladding of the fiber be<br />
transparent enough to the UV light to reach the core and was there a light source with enough power and spatial<br />
coherence to form interference fringes of sufficient contrast and intensity to write FBGs? Luckily, the cladding of<br />
most optical fibers (even those at the time) is made of very pure undoped synthetic silica, whose absorption edge<br />
lies beyond 180 nm in the UV. Since the most prominent defect band thought responsible for the formation of<br />
FBGs is a germanium oxygen deficient color center with absorption near 242 nm, the cladding would not pose a<br />
problem. For the UV source there was not much to choose from in the late 1980s. They had to use a rather intricate<br />
system made up using a tunable excimer-pumped dye laser, operated at a wavelength in the range of 486-500 nm,<br />
along with a frequency-doubling crystal to provide UV light near 242 nm adequate coherence. The advantage of that<br />
particular system was that they could tune the laser wavelength and thus prove that the FBG writing process was more<br />
efficient near 242 nm. Of course, forming the interference fringe pattern from the outside also allowed the writing of<br />
gratings with almost arbitrary periods, and hence to make wavelength filters and sensors for practical wavelength<br />
bands (near 800, 1300, and 1500 nm for instance). The same group also demonstrated very early on the formation of<br />
tilted gratings that were used then to tap some of the light from the core of the fiber to make it radiate outside [9].<br />
But back to sensors! The FBG is essentially a wavelength filter, bandpass in reflection and band reject in<br />
transmission. Since the period of the grating necessary to reflect core guided light in glass fibers is of the order of<br />
half a micrometer, even millimeter long gratings have Q-factors (number of grating periods / grating length) of at<br />
least 2000, resulting in pass bands that are of the order of 1 nm and less for light at visible and near IR wavelengths.<br />
Furthermore the combined thermo-optic coefficient of the reflection wavelength (hereafter called the “Bragg”<br />
wavelength) was soon calculated and measured to be near 10 pm/°C, arising from a combination of the (weak)<br />
coefficient of thermal expansion of silica glass and the relatively stronger thermo-optic coefficient of the refractive<br />
index (+10 -5 /°C). Similarly for strain, whereas fibers can be stretched without breaking by as much as 1%, with a<br />
corresponding change in the period of the FBG [7, 10]. Therefore with suitable instrumentation to launch light with<br />
at least a few nanometers of bandwidth and to measure the reflected optical spectrum (fused fiber couplers for single<br />
mode fibers and optical spectrum analyzers had been invented and mass produced by the late 1980s!), a fiber with<br />
an embedded FBG could be used to measure temperature or strain in very small spaces, remotely from relatively<br />
large distances, and in areas where electrical devices were not welcomed. Interestingly, the original patent on side<br />
writing, filed in October of 1986, explicitly mentions temperature and strain sensing as claimed applications and<br />
provides the equations to calculate the wavelength shifts of FBGs as a function of these two parameters. Knowing<br />
that Meltz’s group at UTC was concerned with the development of fiber optic (strain and temperature) sensors from<br />
at least the mid 1970s (US patents from 1981), it is fair to say that the need for better fiber sensors was the main<br />
reason FBGs were developed in the first place. The attractive sensing features of FBGs were immediately<br />
recognized by the photonics community and led to an incredible flurry of activity worldwide from the early 1990s<br />
onward to this day. Another aspect of optical sensing where FBGs were recognized to have great potential was in<br />
the development of narrowband, high coherence lasers through the use of FBGs as mirrors for fiber laser cavities<br />
and also as pigtailed output couplers for semiconductor lasers [11-13]. These efforts were made in parallel with<br />
similar research activities to find applications for FBGs in other fields, especially telecommunications. The 1990s<br />
were the golden age of optical telecommunications research as the growth of the demand for long distance<br />
bandwidth was fueled by the birth of the internet and fed by several breakthroughs (low loss single mode fibers and<br />
erbium-doped fiber optical amplifiers). Just to give a flavor of the enthusiasm associated with FBGs in that field, in<br />
2000, near the peak of what has been called the “telecom bubble” the market forecast for FBGs in telecom alone was<br />
predicted by some analysts to reach 1 B$/year within the first half of the decade. At grating prices between 10 $ and<br />
1000 $ depending on the application, that would have been a lot of gratings.<br />
Back to sensors again! It is not so well known as I write this that a large fraction of the early research on FBGs was<br />
for sensing applications. The Optical Fiber Sensors conference had had its first occurrence in 1983, and the field was
Fiber Bragg Grating Sensors: A Look Back Fiber Bragg Grating Sensors 3<br />
certainly as exciting during the ensuing years as the optical communications field. In parallel, the FBG community<br />
organized itself and pioneers of the field, Francois Ouellette and Raman Kashyap organized the first two<br />
international conferences or workshops on the topic in 1991 and 1993 in Quebec City [14-15]. It is actually<br />
fascinating to browse through the proceedings of these conferences to see who was involved then and what they<br />
were working on. Surprisingly, some of the research topics discussed then are still very relevant today, such as the<br />
photosensitivity of rare earth doped glasses and the UV-induced birefringence of FBGs. As indicated earlier, these<br />
two conferences, and also the following one that was organized as an Optical Society of America Topical meeting in<br />
Portland, Oregon in 1995 [16], also included what was known then as “self-organization” and quadratic<br />
nonlinearities in optical fibers and waveguides. For the most part these were concerned with light-induced<br />
modifications of glassy materials to induce a second order nonlinearity and hence fabricate devices for second<br />
harmonic generation and for electro-optic modulation in glass fibers and waveguides. Achieving these goals with<br />
enough efficiency for practical devices would have had a tremendous impact in photonics, possibly as important as<br />
the invention of the EDFA.<br />
A MOST NOTABLE YEAR: 1993<br />
The year 1993 is possibly the single most important year in the history of FBGs, with the discovery of two more<br />
enabling technologies on the path to wide acceptability and opening up real life applications: the phase mask and<br />
hydrogen-loading.<br />
First, looking back to 1991, and following the publication of the side-writing paper by Meltz et al., Ken Hill’s group<br />
at the Communications Research Center resumed an intense research effort in FBGs. They had both significant<br />
hurdles to overcome and also some distinct advantages over other groups that jumped in this field. The hurdle was<br />
that the laser source that was used for side writing at the time was by no means easy to put together or to operate.<br />
However, Hill’s group had an old excimer laser available and they were actively working on SHG and selforganization<br />
at the time, using intense pulsed light to fabricate point-by-point mode converter gratings in fibers (the<br />
ancestors of Long Period Gratings), using side exposure [17]. Of course they had intimate knowledge of every<br />
aspect of FBG modeling and applications. It is said that necessity is the mother of invention. The lack of coherence<br />
of the excimer laser turned out to be the main reason Ken Hill was driven to look for a way to generate an<br />
interference pattern of UV light without an interferometer. This is how the phase mask came about [18]: using a<br />
diffractive optical element to generate two light beams from a single one, at such an angle between the beams that<br />
the resulting interference pattern has the necessary periodicity to print a FBG in a fiber. Because the phase mask can<br />
be placed so close to the fiber (within microns of the fiber surface) the two interfering beams generated by the phase<br />
mask need only to have spatial coherence of the order of a few tens of microns for their interference pattern to have<br />
sufficient visibility to write the gratings. Relatively simple calculations and tests confirmed all the necessary features<br />
of the phase mask and a supplier was found to fabricate them. It is worth mentioning that the invention of the phase<br />
mask was contested in court by another group who had a similar idea for writing FBGs, but with using an amplitude<br />
mask instead. Because it is possible to null the zeroth-order transmission with a phase mask, the FBG writing<br />
efficiency is greatly optimized and amplitude masks are no longer used for FBGs. The second crucial development<br />
in 1993 was the discovery that the diffusion of molecular hydrogen gas in optical fibers at relatively high pressure<br />
and for a duration long enough to saturate the fiber led to an extremely large enhancement of the photosensitivity of<br />
the fibers to UV light [19] (it is interesting to note that a similar enhancement had been observed a few years earlier<br />
for SHG experiments in hydrogen-loaded fiber by Francois Ouellette, a member of Ken Hill’s group at the time<br />
[20]). Instead of FBGs with grating modulation amplitudes barely reaching 10 -4 without hydrogen, modulation<br />
amplitudes approaching 10 -2 were achieved. This improvement by two orders of magnitude not only increases the<br />
reflectivity of gratings, it is actually crucial in making some devices possible at all, such as FBG-based gain<br />
equalizing filters for which the whole dynamic range of the photosensitive process is necessary to achieve the<br />
product specifications.<br />
It was also at that time that “photosensitive” fibers were invented, i.e. fibers with dopant materials and profiles<br />
customized to enhance the formation of permanent refractive index changes under UV irradiation [21]. One of the<br />
driving forces for such developments was the perceived need to avoid hydrogen-loading, since that technique<br />
provided as much photosensitivity as needed in low cost, high quality standard telecommunications fiber. The<br />
problem with hydrogen-loading is that there is a safety issue related to handling this gas at high pressure and that the
Fiber Bragg Grating Sensors: Recent Advancements, Industrial Applications and Market Exploitation, 2011, 9-34 9<br />
Fiber Bragg Gratings: Advances in Fabrication Process and Tools<br />
Kate Sugden* and Vladimir Mezentsev<br />
Photonics Research Group, Aston University, Aston Triangle, Birmingham B4 7ET, United Kingdom<br />
Andrea Cusano, Antonello Cutolo and Jacques Albert (Eds)<br />
All rights reserved - © 2011 <strong>Bentham</strong> <strong>Science</strong> Publishers Ltd.<br />
CHAPTER 2<br />
Abstract: Successful commercialization of a technology such as iber Bragg ratings requires the ability to<br />
manufacture devices repeatably, quickly and at low cost. Although the first report of photorefractive gratings was<br />
in 1978 it was not until 1993, when phase mask fabrication was demonstrated, that this became feasible. More<br />
recently, draw tower fabrication on a production level and grating writing through the polymer jacket have been<br />
realized; both important developments since they preserve the intrinsic strength of the fiber. Potentially the most<br />
significant recent development has been femtosecond laser inscription of gratings. Although not yet a commercial<br />
technology, it provides the means of writing multiple gratings in the optical core providing directional sensing<br />
capability in a single fiber. Femtosecond processing can also be used to machine the fiber to produce micronscale<br />
slots and holes enhancing the interaction between the light in the core and the surrounding medium.<br />
INTRODUCTION<br />
When Hill discovered the formation of photorefractive gratings in 1978 few would have imagined the commercial<br />
promise this technology would eventually hold [1-2]. The Hill gratings were fabricated using the interference<br />
between counter propagating laser beams from an Argon ion laser operating at 488 nm. The gratings were long,<br />
which meant they were of extremely narrow bandwidth, and they reflected the same wavelength of light that was<br />
used to fabricate them. These two factors limited their practical applications in sensing.<br />
An approach yielding somewhat more useful devices was demonstrated in 1986 [3], where periodic filters were<br />
written using a photoresist layer and surface relief etching. The cladding layer of the fiber was polished down to<br />
give a flat surface which was overlaid with the photoresist and then exposed and etched. Strong gratings with greater<br />
than 90% reflectivity were made. However, although similar to conventional fiber Bragg gratings in reflection this<br />
process induced large transmission losses due to modal out-coupling which ultimately limited the number of<br />
gratings that could be interrogated on a single piece of fiber. In addition, because of the multistage process, this<br />
approach was never going to be one of low cost volume fabrication.<br />
What was arguably the most significant development came in 1989 when a bulk optic interferometer was used to<br />
directly write gratings into the fiber using side illumination with a UV laser [4]. Here the writing method itself was<br />
not that novel but the realization that the photorefractive change with ultraviolet light was enough to allow direct<br />
side writing of devices was a key development. This significant advance heralded a period of rapid development in<br />
the field of fiber Bragg gratings and, combined with the demonstration of phase mask writing technique in 1993 [5-<br />
6], provided the foundation of most of the work in this area today.<br />
This <strong>chapter</strong> concentrates on the areas of the fabrication of fiber Bragg gratings that are most relevant to the sensing<br />
field today. It first summarizes the background, including photosensitivity, then looks at the different grating types that<br />
can be fabricated, grating stability and choice of fiber and laser sources. It then reviews the more practical schemes<br />
proposed for fabricating fiber Bragg gratings, including the use of interferometers, phase masks, phase mask<br />
interferometers, direct writing, and higher order masks. Draw tower fabrication, writing through the coating and<br />
packaging are also discussed in this section. The final section of this <strong>chapter</strong> is given over to the fabrication of fiber<br />
Bragg gratings with femtosecond lasers which is one of the most exciting developments in the field in recent years.<br />
*Address correspondence to this author Dr. Kate Sugden: Photonics Research Group, Aston University, Aston Triangle, Birmingham B4<br />
7ET, United Kingdom; Tel: +44 121 204 3498; Fax: +44 121 204; Email: k.sugden@aston.ac.uk
10 Fiber Bragg Grating Sensors Sugden and Mezentsev<br />
BACKGROUND<br />
Photosensitivity<br />
Before discussing the techniques for fabricating fiber Bragg gratings it is important to understand the underlying<br />
mechanisms. The term ‘photosensitivity’ is used to describe the permanent change in the refractive index of a<br />
material by exposure to light. It is this photosensitivity effect that is exploited in the fabrication of fiber Bragg<br />
gratings since it enables a variation in laser light intensity to be recorded directly into the fiber as a refractive index<br />
variation. The result of this is seen in (Fig. 1) which shows a microscope image of a number of grating fringes<br />
inscribed in the core of a single mode optical fiber.<br />
Figure 1: Microscope image of fiber Bragg grating fringes written in the core of a fiber [7].<br />
The majority of commercial gratings to date have been written using ultraviolet (UV) lasers in a single photon<br />
excitation of germanosilicate based defects. Hill’s early demonstration of photosensitivity using a 488 nm laser to<br />
produce Hill gratings relied on a two-photon process [8] and was related to an absorption band around 240 nm.<br />
In the single photon regime the ultraviolet radiation interacts with various defect centers in the fiber [9-10]. The<br />
exposure initiates a bleaching in the absorption band at 240 nm and the creation of other absorption bands that lead<br />
to a refractive index change that can be described using the Kramers-Kronig relationship [11]. The photosensitivity<br />
in germanosilicate fibers is generally proportional to the concentration of the Ge dopant.<br />
It is possible to produce significantly larger refractive index changes by ‘hydrogen loading’ the fiber [12]. Hydrogen<br />
loading involves soaking the fiber in a high pressure hydrogen environment either at room temperature for 1-2<br />
weeks or at an elevated temperature for a shorter time. For example, at 80°C the fiber only requires around 24 hours<br />
to absorb sufficient hydrogen.<br />
By studying the growth dynamics of gratings in a variety of different fibers, laser powers and wavelengths a number<br />
of distinct gratings types have been identified. The distinctions between them depend on the exposure conditions,<br />
the type of induced changes within the glass, and the final grating characteristics.<br />
The technology developed largely due to the push of the telecommunications market which had requirements for<br />
long lifetimes and high stability but where devices operate in relatively benign environments below 100°C. Since<br />
sensing gratings are often used in harsh environments such as in oil and gas applications, careful consideration needs<br />
to be given to the temperatures the gratings will encounter over their lifetime. Tests must carried out on the thermal<br />
decay rate of gratings made under the same fabrication conditions as the gratings to be deployed (power, exposure<br />
time, hydrogenation conditions, fiber type) before setting up a successful anneal regime since all of these things<br />
affect the decay rate and long-term stability.<br />
Grating Types<br />
A number of different grating types have been identified to date, however initially the names given to these different<br />
types were somewhat misleading. A new classification system was recently proposed [13] which groups grating<br />
types according to the similarity in growth mechanisms and this system is followed here.
Fiber Bragg Gratings: Advances in Fabrication Process and Tools Fiber Bragg Grating Sensors 11<br />
There are three basic types of grating associated with UV inscription and these are known as Type I, Type II and<br />
Regenerated. Type I gratings are associated with refractive index changes that occur below the damage threshold of<br />
glass whereas Type II gratings are associated with changes in refractive index above the damage threshold of glass.<br />
Each type is summarized below. There are several Type I sub-classes that are identified according to the writing<br />
conditions and properties of the gratings.<br />
(Fig. 2) shows three different types of gratings written with the same period phase mask. The three gratings reflect<br />
light at different wavelengths. This difference in the reflected wavelength occurs because the refractive index<br />
change produced by each process is associated with a different mechanism and has a different value.<br />
Figure 2: The spectra of Type I (blue), Type IHp (red), and Type In (green) gratings written in B/Ge fiber [14].<br />
Type I<br />
Type I – These are currently the most common gratings used and often referred to as standard gratings. These<br />
gratings can be fabricated in most types of germanosilicate fiber. In the reflected spectrum there is negligible light<br />
lost to the cladding modes or by absorption. The refractive index change associated with this process is usually<br />
(although not in the case of Type In) positive i.e. the average refractive index in the exposed area is higher than the<br />
unexposed core.<br />
The underlying mechanism, at least for small refractive index changes, is a one-photon absorption effect [5]. The<br />
process works by exciting defect centers where the energy levels correspond to oxygen deficiency centre (ODC)<br />
absorption bands at wavelengths around 244 nm and 320 nm. These same bands can also be excited by multi-photon<br />
processes using femtosecond lasers, this will be discussed later in the <strong>chapter</strong>. For large refractive index changes it is<br />
likely that some type of densification change is also involved [15].<br />
Erdogan showed that Type I Bragg gratings exhibited a temperature dependent decay [16]. It was shown that the<br />
decay in the reflectivity could be characterized using a power law. Immediately after the inscription the decay rate<br />
was at its maximum and over time this decay rate decreased. The starting rate of decay was dependent on the<br />
surrounding temperature. This decay is due to the thermal depopulation of the trapped excited states created during<br />
the UV irradiation. Elevating the temperature gives the carriers in the shallowest traps enough energy to escape and<br />
return to the ground state. The carriers remaining are associated with more stable traps, however if the temperature<br />
is steadily increased progressively more of these will decay resulting in a further decrease in the grating strength.<br />
Since it is the most unstable carriers that decay at low temperature Bragg gratings can be annealed at an elevated<br />
temperature prior to use and this removes the least stable part of the induced refractive index change. Post-anneal the<br />
grating exhibits a much greater level of stability over the longer term. This annealing process can also be used to<br />
stabilize other Type I grating sub-types although for each type the related decay properties must be individually<br />
characterized.
Fiber Bragg Grating Sensors: Recent Advancements, Industrial Applications and Market Exploitation, 2011, 35-52 35<br />
Fiber Bragg Gratings: Analysis and Synthesis Techniques<br />
Johannes Skaar*<br />
Andrea Cusano, Antonello Cutolo and Jacques Albert (Eds)<br />
All rights reserved - © 2011 <strong>Bentham</strong> <strong>Science</strong> Publishers Ltd.<br />
CHAPTER 3<br />
Department of Electronics and Telecommunications, Norwegian University of <strong>Science</strong> and Technology, Trondheim,<br />
Norway; University Graduate Center, Kjeller, Norway<br />
Abstract: Common methods for modeling, analysis, and synthesis of Fiber Bragg Gratings are reviewed in detail,<br />
including coupled-mode theory, transfer matrix methods, and layer-peeling algorithms.<br />
INTRODUCTION<br />
In order to design, fabricate, and use fiber Bragg gratings in various applications, tools for analysis, synthesis and<br />
characterization are crucial. For example, when designing gratings, a suitable mathematical model is central to<br />
understand the relation between the spatial grating structure and the reflection and transmission spectra. Moreover,<br />
fabrication of gratings is facilitated by synthesis methods, to extract the actual spatial structure from spectral<br />
measurements. Similarly, when using fiber gratings as sensor elements, synthesis methods can be used to extract any<br />
distributed sensing parameter (e.g., temperature, strain, etc.) along the fiber.<br />
A simple model of a fiber Bragg grating is the following: the grating reflects a narrow band of the incident optical<br />
field by successive, coherent scattering from the index variations. When the reflection from a crest in the index<br />
modulation is in phase with the next one, we have maximum mode coupling or reflection. Then the Bragg condition<br />
is fulfilled, i.e.,<br />
λB eff<br />
2n<br />
Λ , (1)<br />
where λB is the Bragg wavelength, neff is the effective modal index, and Λ is the perturbation period. Although this<br />
model captures one of the most important properties of fiber Bragg gratings, it cannot tell us how much light is<br />
actually reflected, how large bandwidth, detailed spectral dependence including reflectivity and group delay, impact<br />
of perturbations or imperfections, and so forth. We thus need a model that can describe the detailed wave<br />
propagation in fiber Bragg gratings, and therefore how the optical filter characteristics relate to the amplitude and<br />
phase modulated quasi-periodic refractive index profile.<br />
The most common mathematical model governing wave propagation in fiber Bragg gratings is coupled-mode<br />
theory. Other techniques are available; however, we will only consider coupled-mode theory since it is by far the<br />
most popular method. Coupled-mode theory is relatively straightforward and intuitive, and for most practical fiber<br />
gratings of interest it is accurate. In Section 2 we will derive the coupled-mode equations from the scalar wave<br />
equation. Once the governing equations are established we use them in Section 3 for analyzing two important<br />
special cases for which closed-form expressions for the reflectivity exist; uniform gratings and weak gratings. In<br />
Section 4 we develop general numerical techniques for computing the spectrum associated with any grating profile.<br />
Section 5 contains a description of the inverse problem, namely to obtain the grating profile from the reflection<br />
spectrum. Finally, we sum up some important spectral properties and constraints in Section 6.<br />
COUPLED-MODE THEORY<br />
The relation between the spectral dependence of a fiber grating and the corresponding grating structure is usually<br />
described by coupled-mode theory. Coupled-mode theory is described in a number of texts; detailed analysis can<br />
be found in Refs. [1-5]. The notation in this section follows most closely that of Refs. [1, 4]. We assume that the<br />
fiber is lossless and single mode in the wavelength range of interest. In other words, only one forward and one backward<br />
*Address correspondence to this author Professor Johannes Skaar: UNIK – University Graduate Center, Postboks 70, NO-2027 Kjeller,<br />
Norway; Tel: +47 64844748; Fax: +47 63818146; Email: johannes.skaar@iet.ntnu.no
36 Fiber Bragg Grating Sensors Johannes Skaar<br />
propagating modes are considered. Moreover, we assume that the fiber is weakly guiding, i.e., the difference<br />
between the refractive indices in the core and the cladding is very small. Then the electric and magnetic fields are<br />
approximately transverse to the fiber axis, and we can ignore all polarization effects due to the fiber structure and<br />
consider solely the scalar wave equation [1]. For analysis and synthesis of birefringent gratings or gratings in<br />
multimode fibers the reader is referred to Refs. [6-7]. The fiber axis is oriented in the z direction and we assume that<br />
the electric field is x-polarized. The implicit time dependence is exp(–iωt); thus a forward propagating wave with<br />
propagation constant β > 0 and frequency ω > 0 has the form exp[i(βz–ωt)].<br />
The grating is treated as a perturbation of the fiber. The unperturbed fiber has refractive index profile n(x,y), and the<br />
perturbed fiber has the z-dependent index n(x,y,z). Both fibers are assumed to be weakly guiding, so in the core and<br />
the cladding we have n(x,y,z) n(x,y) neff ncl nco. Here ncl and nco are the indices in the cladding and the<br />
core, respectively, and neff is the effective index of the supported mode in the absence of the grating. We write the<br />
total electric field as a superposition of the forward and backward propagating modes,<br />
Ex <br />
<br />
(x,y,z) b (z)Ψ(x,<br />
y) b (z)Ψ(x,<br />
y) , (2)<br />
where the coefficients b± contain all z-dependence of the modes. It is clear that b± are dependent on frequency since<br />
they include the harmonic propagation factor exp(±iβz), with β = β(ω) = neff ω/c the scalar propagation constant. The<br />
transverse dependence is described by the function Ψ, which satisfies the scalar wave equation for the unperturbed<br />
fiber,<br />
2 2 2 2<br />
k n (x,y) β Ψ 0<br />
t (3)<br />
where t 2 = ∂ 2 /∂x 2 + ∂ 2 /∂y 2 , and k = ω/c is the vacuum wavenumber. The total electric field Ex must satisfy the<br />
scalar wave equation for the perturbed fiber, i.e.,<br />
2 2 2<br />
2 2<br />
n (x,y,z) / z<br />
E 0<br />
t x<br />
k (4)<br />
By substituting Eq. (2) into Eq. (4), and using Eq. (3), we obtain<br />
2<br />
<br />
z<br />
2<br />
(b<br />
<br />
b )Ψ [β k (n<br />
<br />
2<br />
2<br />
2<br />
n<br />
2<br />
)](b<br />
<br />
b )Ψ 0<br />
Multiplication by Ψ and integration over the xy-plane leads to<br />
2 2<br />
b<br />
d b<br />
2<br />
(β<br />
2 2<br />
<br />
d<br />
dz<br />
dz<br />
where we have defined the coefficient D as<br />
D(z) <br />
k<br />
2n<br />
co<br />
(n<br />
2<br />
n<br />
2<br />
Ψ<br />
dA<br />
<br />
2knco<br />
D(z))(b<br />
b ) 0<br />
(6)<br />
2<br />
)Ψ<br />
2<br />
dA<br />
The integrations extend over the entire xy-plane. Eq. (6) can be decomposed into the following set of first order<br />
differential equations [1]<br />
db<br />
i(β D)b<br />
iDb<br />
, (8a)<br />
dz<br />
db<br />
i(β D)b<br />
iDb<br />
, (8b)<br />
dz<br />
(5)<br />
(7)
Fiber Bragg Gratings: Analysis and Synthesis Techniques Fiber Bragg Grating Sensors 37<br />
as is readily realized by differentiation and summation of the two Eqs. (8). This decomposition amounts to<br />
separating the total field in Eq. (2) into its forward and backward propagating components. Indeed, when n = n, we<br />
observe that the solution of Eqs. (8) reduces to b±(z) = B± exp(±iβz) with constant B±, i.e., b± correspond to the<br />
forward and backward propagating modes. In the absence of the grating, n = n, the modes propagate without<br />
affecting each other; otherwise the modes will couple to each other through the quantity D(z).<br />
For a fiber grating, the z-dependence of the index perturbation is approximately quasi-sinusoidal in the sense that it<br />
can be written<br />
2 2 2π<br />
<br />
n n Δεr,ac(z)<br />
cos<br />
z θ(z) Δεr,dc(z)<br />
, (9)<br />
Λ<br />
<br />
<br />
where Λ is a chosen design period so that θ(z) becomes a slowly varying function of z compared to a period Λ. The<br />
functions Δεr,ac(z) and Δεr,dc(z) are real and slowly varying, and satisfy<br />
r,ac<br />
2<br />
co<br />
2<br />
co<br />
|Δ (z)| n , |Δ (z)| n . (10)<br />
r,dc<br />
It is therefore convenient to express D as a quasi-sinusoidal function<br />
D(z)<br />
2π<br />
* 2π<br />
<br />
κ(z) exp i<br />
z κ (z) exp<br />
i z σ(z) , (11)<br />
Λ<br />
Λ<br />
<br />
<br />
where κ(z) is a complex, slowly varying function of z and σ(z) is a real, slowly varying function that accounts for the<br />
dc index variation from εr,dc(z). In order to simplify Eqs. (8), we define new field amplitudes u and v by setting<br />
b (z)<br />
<br />
<br />
<br />
π<br />
z<br />
u(z) exp i z exp(<br />
i<br />
σ(z')dz' )<br />
(12a)<br />
Λ<br />
0<br />
π <br />
z<br />
b(z)<br />
v(z) exp i z exp( i0<br />
σ(z')dz' ) . (12b)<br />
Λ <br />
By substituting Eqs. (11) and (12) into Eqs. (8), ignoring the terms that are rapidly oscillating since they contribute<br />
little to the growth and decay of the amplitudes, we arrive at the coupled-mode equations<br />
du<br />
dz<br />
iδu<br />
q(z)v , (13a)<br />
dv *<br />
iδv<br />
q (z)u . (13b)<br />
dz<br />
In Eqs. (13) we have defined the wavenumber detuning δ = β – π/Λ, and the coupling coefficient q of the grating<br />
q(z)<br />
z<br />
0<br />
iκ(z)<br />
exp ( 2i<br />
σ(z')dz' ) . (14)<br />
Note that all phase factors in Eqs. (12) are independent of the propagation constant β and therefore the frequency.<br />
Hence, we can simply treat the new variables u and v as the fields themselves once the reference planes have been<br />
fixed, since they differ only from b± by constant phase factors. For example, the reflection coefficient b–(z0)/b+(z0)<br />
can as well be computed by the expression v(z0)/u(z0) once the position z0 has been fixed because the two<br />
expressions only differ by a constant phase factor 1 . Also note that all functions u, v, and q are slowly varying with z<br />
compared to the period Λ because β π/Λ when the wavelength is close to the Bragg wavelength λB = 2neff Λ.<br />
1 Strictly speaking, σ(z) is dependent on frequency (see Eq. (16)). However, since fiber gratings usually have narrow bandwidths, this dependence<br />
can be ignored to a very good approximation.
Fiber Bragg Grating Sensors: Recent Advancements, Industrial Applications and Market Exploitation, 2011, 53-77 53<br />
Andrea Cusano, Antonello Cutolo and Jacques Albert (Eds)<br />
All rights reserved - © 2011 <strong>Bentham</strong> <strong>Science</strong> Publishers Ltd.<br />
CHAPTER 4<br />
Photonic Bandgap Engineering in FBGs by Post Processing Fabrication<br />
Techniques<br />
Andrea Cusano 1,* , Domenico Paladino 1 , Antonello Cutolo 1 , Agostino Iadicicco 2 and<br />
Stefania Campopiano 2<br />
1 Optoelectronic Division – Engineering Department, University of Sannio, Corso Garibaldi 107, 82100 Benevento,<br />
Italy and 2 Department of Technology, University of Naples “Parthenope”, Centro Direzionale di Napoli Isola C4,<br />
80143 Napoli, Italy<br />
Abstract: The incredible growth of Fiber Bragg Gratings (FBGs) from both the research and the commercial<br />
points of view, has forced the development of several methods able to tailor their spectral characteristics for<br />
specific applications. Basically, two different approaches can be adopted to specialize the grating spectra: one is<br />
based on complex grating profiles induced directly at the fabrication stage. Another, more interesting, approach<br />
relies on post processing methodologies that introduce localized defects along the grating, breaking its periodic<br />
structure. The presence of the defects leads to the formation of allowed bands or defect states within the grating<br />
bandgap. The introduction of finer scale spectral features results in new interesting perspectives in both<br />
telecommunications and sensing fields. In addition, the possibility to accurately control the defect states spectral<br />
features – depth, bandwidth, and spectral position – could allow the complete engineering of the FBG bandgap,<br />
opening the way to the realization of several new advanced and attractive photonic devices. This <strong>chapter</strong> is<br />
focused to review the main advancements in FBGs post processing to easily control the spectral features of the<br />
final device for specific applications.<br />
INTRODUCTION<br />
Thanks to their peculiar spectral characteristics, Fiber Bragg Gratings (FBGs) are widely popular devices within and<br />
out the photonic community [1-2]. Basically – by means of a UV induced periodic modulation of the core refractive<br />
index along the hosting optical fiber – FBGs are able to selectively reflect only a narrow wavelength range (typically<br />
0.1-1 nm), giving rise to an effective Photonic BandGap (PBG). Consequently, FBGs – as they stand – represent an<br />
useful tool to manipulate the propagating features of the light traveling along an optical fiber [3], and can be usefully<br />
exploited to develop wavelength encoded sensing elements in numerous application fields [4-5]. As a consequence<br />
of the technological assessment in FBGs fabrication, in the mid 1990’s many research groups have been engaged in<br />
the study and realization of new fiber grating devices through more complex refractive index modulation profiles [6-<br />
12]. The common idea was to complicate the core refractive index modulation allowing further manipulation of the<br />
light propagating within the fiber grating structures. In this way, it is possible to realize fiber grating with peculiar<br />
spectral characteristics either in reflection and transmission.<br />
Since FBGs can be considered one-dimensional PBG structures realized within optical fibers by UV writing, among<br />
the different complex grating structures, a particular and interesting approach relies on the inclusion of local defects<br />
breaking the FBG periodicity along the grating axis. One of the most important modern research field, in fact, is<br />
related to the novel and unique way to control many features of the electromagnetic radiation by opportune PBG<br />
structures. These artificial structures are characterized by one-, two-, or three-dimensional periodic arrangements of<br />
dielectric materials and are commonly called Photonic Crystals (PhCs) [13].<br />
PhCs are the optical analogue to electronic semiconductors, i.e., PhCs can be considered semiconductors for photons:<br />
in the electronic case, the wave functions of electrons interact with the periodic potential of the atomic lattice and, for<br />
a certain range of energies (similar to the frequencies of photons), electronic states cease to exist, thus giving rise to<br />
an electronic bandgap. For PhCs, the analogous of the electronic potential in an atomic crystal is the spatial<br />
distribution of the dielectric constant of the PhC structure. Moreover, due to a periodic interaction, certain PBGs<br />
appear, forbidding the propagation of certain frequency ranges of light. In such a structure, a line or point defect, which<br />
*Address correspondence to this author Professor Andrea Cusano: Optoelectronic Division – Engineering Department, University of Sannio,<br />
Corso Garibaldi 107, 82100 Benevento, Italy; Tel: +39 0824305835; Fax: +39 0824305846; Email: a.cusano@unisannio.it
54 Fiber Bragg Grating Sensors Cusano et al.<br />
breaks the lattice periodicity, can be introduced wherein a mode is localized in virtue of being suppressed by the<br />
lattice. In practice, defects break the PhC structure symmetry and may give rise to a defect state or an allowed<br />
frequency within the PBG region. Several research groups proposed and experimentally demonstrated optical<br />
trapping and dropping functions by exploiting resonant tunneling [14], cavity wave-guiding coupling phenomena in<br />
one-dimensional [15-16], two-dimensional [17], and three-dimensional [18] structures.<br />
In the case of FBGs, the introduction of local defects introducing a phase shift along the grating can be substantially<br />
obtained by adopting two different methods:<br />
- by acting at the fabrication stage through the modification of the photo-induced core refractive index<br />
modulation generating punctual or distributed phase shift;<br />
- by post processing approaches aimed to locally modify the FBG physical and/or geometrical features<br />
generating distributed phase shift.<br />
The common spectral effect is to induce narrow transmitted windows or defect states within the standard FBG stopband.<br />
Clearly, narrower spectral features would further increase the FBG potentialities in both telecommunications<br />
and sensing.<br />
Even if this <strong>chapter</strong> is focused on the second class of FBG based devices, here it is important to briefly summarize<br />
also the main milestones characterizing the first category.<br />
In 1986, it was originally demonstrated by Alferness et al. [19] that the insertion of a single quarter-wave phase shift<br />
at the center of a Bragg grating waveguide opens a band-pass peak within the stop-band.<br />
The same principle of operation was adopted in 1990 by Reid et al. to realized phase-shifted Moiré grating<br />
resonators [20]. The Moiré approach to fiber grating transmission filter fabrication was reported for application to<br />
side-etched, surface-relief structures. Later, the same technique was adopted using the UV direct-write grating<br />
fabrication method and uniform FBGs [21]. The Moiré grating resonators were fabricated using the same technique<br />
as for standard FBGs, but adopting the holographic exposure. Moirè gratings were formed by a double exposure to<br />
two interference patterns of slightly different periods.<br />
In 1994, Agrawal and Radic modeled the response of phase-shifted FBGs and showed that the transmitted peak<br />
shifts with the amount of the phase shift in the middle of the grating [22]. FBGs fabricated with single or multiple<br />
phase shifts at various precise locations along the grating have been proposed for creating narrow spectral<br />
resonances for both filtering and laser applications [23-24]. Such structures can be produced at the fabrication stage<br />
by using specially designed phase masks incorporating the required phase shifts with restrictions on the magnitude,<br />
position, and number [24-26]. However, the cost of this procedure, the dependence on the operating wavelength and<br />
the need of a large variety of phase masks to produce structures with different and optimized spectral properties for<br />
specific applications are the major limitations of this approach.<br />
Phase shifts were also realized by detuning the relative position between the phase mask and the optical fiber<br />
hosting the grating at properly selected longitudinal positions [27-28]. In this case a piezoelectric ceramic translation<br />
stage was used to change the relative position during the fabrication stage, but, in principle, it would be extremely<br />
useful to separate the FBG fabrication stage from the successive introduction of defects along their structure. In this<br />
sense, during the last decades several mechanisms have been adopted to introduce phase shifts along FBGs,<br />
allowing also a more accurate and adjustable control of the induced phase shift amount.<br />
In particular, many research studies focused their attention on the development of wavelength independent post<br />
processing techniques able to induce distributed phase shifts along the grating structure enabling the tailoring of the<br />
bandgap features. In this <strong>chapter</strong>, the main technological steps are carefully reviewed.<br />
POST PROCESSING APPROACHES: A VIEW BACK<br />
The first attempt to introduce a defect along a FBG by post processing methods was dated 1994 by Canning and Sceats<br />
[29]. Starting from the demonstration of grating writing by point-by-point technique [30] – having the significant<br />
advantage of allowing arbitrary phase gratings to be written – here the basic idea relied on the demonstration that a
Photonic Bandgap Engineering in FBGs by Post Processing Fabrication Techniques Fiber Bragg Grating Sensors 55<br />
local UV post processing treatment of a periodic grating can be used to generate a complex grating. The local<br />
irradiation by UV light raises the general refractive index at a certain region along the fiber grating as shown in (Fig. 1).<br />
Such post processing produces two gratings out of phase with each other which act as a wavelength selective Fabry-<br />
Pérot resonator allowing light at the resonance to penetrate the stop-band of the original grating. The resonance<br />
wavelength depends on the size of the phase change. For the experimental realization an homemade strong uniform 4 cm<br />
long FBG (reflectivity of more than 99% and Full Width at Half Maximum (FWHM) of about 0.15 nm, kL ~ 3) was<br />
selected. 240 nm light was focused directly onto the centre of the grating over a length of about 1 mm. The FBG<br />
transmission spectrum was monitored with a resolution of 1 pm. A defect state was observed to grow into the reflection<br />
bandwidth and after 12000 shots the transmission spectrum of (Fig. 2) was obtained. Continued processing resulted in<br />
no further changes, indicating that the processed region had reached a saturated level of index change.<br />
Figure 1: Introduction of a phase shift by raising the refractive index at a point along the FBG. (J. Canning et al. (1994),<br />
“π-phase-shifted periodic distributed structures in optical fibres by UV post-processing,” Electron. Lett. 30, 1344-1345. © IEEE.<br />
Reproduced with permission)<br />
transmission<br />
1.2<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
0.0<br />
q1<br />
L<br />
1<br />
UV beam<br />
q2<br />
1519.8 1519.0 1520.0<br />
wavelength,nm<br />
Figure 2: Normalized transmission spectrum of the grating after UV post-processing. (J. Canning et al. (1994), “π-phase-shifted<br />
periodic distributed structures in optical fibres by UV post-processing,” Electron. Lett. 30, 1344-1345. © IEEE. Reproduced with<br />
permission)<br />
The experimental results revealed also that, once the phase shift was introduced, an overall broadening of the whole<br />
bandwidth of the spectrum results. Theoretical calculations confirmed the experimentally measured broadening of<br />
the bandwidth. The obtained defect state demonstrated that a phase shift close to π was introduced. The narrowness<br />
of this defect state – close to the utilized resolution, in this case – is influenced by several factors, but, above all, by<br />
the strength of the grating. The most obvious applications include production of very narrowband transmission<br />
filters. In addition, the length of the grating resulted in high sensitivity to external perturbations such as temperature<br />
and strain where potential applications in sensor devices involving detuning of the resonator to detect small changes<br />
in the measurand can be envisaged. It is worth noting that such a technique could be used to introduce multiple<br />
phase shifts along the same grating to produce other devices such as comb filters.<br />
This technique was used in 2005 by Canning and associates to finely tune the phase shift notch spectral position in a<br />
Distributed Feed-Back (DFB) Photonic Crystal Fiber (PCF) laser configuration [31]. The DFB laser structure relied<br />
L 2
78 Fiber Bragg Grating Sensors: Recent Advancements, Industrial Applications and Market Exploitation, 2011, 78-98<br />
Fiber Bragg Grating Interrogation Systems<br />
José Luís Santos 1,2,* , Luís Alberto Ferreira 1,3 and Francisco Manuel Araújo 1,3<br />
Andrea Cusano, Antonello Cutolo and Jacques Albert (Eds)<br />
All rights reserved - © 2011 <strong>Bentham</strong> <strong>Science</strong> Publishers Ltd.<br />
CHAPTER 5<br />
1 INESC Porto, Rua do Campo Alegre 687, 4169-007 Porto, Portugal; 2 Faculdade de Ciências da Universidade do<br />
Porto, Rua do Campo Alegre 687, 4169-007 Porto, Portugal and 3 FiberSensing S.A., Rua Vasconcelos Costa 277,<br />
4470-640 Maia, Portugal<br />
Abstract: Fiber Bragg Gratings are structures with remarkable characteristics that have induced new qualitative<br />
developments in the broad field of optical fiber technology, most notably in optical communications and in<br />
optical sensing. When these devices are applied for sensing, the underlying concept is the modulation of the<br />
grating Bragg wavelength by the measured and, therefore, a central issue is the sensitive and accurate conversion<br />
of the resonant wavelength into a proportional electrical signal with the adequate format for further processing.<br />
This topic is broadly known as Fiber Bragg Grating interrogation and is the subject of the present <strong>chapter</strong>. It is<br />
organized in two parts: in the first one, the techniques developed by the scientific community looking for this<br />
functionality are reviewed, with emphasis on the identification of general conceptual classes where they fit; in the<br />
second part, illustrative and state-of-the-art commercial Fiber Bragg Grating interrogation systems are described.<br />
INTRODUCTION<br />
Since the beginning the optical fiber sensing concept promised novel developments in the vast area of<br />
instrumentation and measurement, conviction anchored in the unique feature of this technology which is the fact that<br />
the optical fiber is simultaneously sensing element and communication channel. Therefore, there is no need to<br />
consider a specific telemetry system that is always required in other sensing technologies and, probably more<br />
important, this dual characteristic of the optical fiber opens the natural consideration of fiber optic sensing networks<br />
for multi-point or distributed measurement. Up to the outcome of the fiber Bragg grating technology, a vast number<br />
of sensing solutions based on optical fibers were demonstrated, but only a few could overcome the demanding<br />
obstacles that are present in the industrialization and commercialization paths. In that sense, and certainly by other<br />
reasons, fiber Bragg gratings (FBG) were a technological breakthrough. In fact, these structures can be considered as<br />
almost ideal sensing elements. One central feature is the circumstance that the measurand information is, in the vast<br />
majority of situations, encoded in the resonance wavelength, which is an absolute parameter and provides unique<br />
self-referencing capacity. There are also other favorable characteristics, common to the optical fiber sensing<br />
technology, such as small size and weight, as well as immunity to electromagnetic interference. Additionally, the<br />
wavelength encoded characteristic of these devices confer them excellent multiplexing capability, which permits to<br />
address a large number of sensors with a single optical fiber cable [1].<br />
In principle, the action of a particular measurand on a fiber Bragg grating can affect one or more of the<br />
characteristics of the device spectral signature, i.e., its resonance wavelength, the spectral width or the reflectivity.<br />
However, in most of the cases reported up to now the focus has been on the resonance wavelength direct modulation<br />
and, therefore, the subject known as FBG interrogation addresses the problem of transforming this wavelength<br />
modulation into an optical intensity modulation, and later into a corresponding electrical signal compatible with the<br />
common standards of instrumentation. Desirably, this transformation is accurate in the determination of the absolute<br />
Bragg wavelength and sensitive in the detection of small Bragg wavelength shifts, a perspective that in practice<br />
needs to be balanced by the factors complexity and cost in a particular FBG interrogation configuration.<br />
This <strong>chapter</strong> addresses the problem of interrogation of fiber optic sensors based on Bragg gratings. The first part<br />
reviews the interrogation concepts developed so far, emphasizing their core characteristics, performance and<br />
requirements, while the second part describes some state-of-the-art equipments commercially available and<br />
illustrative of the main FBG interrogation approaches.<br />
*Address correspondence to this author Professor José Luís Santos: INESC Porto, Rua do Campo Alegre 687, 4169-007 Porto, Portugal; Tel:<br />
+351 220 402 302; Fax: +351 220 402 437; Email: josantos@fc.up.pt
Fiber Bragg Grating Interrogation Systems Fiber Bragg Grating Sensors 79<br />
FIBER BRAGG GRATING INTERROGATION<br />
There are several types of fiber Bragg gratings with specific optical transfer functions, but the standard structure has<br />
a reflectivity function with an approximately Gaussian shape centered at the resonance wavelength of the device and<br />
a spectral width around 0.2 nm. When a particular measurand direct or indirectly affects the effective refractive<br />
index of the core mode, or the period of the grating refractive index modulation, or both, then the resonance<br />
condition changes and there is a spectral shift of the grating signature. FBG interrogation means to convert this<br />
wavelength shift into a variation of an electrical signal with adequate characteristics to obtain the information about<br />
the measurand. The general principle of FBG interrogation is shown in (Fig. 1). The optical source can be a<br />
broadband source (LED, SLD, ASE, Supercontinuum), in which case it operates passively and normally there is no<br />
control from the processing unit. This is not the case when spectrally narrow illumination is used, most of the cases<br />
from laser sources, in which the wavelength modulation of the emitted light can be a component of the FBG<br />
interrogation technique.<br />
<br />
Vout f B<br />
<br />
Figure 1: General layout for interrogation of fiber Bragg grating sensors.<br />
B<br />
Mensurand( )<br />
Table 1 groups and classifies the FBG interrogation techniques that have been proposed along the years. As will be<br />
described next sections, the underlying physical principles are diverse, which is also reflected in the performances<br />
achievable with each particular configuration. Independently of the specificities of each FBG interrogation<br />
approach, some general goals can be indicated. First of all, an appropriated interrogation technique must provide a<br />
reproducible transduction of the Bragg wavelength. Additional merit factors will be high sensitivity, large<br />
measurement range, immunity to optical power fluctuations, low environmental susceptibility, amenability to sensor<br />
multiplexing, simplicity and low cost. Certainly, some of these characteristics do not coexist and, therefore, for any<br />
particular demodulation solution a compromise is required.<br />
Bulk Optics<br />
The first group of interrogation systems indicated in Table 1 is based on utilization of diffraction gratings and<br />
volume holograms. The designation “bulk” is normally utilized to mention conventional optical devices, i.e., not<br />
supported in optical fibers. Of historical importance is the utilization of a monochromator, by Morey et al. [2], to<br />
perform in 1989 some of the first studies involving fiber Bragg gratings. Normally, the diffraction gratings are<br />
integrated in equipment such as monochromators and optical spectrum analyzers, but can also be used externally in<br />
association with CCD cameras (Blair et al. [3], Xu et al. [4]). These systems are versatile and can interrogate a large<br />
number of sensors. However, to obtain good sensitivity in the determination of the gratings resonance wavelengths it<br />
is required high density diffraction gratings associated with high precision rotation stages or, alternatively, large<br />
distances between the diffraction gratings and the detectors. This increases the size of the systems and their cost,<br />
eventually making them an attractive solution only in the cases where it is needed to read a large number of Bragg<br />
FBG
80 Fiber Bragg Grating Sensors Santos et al.<br />
sensors. However, for applications in which the requested sensitivity is not demanding, the utilization of<br />
spectrometers based on low resolution diffraction gratings and small CCD arrays can be a favorable solution,<br />
particularly when sub-pixel algorithms are explored (Askins et al. [5], Liu et al. [6]). Besides these situations, the<br />
utilization of monochromators and optical spectrum analyzers is essentially confined to the laboratory environment.<br />
Table 1: Interrogation techniques of FBG based sensors.<br />
Group Principle Configuration References<br />
Monochromator [2]<br />
Bulk Optics Diffraction<br />
Optical Spectrum Analyzer<br />
Diffraction Grating + CCD<br />
[3-4]<br />
[5-6]<br />
Volume Hologram + CCD [7]<br />
Bulk Filter Transmissive [8-9]<br />
Integrated Optic Filter Arrayed Waveguide Grating [10-11]<br />
Biconical Filter [12]<br />
Long-Period Grating [13-15]<br />
Passive Edge Filtering Optical Fiber Filters<br />
Fused Coupler [16-17]<br />
Chirped Bragg Grating [18-19]<br />
Sagnac Loop [20]<br />
Sources/Detectors<br />
Source Spectrum<br />
Detector Spectral Responsivity<br />
[21-22]<br />
[23]<br />
Fabry-Pérot [24-27]<br />
Bulk Optical Filters<br />
Acousto-Optic [28-30]<br />
Hybrid Optical MEMS [32]<br />
Receiving Bragg Grating [34-36]<br />
Active Bandpass Optical Fiber Filters<br />
WDM Coupler [37]<br />
Filtering<br />
Dynamic Long-Period Grating [38]<br />
Optical Sources<br />
Singlemode Laser Diode<br />
Multimode Laser Diode<br />
[39-43, 46]<br />
[44]<br />
Carrier Generation<br />
Receiving Bragg Grating<br />
Multimode Laser Diode<br />
[50-51]<br />
[48-49]<br />
Passive Chirped Grating + Sagnac loop [53]<br />
Homodyne Mach-Zenhder [52, 54-56]<br />
Interferometric Carrier Generation Mach-Zenhder [57, 61]<br />
Fourier Domain Variable OPD Scan [62, 65]<br />
Optical Coherence Function [66]<br />
Laser Sensing FBG Laser Cavity FBG as Cavity Mirror [67-74]<br />
Angular Dispersion Receiving Blazed FBG + CCD [75-76]<br />
Miscellaneous<br />
OTDR [77-80]<br />
Wavelet Processing [81]<br />
A different approach consists in using volume holograms written in photo-refractive materials as diffractive<br />
elements (James et al. [7]). In such cases some benefits can result, considering it is then practical to store in a single<br />
passive device a large number of filters with different spectral responses, which can later be tuned in order to fulfill<br />
the needs of a specific application. This flexibility, as well as the ability to tune the filters central wavelengths by<br />
application of electric fields to the material where the hologram is written, indicate that the utilization of these<br />
elements is a conceptually elegant solution for interrogation of fiber Bragg sensors. However, their bulk nature and<br />
high insertion losses are factors that probably will limit their dissemination in this field.
Fiber Bragg Grating Sensors: Recent Advancements, Industrial Applications and Market Exploitation, 2011, 99-115 99<br />
Multiplexing Techniques for FBG Sensors<br />
Manuel López-Amo 1,* and Jóse Miguel López-Higuera 2<br />
Andrea Cusano, Antonello Cutolo and Jacques Albert (Eds)<br />
All rights reserved - © 2011 <strong>Bentham</strong> <strong>Science</strong> Publishers Ltd.<br />
CHAPTER 6<br />
1 Grupo de Comunicaciones Ópticas, Universidad Pública de Navarra, Campus de Arrosdía, E-31006-Pamplona,<br />
Spain and 2 Grupo de Ingeniería Fotónica, Universidad de Cantabria, ETSII y Telecomunicación, Avda. de los<br />
Castros s/n E-39005 Santander, Spain<br />
Abstract: One of the main goals of fiber optic sensor technology is to multiplex together a high number of<br />
sensors in the same fiber in order to share expensive terminal equipment and reduce the size and weight of the<br />
optical cable. Both passive and active multiplexing techniques are used for telemetry in sensor networks. The<br />
different multiplexing techniques for Fiber Bragg Grating sensors will be described and compared here, including<br />
networks using optical amplification and lasing multiplexing systems.<br />
INTRODUCTION<br />
Multiplexing is the simultaneous transmission of two or more information channels along a common path. A fiber sensor<br />
system includes three main parts or subsystems: the sensing elements or transducers, the optical fiber channel and the<br />
optoelectronic unit [1]. Because of the high cost of the last subsystem, when it is possible to multiplex a high number of<br />
sensing points in the same network using a common optoelectronic unit, the cost per sensing element decreases.<br />
125 Systems employing multiplexed arrays of fiber Bragg gratings (FBGs) (surface mounted or embedded in<br />
structures and materials) have performed successfully in numerous field trials and applications involving a wide<br />
variety of structures, as will be shown in the next <strong>chapter</strong>s. There are numerous companies that commercialize FBG<br />
sensors, arrays of FBGs and optoelectronic units for FBGs multiplexed networks.<br />
The development of the technology and components used in the optoelectronic units and multiplexing networks has<br />
been helped by the fast growing of the fiberoptic telecommunications technology. High performance tunable lasers,<br />
couplers, optical switches, optical amplifiers, filters and detectors are available for sensors multiplexing due to the<br />
major market that supposes telecommunications. Furthermore, certain multiplexing techniques, like Wavelength<br />
Division Multiplexing, come from the telecommunications side. However, other multiplexing techniques have been<br />
developed or adapted specifically for fiber optic sensors multiplexing [2].<br />
In this <strong>chapter</strong> we are going to focus on the most utilized techniques for FBGs multiplexing, omitting other<br />
techniques, like coherence division multiplexing or polarization division multiplexing, that are little tied to FBGs.<br />
The sensor networks can be designed using only passive fibers (without utilizing optical gain) or introducing optical<br />
amplification in some key parts of the networks and hence passive or active networks can be developed [3-4].<br />
Especially interesting for long distance applications are the systems that transform the multiplexing structure in a<br />
multiwavelength laser by the utilization of optical amplifiers.<br />
The aim of the <strong>chapter</strong> is to give a comprehensive overview of multiplexing techniques for FBGs, including a<br />
review of the most important achievements in modern multiplexing systems.<br />
GENERAL CONCEPTS<br />
FBG based sensors are wavelength selective devices that may be interrogated using different modulation formats. It<br />
is possible to interrogate them by using their reflected signal or the transmitted one. However, the reflective<br />
response is the most usually employed.<br />
A variety of multiplexing techniques based on different modulation formats have been developed, each one with<br />
their specific advantages for a particular application. As it is shown in (Fig. 1), the modulation formats generally fall<br />
*Address correspondence to this author Professor Mauel López-Amo: Grupo de Comunicaciones Ópticas, Universidad Pública de Navarra,<br />
Campus de Arrosadía, E-31006-Pamplona, Spain; Tel: +34 948169055; Fax: +34 948169720; Email: mla@unavarra.es
100 Fiber Bragg Grating Sensors López-Amo and López-Higuera<br />
into one of the following categories: Wavelength Division Multiplexing, Time Division Multiplexing (TDM),<br />
Frequency Division Multiplexing, Coherence Multiplexing and Polarization Division Multiplexing [5]. The last two<br />
techniques are not usually employed for FBGs multiplexing, but we must consider also hybrid approaches<br />
(simultaneous utilization of two modulation formats inside the same network).<br />
Figure 1: Multiplexing modulation formats: (a) Wavelength Division Multiplexing (WDM); (b) Time Division Multiplexing<br />
(TDM); (c) Frequency Division Multiplexing (FDM); (d) Coherence Multiplexing (CM) and (e) Polarization Division<br />
Multiplexing (PDM).<br />
When a single fiber is devoted to each sensor, we also use the term Spatial Division Multiplexing as another<br />
multiplexing technique. As depicted in (Fig. 2), to build up a sensor network, four basic configurations or topologies<br />
can be used. In all of them, the optimized received signal power is a key factor. The optoelectronic unit of the<br />
system must receive from the network an optical power level that ranges from the saturation power of the detector to<br />
the minimum accepted power (in which the system works with an acceptable signal to noise ratio). This difference is<br />
called the dynamic range of the system, which is another important design factor. Couplers and optical circulators<br />
are basic devices for networking, being their losses a fundamental parameter in the network behavior. In the star<br />
topology, the chosen splitting ratio uses to be 1/M, where M is the number of sensors to be multiplexed. However, in<br />
branched buses (buses which includes couplers for power distribution) or dual buses, the selection of the coupling<br />
factor is not an easy task [6]. The usual case, in which it is desirable that all couplers have the same splitting ratio,<br />
the optimum value should be 1/M in a branched single bus configuration. On the other hand, the received signal<br />
level from the M sensors may be equalized by tailoring the coupling ratio of each coupler. This last strategy is not<br />
usually selected, because of economical and operative reasons.<br />
a)<br />
b)<br />
c)<br />
O.S<br />
O.D<br />
O.S<br />
T<br />
T<br />
O.D<br />
T<br />
T<br />
O.S<br />
T<br />
O.D<br />
T<br />
1<br />
3<br />
2<br />
C<br />
Star<br />
coupler<br />
1xM<br />
Serial Bus<br />
S1 S2 SM-1 S M<br />
Dual Bus<br />
S1 S2 SM-1 S M<br />
Figure 2: Typical fiber optics sensor network topologies: (a) serial bus; (b) dual bus; (c) star and (d) tree topologies.<br />
S1<br />
S2<br />
SM-1<br />
S M<br />
d)<br />
O.S<br />
T<br />
O.D<br />
T<br />
1 2<br />
3<br />
Tree<br />
O.D<br />
T<br />
S1<br />
S2<br />
S M-1<br />
S M
Multiplexing Techniques for FBG Sensors Fiber Bragg Grating Sensors 101<br />
Multiplexing optical fiber sensors includes four basic tasks [3]: i) Launching an optical signal into the network having a<br />
suitable power, spectral distribution, polarization and modulation; ii) The detection of the signal which has been coded or<br />
modulated by the sensor element, and sent back to the optoelectronic unit by transmission or reflection; iii) Uniquely<br />
identifying the information corresponding to each sensor in the network, by means of proper addressing, polling and<br />
decoding; iv) The evaluation of the generated optical signals by modulating each sensor with a calibrated electrical signal.<br />
To efficiently design the most appropriate sensor network, it is necessary to take into account different aspects: the<br />
modulation and coding format of the optical signal, the network topology, the inclusion or not of optical<br />
amplification technologies, the decoding method for the received signal, and the type or types of sensors<br />
multiplexed in the same network. Finally, it must be also considered the economic conditions, which would<br />
eventually determine the most appropriate network (see Fig. 3).<br />
Figure 3: Illustration of the interdependence of the main concepts on a typical fiber optics sensor network.<br />
It must be noted that there is not a single ‘ideal’ approach for all the applications. However, there is indeed an<br />
optimum approach for each application. The choice of a different multiplexing technique depends on the<br />
requirements of the sensor network. The relative importance of parameters such as cost, noise, Bandwidth (BW),<br />
and flexibility constitutes the basis for making the right selection.<br />
SPATIAL DIVISION MULTIPLEXING<br />
This technique is also known as fiber multiplexing, and it is the first natural approach to build up fiber optic sensor<br />
networks. As illustrated in (Fig. 4), a single light source feeds M fiber channels and interrogates M sensing points,<br />
sharing in this way the cost of the laser or the broadband incoherent source over the entire network. In the typical<br />
reflective configuration, M fibers (one per path) are used to interrogate the M FBG sensors. By using optical<br />
circulators located at the M outputs of the optical coupler (OC), each sensor response is detected by an optical<br />
detector (OD). In the unusual transmissive response (dashed lines), 2·M fibers are needed. In the reflective case, the<br />
optical power at each detector drops by a factor of M, which corresponds to a signal power drop of 1/M 2 . Assuming<br />
shot noise limited operation, the signal to noise ratio (SNR) in fiber multiplexed systems falls to (1/M 2 )/(1/M) = 1/M,<br />
i.e. the SNR degradation in decibels is given by 10·logM.<br />
Figure 4: Illustration of a spatial or fiber multiplexed network, using one optical source, M detectors on the Optoelectronic Unit.<br />
Continuous line: reflective configuration. Dashed line: transmissive configuration.
116 Fiber Bragg Grating Sensors: Recent Advancements, Industrial Applications and Market Exploitation, 2011, 116-142<br />
Andrea Cusano, Antonello Cutolo and Jacques Albert (Eds)<br />
All rights reserved - © 2011 <strong>Bentham</strong> <strong>Science</strong> Publishers Ltd.<br />
CHAPTER 7<br />
Polarization Properties of Fiber Bragg Gratings and Their Application for<br />
Transverse Force Sensing Purposes<br />
Christophe Caucheteur*, Sébastien Bette, Marc Wuilpart and Patrice Mégret<br />
Electromagnetism and Telecommunications Department, Faculté Polytechnique, Université de Mons, Boulevard<br />
Dolez 31, 7000 Mons, Belgium<br />
Abstract: Due to the lateral inscription process, Fiber Bragg Gratings (FBGs) written into standard single-mode<br />
fiber can exhibit birefringence. The birefringence value is however too small to be perceived in the FBG spectral<br />
response but it can lead to significant polarization-dependent properties such as polarization dependent loss<br />
(PDL) and differential group delay (DGD). A transverse force applied on the FBG can enhance the birefringence,<br />
which increases both the PDL and DGD values. Hence, although these properties are not desired in<br />
telecommunication applications, they can be advantageously used for transverse force sensing measurements,<br />
allowing the use of FBGs written into standard single-mode optical fiber, which fail to work when they are<br />
interrogated through amplitude spectral measurements. This <strong>chapter</strong> first analyzes the normalized Stokes<br />
parameters, PDL and DGD evolutions with wavelength when the FBG parameters and the birefringence value are<br />
modified. It then focuses on the realization of a transverse force sensor based on the monitoring of the PDL and<br />
DGD evolutions.<br />
INTRODUCTION<br />
Fiber Bragg grating (FBG) sensors are widely used to measure temperature and axial strain. During the last decade,<br />
there has been increasing attention on the application of FBG sensors in smart structures, where sensors are<br />
deployed to measure not only the axial strain but also the transverse force. Transversal strain sensors are of high<br />
importance in the civil engineering domain. Moreover, FBGs can be used in embedded applications for structural<br />
health monitoring. As an example, in composite materials, sensors are used for cure monitoring during the<br />
manufacturing process and also for health monitoring of the final product.<br />
It is known that the transverse strain or force induces birefringence in FBGs written into standard fiber and thus cause<br />
their reflection band to split into two bands characterized by orthogonal polarizations [1]. The amount of<br />
birefringence is directly linked to the transverse force value [2]. However, for force values below a few hundreds of<br />
Newtons, the mechanically-induced birefringence is not sufficient to reach the complete separation between the two<br />
bands corresponding to the two orthogonal polarization modes. They overlap and the resulting amplitude spectrum is<br />
completely deformed. Accurate measurements are therefore not possible with a single FBG fabricated into a standard<br />
single mode fiber (SMF), unless a polarizer is optimized in front of the sensor to discriminate between the amplitude<br />
spectra corresponding to the two polarization modes [3]. Such discrimination is essentially visual and thus fails in<br />
being automated and precise. It is the reason why FBGs written into polarization maintaining fiber (PMF or highbirefringence<br />
fiber) have been privileged for transverse force sensing purposes [4].<br />
PMFs generally contain stress applying regions in the cladding that break the circular symmetry of the fiber cross<br />
section and lead to an important intrinsic fiber birefringence of the order of 10 -4 . The pristine birefringence in PMF is<br />
hundreds or thousands times higher than in standard SMF. This stress-induced birefringence leads to different<br />
propagation constants for the two orthogonal polarization modes (also called slow and fast modes – the slow mode<br />
corresponds to the axis passing through the stress applying regions) so that the FBG fabrication in PMF results in the<br />
formation of dual well-separated resonance bands [5]: the wavelength separation between these two bands, which is<br />
linked to the birefringence value, is generally of the order of a few hundreds of picometers [6-8].<br />
When a transverse force is applied on an FBG written into a PMF, the resonance bands amplitudes and central<br />
*Address correspondence to this author Dr. Christophe Caucheteur: Electromagnetism and Telecommunications Department, Faculté<br />
Polytechnique, Université de Mons, Boulevard Dolez 31, 7000 Mons, Belgium; Tel: +32 65374149; Fax: +32 65374199; E-mail:<br />
christophe.caucheteur@umons.ac.be
Polarization Properties of Fiber Bragg Gratings Fiber Bragg Grating Sensors 117<br />
wavelengths are simultaneously modified. The rate of change depends on the direction of the applied force with respect<br />
to the PMF slow and fast axes [6]. Moreover, the maximum wavelength shift of a given resonance band is obtained<br />
when the loading direction is oriented along the corresponding fiber axis. It corresponds to the minimum sensitivity<br />
for the orthogonal axis. Hence, the resonance band wavelength shift exhibits a periodic dependence on the load<br />
orientation, with a period of about 180° [6, 8-10]. The sensitivity of the slow axis has been reported higher than that<br />
of the fast axis [8]. Depending on the fiber birefringence value, it can reach a maximum of about 0.2 nm/(N/mm) for<br />
the slow axis. However, as the temperature and axial strain effects also shift the resonance bands [5], monitoring<br />
separately the resonance bands shifts does not provide temperature-insensitive and axial strain-insensitive transverse<br />
force measurements. Therefore the operating principle of transverse force sensing with FBGs written in PMF results<br />
in the monitoring of the wavelength spacing between the two resonance bands [6-10]. This provides measurements<br />
of both the amplitude and direction of the transverse force thanks to the fiber-orientation dependence of the<br />
amplitude spectral response. Accurate measurements require a careful control of the state of polarization of the input<br />
signal at 45° between the fiber axes so that the two orthogonal modes exhibit the same optical power. Because the<br />
state of polarization is varied over the distance and the transversal force modifies the birefringence value, this<br />
requirement is difficult to ensure in practice so that important measurement errors can readily occur. To avoid this<br />
important complexity as well as the high cost linked to the use of PMF, an interesting alternative remains the use of<br />
FBGs written in standard SMF. For the reasons outlined above, amplitude spectral measurements of standard FBGs<br />
are not suited for transverse force sensing purposes. Recent researches have demonstrated that the polarization<br />
dependent properties related to standard FBGs can be used for this purpose [11]. Due to the lateral inscription<br />
process, photo-induced birefringence is present in FBGs written into standard SMF [12]. The birefringence value,<br />
typically of the order of 10 -6 , is too small to be perceived in the FBG amplitude spectral response. However, it leads<br />
to significant polarization-dependent properties such as polarization dependent loss (PDL) and differential group<br />
delay (DGD). The birefringence value can be mechanically enhanced by a transverse force [2], which increases both<br />
the PDL and DGD values. As a result, the PDL and DGD evolutions with wavelength can be monitored for<br />
transverse force sensing purposes, allowing to use FBGs written into standard SMF, which fail to work when they<br />
are interrogated by means of amplitude spectral measurements.<br />
The aim of this <strong>chapter</strong> is double: it first defines and analyzes the evolutions with wavelength of the normalized<br />
Stokes parameters, PDL and DGD associated to the transmitted signal of uniform FBGs, which completes the study<br />
of the amplitude and phase spectral characteristics presented in Ref. [13] when birefringence is taken into account. It<br />
then details the exploitation of the polarization related phenomena in FBGs for transverse force sensing purposes.<br />
In the following, Section 2 introduces some concepts of light polarization in optical fiber while Section 3 details and<br />
analyzes the polarization dependent phenomena in FBGs. Their dependence with respect to the FBG parameters and<br />
the birefringence value is identified in Section 4. Finally, the feasibility of using the FBGs polarization dependent<br />
properties for transverse strain sensing is demonstrated in Section 5.<br />
REVIEW OF SOME CONCEPTS OF LIGHT POLARIZATION IN OPTICAL FIBER<br />
This section reviews some important concepts of light polarization in optical fiber. These concepts are essentially<br />
the Jones and Stokes formalisms and will be useful for the definition of the polarization dependent properties<br />
associated to FBGs, which is the subject of the next section.<br />
State of Polarization<br />
A polarized lightwave signal propagating in an optical fiber or in free space is represented by electric and magnetic<br />
field vectors perpendicular to each other in a transverse plane itself perpendicular to the direction of propagation.<br />
Commonly, the attention is focused on the electric field since it has the most direct effect during the interaction<br />
between wave and matter [14].<br />
The state of polarization is defined as the pattern drawn in the transverse plane by the extremity of the electric field<br />
vector as a function of time at a fixed position in space. A monochromatic light source emits a single frequency and is<br />
always totally polarized. In some cases, the electric field vector can occupy random orientations in the transverse<br />
plane as a function of time. If the orientation changes are fast enough to be beyond observation in the physical context<br />
of a particular measurement or application, the light is said to be unpolarized [15]. It is the case of naturally produced
118 Fiber Bragg Grating Sensors Caucheteur et al.<br />
light, such as sunlight and firelight, which is characterized by a very large spectral width. In that case, the polarization<br />
behavior of each frequency is different and it is therefore impossible to clearly define a fixed polarization state.<br />
Between the cases of totally polarized and unpolarized light, we find partially polarized light for which the electric<br />
field vector is still characterized by random orientations but stays around a polarization state of maximum probability.<br />
It is typically the case when an optical source emits a quasi-monochromatic wave with a small but not null spectral<br />
width. Hence, each spectral component of the wave produces a different polarization state at a fixed point of space but<br />
all these states remain distributed around the same state.<br />
Partially polarized light can be represented as a superposition of both fully polarized and completely unpolarized<br />
lightwaves. The degree of polarization (DOP) describes partially polarized light and is defined as the ratio of the<br />
intensity of the totally polarized component to the total intensity of the wave. The degree of polarization varies from<br />
0 for unpolarized light to 1 for totally polarized light and takes intermediate values for partially polarized light. In<br />
free space propagation, the degree of polarization of a wave is maintained. In a transmission medium, it can evolve<br />
depending on the spectral width of the source and the dispersive properties of the transmission path.<br />
The Polarization Ellipse<br />
For a plane and monochromatic lightwave (DOP = 1) propagating along the z direction of a Cartesian coordinate<br />
system (x, y, z), the transverse components of the electric field can be expressed as [14]:<br />
Ex 0 x<br />
x<br />
(1)<br />
(z,t) E cos(ω<br />
t δ k z)<br />
Ey 0 y<br />
y<br />
(2)<br />
(z,t) E cos(ω<br />
t δ k z)<br />
where E0x and E0y are the amplitudes of the x and y components, x and y are the corresponding phases. The<br />
elimination of (·t – k·z) from Eqs. (1) and (2) yields the following relationship:<br />
E<br />
E<br />
2<br />
x<br />
2<br />
0x<br />
2<br />
Ey<br />
Ex<br />
Ey<br />
2<br />
2 cos δ sin δ<br />
(3)<br />
2<br />
E E E<br />
0 y<br />
0x<br />
0 y<br />
where = y – x is the phase difference between the two components.<br />
Eq. (3) is the equation of an ellipse drawn in the transverse plane by the extremity of the electric field vector at a<br />
fixed point in space as a function of time. The state of polarization of a fully polarized lightwave is therefore in<br />
general elliptical. The ellipse defined by Eq. (3) is illustrated in (Fig. 1) where one can see that parameters other than<br />
E0x, E0y and can be used to define a state of polarization. These parameters are the angle , called the azimuth,<br />
between the major axis and the x axis, the sense of rotation of the electric field vector materialized by an arrow on<br />
the ellipse and the ellipticity degree de defined as:<br />
Figure 1: Polarization ellipse.
Fiber Bragg Grating Sensors: Recent Advancements, Industrial Applications and Market Exploitation, 2011, 143-170 143<br />
Fiber Bragg Grating Sensors in Civil Engineering Applications<br />
Jinping Ou 1,2,* , Zhi Zhou 1,3 and Genda Chen 3<br />
Andrea Cusano, Antonello Cutolo and Jacques Albert (Eds)<br />
All rights reserved - © 2011 <strong>Bentham</strong> <strong>Science</strong> Publishers Ltd.<br />
CHAPTER 8<br />
1 School of Civil Engineering, Harbin Institute of Technology, Harbin, Heilongjiang, 150090, P.R. China; 2 School of<br />
Civil and Hydraulic Engineering, Dalian University of Technology, Dalian, 116024, P.R. China and 3 Center for<br />
Infrastructure Engineering Studies, Missouri University of <strong>Science</strong> and Technology, Rolla, MO 65409-0710, USA<br />
Abstract: Fiber Bragg Gratings (FBG) have been regarded as one of the most promising local monitoring sensors<br />
and are widely deployed in civil infrastructures. In this <strong>chapter</strong>, those FBG-based sensors aiming at civil<br />
structures have been briefly presented, including direct packaged strain and temperature sensors, indirect sensors<br />
constructed using FBG as a sensing element, and FBG based smart civil structures. Specific issues concerning<br />
methods of effectively and correctly applying the FBG sensors to civil infrastructures have been discussed. Those<br />
issues involve sensor installation technique, strain transfer based error modification, and temperature<br />
compensation. Finally, more than ten case studies of critical infrastructures outfitted with FBG sensors have been<br />
reported. Both research and practical applications show that FBG sensors are now competitive with conventional<br />
electrical sensors in long-term structural health monitoring.<br />
INTRODUCTION<br />
Civil infrastructure can be defined as a network of large scale structures with long service lives, existing in harsh<br />
operating environments with limited maintenance budget. For example, long-span bridges are often constructed to<br />
link ground transportation networks across rivers for 75 to 100 years. During a typical lifespan, civil infrastructure is<br />
inevitably subjected to environmental effects, long-term loading effects, material deterioration and/or extreme<br />
loading effects such as corrosion, fatigue, aging, and/or earthquakes and hurricanes. Predictably, the accumulated<br />
damage on a structure will result in the degradation of structural performance; the structural loading resistance can<br />
be greatly lowered, possibly leading to disasters in cases of extreme loading. Therefore, Structural Health<br />
Monitoring (SHM) has recently become an important innovative technology for both the detection and<br />
characterization of a damage process. Understanding this process will ensure the ability to maintain healthy<br />
conditions for civil infrastructures [1].<br />
China has constructed a number of long-span bridges annually since the early 1990s. For example, the Sutong cablestayed<br />
bridge has a main span of 1088 m, and the Hanzhou Bay Bridge is 36 km in length. Similarly, other<br />
structures, such as the Olympics Swimming Center, Bird Nest, Three Gorges Dam, Guangzhou New TV Tower, and<br />
Tibet Railway have also been rapidly developed. In the near future, more skyscrapers and long-span structures<br />
expects to be constructed. Maintaining structural health throughout their lifespan and mitigating potential<br />
catastrophes will be a major challenge.<br />
Traditional sensors such as electrical strain gauges and vibration wires are unable to meet the ever-increasing demand<br />
for long-term monitoring of large-scale infrastructure in terms of durability. In recent years, Optical Fiber (OF)<br />
sensors have been used for long-term SHM due to their distinct advantages, such as electromagnetic immunity,<br />
compact size, corrosion resistance, and the ability for a multiplexed sensor array to be distributed along a single fiber.<br />
To date, Fiber Bragg Grating (FBG) has attracted the most attention in various SHM applications in composite<br />
structures [2-5]. The technology has been validated and used for case studies on more than 100 real-world structures.<br />
In 2003, approximately 1,700 FBG sensors were installed on the Dongying Yellow River Bridge for long-term SHM,<br />
making it the largest FBG case study. In 2006, 179 FBG strain sensors and 24 temperature sensors were used to monitor<br />
stress and temperature levels during the construction of a national swimming center in China. FBG sensors were<br />
also attached to rails along a railway to monitor dynamic strain based on the Wavelength Division<br />
Multiplexer/demultiplexer (WDM) [6]. FBG temperature sensors were also integrated into a fire alarm system in<br />
*Address correspondence to this author at Professor Jinping Ou: School of Civil Engineering, Harbin Institute of Technology, Harbin,<br />
150090, P.R. China; Tel: +8645186282209; Fax: +8645186282209; E-mail: oujinping@hit.edu.cn
144 Fiber Bragg Grating Sensors Ou et al.<br />
highway tunnels at Luliang Mountain, Baiyan Brook, and the Zhang Jiachao Tunnel in west Shanghai-Chengdu<br />
Highway [7]. Furthermore, FBG strain and temperature sensors were applied to monitor offshore platform Numbers<br />
CB371 and CB32A [8-9].<br />
In this <strong>chapter</strong>, some of the latest advances in FBG sensing technologies concerning long-term monitoring of largescale<br />
infrastructure are reported in two parts. The first part will introduce the latest progress in the development of<br />
direct FBG-based sensors, indirect FBG based sensors, and FBG-based smart structures. The second part will<br />
demonstrate various applications of the FBG sensors and smart structures in more than 10 case studies on critical<br />
infrastructures.<br />
FBG-BASED SENSORS FOR CIVIL INFRASTRUCTURES<br />
Direct OF Sensors<br />
Packaged Strain Sensors<br />
Due to their fragile behavior in a harsh environment, bare FBG sensors cannot be directly placed in civil<br />
infrastructure without proper packaging. Considering the creep and aging effect of glues and the corrosion of metals,<br />
Fiber Reinforced Polymer (FRP) has been extensively studied and successfully applied for FBG and OF packaging.<br />
FRP is an elastic and durable composite material with superior fatigue behavior. FRP is naturally compatible with<br />
optical fiber material because both have a key element of silica. (Fig. 1) shows a Scanning Electron Microscope<br />
(SEM) view of the bonding surface between FRP and bare OF. Through the cross section analysis under axial<br />
strains, the strain transfer error and the error correction coefficient are approximately 3.3% and 1.03, respectively,<br />
for Glass FRP (GFRP) rods of 4~8 mm in diameter, and those for Carbon FRP (CFRP) bars of 4~8 mm in diameter<br />
are approximately 2% and 1.1, respectively, indicating that the FRP is well bonded to the OF. Therefore, FRP is<br />
well qualified for both FBG and OF packaging. Additionally, because FBG or OF sensors with a diameter of<br />
125 μm or less will not affect the mechanical properties of FRP elements, the FRP packaging can be integrated with<br />
bars and plates to promote load capacity. (Fig. 2) validates this theory.<br />
(a) (b)<br />
Figure 1: Interface SEM of bare OF and GFRP/CFRP: (a) GFRP and (b) CFRP.<br />
In order to make full use of the superior durability and elasticity of the FRP material, various FRP-packaged FBG<br />
strain sensors has been developed as represented by those in (Fig. 3). These sensors show excellent characteristics,<br />
including: 1) a strain measurement range of up to 8000~15000 , 2) a strain measurement resolution of 1~2 ,<br />
depending on the FBG interrogator, 3) exceptional precision, with deviations of less than 0.5%, 4) a normalized<br />
sensitivity coefficient of 7.8×10 -7 , 5) a hysteresis error of less than 0.5%, and 6) a fatigue life of over 200 million<br />
cycles at 30001000 . After 12 months of corrosion tests, the FRP-packaged sensors have the same sensing<br />
capabilities as the original ones. They can be designed for various needs and specifically are appropriate in longterm<br />
SHM for large-scale infrastructure. In addition, for a strain measurement of over 15,000 with a resolution of<br />
less than 0.5 , a sensitivity-increasing or decreasing technology based on strain transfer mechanisms has been<br />
developed for various applications [10-13].
Fiber Bragg Grating Sensors in Civil Engineering Applications Fiber Bragg Grating Sensors 145<br />
Stress(MPa)<br />
700<br />
600<br />
500<br />
400<br />
300<br />
200<br />
100<br />
CFRP1<br />
CFRP2<br />
CFRP-FBG1<br />
CFRP-FBG2<br />
0<br />
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000<br />
Strain(X10 -6 )<br />
(a) (b)<br />
Figure 2: Strength comparison of FRP-FBG bar and FRP bar under loading: (a) CFRP and (b) GFRP.<br />
(a)<br />
(c)<br />
Figure 3: Representative FRP-packaged FBG strain sensors: (a) overall embeddable, (b) end-enlarged, (c) weldable, and<br />
(d) long-gauge.<br />
To address the large-scale nature of infrastructure, an integrated global and local SHM technology with a single OF<br />
was developed by combining the Brillouin Optical Time Domain Analysis/Reflectometry (BOTDA/R) with FBG<br />
sensors. The BOTDA/R and FBG are for distributed and local high-precision strain measurements, respectively. The<br />
combined sensing heads can be easily integrated using the FRP-package method and applied at a length of over<br />
5 km, as depicted in (Fig. 4). (Fig. 5) shows that the strain measured with BOTDA/R is in agreement with that from<br />
FBG (marked with dots) as long as the center wavelength of the FBG is far away from the BOTDA/R’s operation<br />
wavelength of 1550 nm. For practical applications, the BOTDA/R-FBG sensor is installed on a structure so that the<br />
Bragg gratings are located in potential damage areas based on a pre-analysis and assessment of the structure. The<br />
actual location of damage areas and their spatial distribution can be identified by the BOTDA/R measurements,<br />
while the detailed damage data can be acquired through the FBG sensors [14]. (Fig. 6) shows some other metalpackaged<br />
FBG sensors developed by Micron Optics, Inc. (MOI). They are examples of products with the advantages<br />
of fast response time, high accuracy, long-term stability and premium performance under harsh environmental<br />
conditions.<br />
Stress(MPa)<br />
500<br />
450<br />
400<br />
350<br />
300<br />
250<br />
200<br />
150<br />
100<br />
50<br />
0<br />
GFRP-FBG1<br />
GFRP-FBG2<br />
GFRP1<br />
GFRP2<br />
-50<br />
0 1000 2000 3000 4000 5000 6000<br />
(b)<br />
(d)<br />
Strain(X10 -6 )
Fiber Bragg Grating Sensors: Recent Advancements, Industrial Applications and Market Exploitation, 2011, 171-184 171<br />
Fiber Bragg Grating Sensors in Aeronautics and Astronautics<br />
Nobuo Takeda 1,* and Yoji Okabe 2<br />
Andrea Cusano, Antonello Cutolo and Jacques Albert (Eds)<br />
All rights reserved - © 2011 <strong>Bentham</strong> <strong>Science</strong> Publishers Ltd.<br />
CHAPTER 9<br />
1 Department of Advanced Energy, Graduate School of Frontier <strong>Science</strong>s, The University of Tokyo, Mail Box 302, 5-<br />
1-5 Kashiwanoha, Kashiwa-shi, Chiba 277-8561, Japan and 2 Department of Mechanical and Biofunctional Systems,<br />
Institute of Industrial <strong>Science</strong>, The University of Tokyo, 4-6-1 Komaba, Meguro-ku, Tokyo 153-8505, Japan<br />
Abstract: Fiber optic sensors, Fiber Bragg Grating (FBG) sensors in particular, are among most promising<br />
sensors for structural health monitoring of aerospace structures in order to assess the safety and durability during<br />
a long period of time in use. These sensors are expected to provide a low-cost maintenance methodology for<br />
carbon fiber reinforced composite structures which are now extensively being used for the primary structures.<br />
This <strong>chapter</strong> presents an overview of current use of FBG sensors in aeronautics and astronautics.<br />
INTRODUCTION<br />
In the aerospace filed, the weight saving of the structures is one of the important demands to increase the flight<br />
performance of airplanes and the mission operability of spacecrafts. Therefore lightweight composite materials, such<br />
as carbon fiber reinforced plastics (CFRP), have been applied to most main members of aerospace structures in<br />
recent years. Whereas the composite materials actualize lightweight high-performance structures, the fracture<br />
process of the materials is really complex due to the combination of different constituents of reinforcing fibers and<br />
matrix. Hence structural health monitoring (SHM) technologies are attracting attention to increase the reliability and<br />
safety in the lightweight design of the aerospace structures.<br />
In the SHM field, currently, the methods using piezoelectric actuators/sensors for generating and detecting<br />
ultrasonic waves and optical fiber sensors for static and dynamic strain sensing are mainly researched. Ultrasonic<br />
propagation techniques use the response of the waves to the geometric change caused by the damage occurrence. On<br />
the other hand, optical fiber sensors basically measure strain with high accuracy in order to detect damages in<br />
structural members. Among the many optical fiber sensors, fiber Bragg grating (FBG) sensor is one of the most<br />
promising sensors for SHM, because the FBGs can measure the strain with a high degree of accuracy and reliability<br />
and are also easy to handle.<br />
Excellent summary of fiber optic sensors including FBG sensors to be used in aerospace applications can be found<br />
in Refs. [1-3]. In this <strong>chapter</strong>, some recent detailed researches using FBG sensors for SHM of aerospace structures<br />
are reviewed based on the authors’ previous review article [4].<br />
SHM OF AIRCRAFT WITH FBG SENSORS<br />
Most researches on SHM with FBG sensors in the aerospace field targets aircrafts in order to decrease the<br />
maintenance cost and increase the safety of the structures. FBG sensors can be easily integrated into structural<br />
materials of aircrafts because optical fibers are thin, lightweight, and flexible.<br />
As an example of early fundamental researches, Wood et al. embedded FBG sensors into CFRP laminates for a<br />
future morphing aircraft and measured the Bragg wavelength using their demodulation system that determined the<br />
locations of multiplexed FBGs by Fourier transformation. The strain of the laminate was successfully measured with<br />
the embedded FBG [5]. They also attempted to combine the single mode optical fiber for the FBG sensor with near<br />
infrared spectroscopy to measure a variety of chemical and physical changes in the airframe structure using a single<br />
embedded fiber [6].<br />
*Address correspondence to this author Professor Nobuo Takeda: Department of Advanced Energy, Graduate School of Frontier <strong>Science</strong>s,<br />
The University of Tokyo, Mail Box 302, 5-1-5 Kashiwanoha, Kashiwa-shi, Chiba 277-8561, Japan; Tel: +81-3-5841-6642; Fax: +81-3-5841-<br />
6642; Email: takeda@smart.k.u-tokyo.ac.jp
172 Fiber Bragg Grating Sensors Takeda and Okabe<br />
ANALYSIS/EVALUATION<br />
(by Kawasaki Heavy Industries, Ltd.)<br />
EMBEDDED SMALL-DIAMETER<br />
OPTICAL FIBER SENSORS<br />
VISUALIZATION<br />
(by Univ. Tokyo)<br />
Figure 1: Arrangement of embedded small-diameter FBG sensors in upper panel of composite fuselage demonstrator and the<br />
impact detection/localization system.<br />
The precise strain measurements with FBG sensors have been applied to detection of damages in aircraft structures.<br />
Guemes et al. bonded or embedded some FBGs into three blade-stiffened CFRP panels with co-cured stiffener webs<br />
[7]. The sensors successfully measured the strain during the buckling behavior of the stiffened panels under the<br />
compression loading. In the stiffened CFRP panels, debonding between the skin panel and the stiffener is a typical<br />
damage. They also detected the debonding damage based on the response of the FBG sensor to the non-uniform<br />
strain distribution in the grating length [8].<br />
Other typical structural panel used in the aerospace structures is a sandwich panel that consists of a lightweight thick<br />
core and two thin high-stiffness facesheets to increase the bending stiffness of the lightweight panel. However, since<br />
the sandwich panels are vulnerable to out-of-plane impact loadings, Bocherens et al. embedded FBG sensors and<br />
multimode optical fibers for the optical time domain reflectometry (OTDR) in a radome sandwich structure for<br />
detection of impact damages [9]. These SHM methods were indicated to be efficient in detecting permanent damage<br />
induced by the impacts.<br />
The FBG strain measurement is also applied to monitoring of the deformation behavior of wing structures. When the<br />
flight speed of aircrafts exceeds a limit, the oscillation of the wing panel diverges and leads to destruction of the<br />
wing. This hazardous phenomenon is called flutter. If the flutter can be detected and suppressed by smart material<br />
systems, the limitation of the flight speed can be increased. Lee et al. embedded FBG sensors in the skin panel of a<br />
1/25th scale wing model of a commercial airliner and conducted a wind tunnel testing [10]. Using their<br />
demodulation system with a wavelength swept fiber laser, the flutter was able to be detected. The deformation<br />
monitoring is also needed for development of “morphing wing” that is a new concept of wing structures. The<br />
morphing wing changes the wing area, sweepback angle, and camber smoothly to maximize the flight performance.<br />
In order to control the wing shape precisely, the shape sensing of the morphed wing is important. Hence Tai<br />
investigated theoretically the applicability of multiple FBG sensors to estimate the shape of the morphed wing [11].<br />
Moreover, modal parameters in vibration are often sensitive to damages in structures. Some attempts have been<br />
conducted using high-speed FBG interrogation systems [12-13]. Also, vibration observations with FBGs are
Fiber Bragg Grating Sensors in Aeronautics and Astronautics Fiber Bragg Grating Sensors 173<br />
combined with smart actuator systems to control the vibrations. For example, the strain measurement with an FBG<br />
was used to suppression of the vortex-induced vibration of a cylinder by piezo-ceramic THUNDER actuators [14].<br />
In a MESEMA project in the aeronautics field, a smart magnetostrictive auxiliary mass damper equipped with an<br />
FBG sensor was developed [15]. This smart device was mounted on a real stiffened aeronautical panel and<br />
succeeded in the reduction of the vibration level.<br />
When the FBG sensors are embedded into composite materials, connection method to the embedded optical fibers is<br />
also an important issue. In order to overcome this problem, various optical fiber connecters integrated in the<br />
composite laminates have been developed [16-17]. The method using polytetra-fluorethylene dummies enabled<br />
trimming of the composite parts with embedded optical fiber connection hardware [17].<br />
At the same time as the above fundamental researches, investigations on practical application to actual aircrafts have<br />
been performed. Daimler Chrysler glued FBG sensors to the surface of a newly developed CFRP wing structure and<br />
tested the sensing performance at the DASA Airbus test center Hamburg for over one year from September 1998 to<br />
December 1999 [18]. They also attempted to embed the sensors into the varnish of structures for protection of the<br />
optical fibers. Then Airbus Espana had evaluated the feasibility of FBGs embedded in CFRP structures for<br />
measurement of manufacturing induced residual stresses and SHM of the large structure [19]. EADS Deutschland<br />
GmbH bonded FBG sensors on the surface of a CFRP aircraft fuselage test barrel and measured the strain change<br />
during the impact and the permanent deformation caused by the impact damage [20]. Alenia Aeronautica in Italy<br />
also has investigated the main requirements for FBG sensors and the application possibility to SHM of aerospace<br />
structures [21]. A micro-size interrogation system is under development for aircraft use [22].<br />
Damage identification has been conducted in some Japanese national projects on fiber optic sensors in aircraft<br />
applications [23]. Development of real-time detection of impact damage with embedded small-diameter optical fiber<br />
sensors was conducted as a collaborative work between the University of Tokyo and Kawasaki Heavy Industries and<br />
demonstrated in a CFRP fuselage structure of 1.5 m in diameter and 3 m long. In the upper panel, the small-diameter<br />
sensors were embedded in order to detect impact-induced damages (Fig. 1) [24-25]. The small-diameter FBG<br />
sensors were used to obtain the impact location through the dynamic strain measurement, and multi-mode smalldiameter<br />
optical fibers were embedded to judge the occurrence of the impact-induced damages using the optical loss<br />
due to bending. An impact detection and localization system was also developed and could successfully detect the<br />
impact locations and impact-induced damages. A special embedded optical connector was developed for a smalldiameter<br />
optical fiber so that it could be trimmed after the autoclave composite fabrication and then connected to a<br />
normal-diameter optical fiber. More details on small-diameter FBG sensors can be found in reference [26].<br />
Another example is a grid structure (Fig. 2) made of CFRP unidirectional composites, named an advanced grid<br />
structure (AGS), has specific characteristics such as simplicity of stress path/damage feature and fail-safe structural<br />
redundancy [27]. The HRAGS system equipped with an SHM system utilizing FBG sensors embedded in every rib<br />
of AGS has been proposed, so that the size and intensity of operational or accidental damages can be evaluated<br />
through the strain measurement of every rib. A recent advanced six-axis-controlled tape placement machine can be<br />
utilized to place CFRP unidirectional tapes and optical fibers with FBG sensors. Recent high-speed optical switch<br />
can also be used to scan all the strain data from a number of FBG sensors. A statistical damage recognition system<br />
for SHM was established to distinguish the damage location from strain data [28].<br />
When new demodulation systems of FBG sensor were developed, their practicality was sometimes demonstrated by<br />
mounting the system in airplanes and monitoring the strain by FBG during the flight. When BAE Systems<br />
developed their FBG sensor system, the system was equipped in the cabin of a test aircraft and FBG strain rosettes<br />
were bonded to the external lower surface of the wing [29]. The flight tests over a two-week period successfully<br />
demonstrated that the FBG sensor system was capable of measuring strains during the actual operation of airplanes.<br />
Recently, Technobis Fiber Technologies developed a high-speed multi sensor FBG interrogator, and they conducted<br />
a flight test using a test aircraft [30]. The system succeeded in measurement of the stress and the temperature of the<br />
aircraft structure during the flight.<br />
As other examples of SHM researches relating to the aircrafts, the dynamic behavior of the main rotor blade in a<br />
helicopter [31] and that of a parachute during inflation [32] were evaluated using FBG sensors.
Fiber Bragg Grating Sensors: Recent Advancements, Industrial Applications and Market Exploitation, 2011, 185-196 185<br />
Fiber Bragg Grating Sensors in Energy Applications<br />
Christopher Barry Staveley*<br />
Smart Fibres, Limited, 12 The Courtyard, Eastern Road, Bracknell, RG12 2XB, United Kingdom<br />
Andrea Cusano, Antonello Cutolo and Jacques Albert (Eds)<br />
All rights reserved - © 2011 <strong>Bentham</strong> <strong>Science</strong> Publishers Ltd.<br />
CHAPTER 10<br />
Abstract: Fiber Bragg Grating sensing systems have found application in numerous market sectors, wherein the<br />
unique capabilities of the technology are either displacing existing, limited sensing technologies, or providing<br />
solutions where sensing was hitherto impractical or impossible. Here, we review the energy market for the<br />
technology, a market where some of the most challenging measurement conditions are presented, yet some of the<br />
largest value propositions can be seen.<br />
INTRODUCTION<br />
Fiber Bragg grating (FBG) technology came to light toward the end of the 1970s. Through the remainder of the 20 th<br />
century, the development of the technology almost exclusively focused on telecoms applications. In the 1990s a few<br />
pioneering Companies recognized the potential sensing opportunities of the technology and began to develop<br />
products towards the markets that might emerge to utilize the same.<br />
After just over a decade of this shift in focus towards sensing, FBG measurement systems are finding increasing<br />
application in a number of industry sectors for design optimization, smart control of active systems, and structural<br />
health monitoring. Of these industry sectors, which include aerospace, civil infrastructure, transportation and<br />
medical, the single largest one is energy.<br />
ENERGY APPLICATIONS<br />
The generation, harvesting, conversion, transportation and storage of energy utilizes an enormous range of industries<br />
and scientific disciplines, and represents one of the single largest global market sectors. As shown in (Fig. 1), the<br />
market is projected to continue growing at a significant rate for the foreseeable future,<br />
Figure 1: Projected growth in global energy demand (source: IEA WEO 2007).<br />
The industry is very diverse, reflecting the diversity of the forms of energy that humankind relies upon in everyday<br />
life, from ancient hydrocarbons that are mined or pumped from subterranean reserves, though nuclear reactions<br />
carried out within safety critical containment systems, to naturally renewable forms of energy such as hydroelectric,<br />
solar, wind and wave.<br />
*Address correspondence to this author Christopher Barry Staveley: Smart Fibres, Limited, 12 The Courtyard, Eastern Road, Bracknell,<br />
RG12 2XB, United Kingdom; Tel: +44 1344 484111; Fax: +44 1344 423241; Email: info@smartfibres.com
186 Fiber Bragg Grating Sensors Christopher Barry Staveley<br />
The forms of distribution of energy are equally diverse, from high power electric cables feeding transformer<br />
stations, to pipelines and ocean vessels carrying hydrocarbons, either in their stable state or super-cooled to become<br />
easier to handle.<br />
It is no wonder then, that the applications found by FBG technologies in this huge, diverse market are themselves<br />
diverse also. This brief summary lists some of these applications, and cites some noteworthy examples of them<br />
being developed. It is, however, far from exhaustive, and seeks only to capture a snapshot in time of this quickly<br />
growing and dynamically changing field of sensing.<br />
Table 1 lists a number of applications that can be found within the energy industry where FBG technology is either<br />
already established or showing promise for future adoption.<br />
Table 1: Example FBG applications in the energy industry.<br />
Upstream Oil and Gas<br />
Multi-drop downhole pressure and temperature sensing for intelligent well optimization<br />
Structural health monitoring of offshore structures<br />
Condition monitoring of subsea pumps<br />
4D seismic monitoring<br />
Midstream Oil and Gas<br />
Pipeline integrity monitoring<br />
Structural health monitoring of LNG carrier cargo containment tanks<br />
Hull stress monitoring of product carriers<br />
Fossil Fuel Power generation<br />
Temperature sensing of power plant boilers and generators<br />
Wind Energy<br />
Structural health monitoring of rotorblades<br />
Independent pitch control of rotorblades<br />
Condition monitoring of rotating turbine machinery<br />
Structural health monitoring of turbine tower and foundations<br />
Tidal Energy<br />
Blade, drivetrain and marine life impact monitoring<br />
Nuclear<br />
In-core temperature measurement of nuclear reactors<br />
Structural health monitoring of containment and cooling infrastructure for nuclear power plants<br />
Electricity Transmission<br />
Power transformer hot spot monitoring<br />
Health and usage monitoring of power transmission and distribution systems<br />
UPSTREAM OIL AND GAS<br />
Upstream is a term commonly used to refer to the searching for and the recovery and production of crude oil and<br />
natural gas. The upstream oil sector is also known as the exploration and production (E&P) sector. There is one fiber<br />
optic sensing technique that is already well established and recognized as a valuable tool in the upstream sector.<br />
Distributed fiber optic sensors (based on Raman or Brillouin light scattering), measure a change in light wave<br />
scattering at different frequencies due to displacements within an optical fiber. The fiber itself is used for sensing
Fiber Bragg Grating Sensors in Energy Applications Fiber Bragg Grating Sensors 187<br />
and, in the case of distributed temperature sensors, also tends to be used for data transmission. Such distributed<br />
sensing systems employ only a single fiber, and can interrogate the fiber over distances as great as around 30 km.<br />
The energy markets for distributed temperature sensing (DTS) are well served by a number of global service<br />
providers including Sensa, Sensornet, Lios, AP Sensing, Omnisens and Sensortran. The experiences gained by the<br />
upstream industry over the past decade with these DTS technologies, particularly for in-well measurements, has<br />
been very useful in paving the way for FBG technology to follow. As a result, two particular applications for FBG<br />
technology in upstream oil and gas have emerged.<br />
Multi-Drop Downhole Pressure and Temperature Sensing for Intelligent Well Optimization<br />
DTS is primarily used in-well for early detection of problems with injectors, producers and downhole pumps, flow<br />
profiling, and well integrity monitoring. One particular DTS campaign is described by Simonits and Franzen [1],<br />
whilst new applications continue to arise such as fracturing optimization, as reported by Shell [2].<br />
In recent years, the benefits offered by adding multi-point pressure and temperature (P/T) gauges in providing higher<br />
accuracy measurements at certain points have become apparent.<br />
Electronic pressure sensors (for example, quartz resonator gauges and piezoresistive sensors) have key limitations in<br />
downhole monitoring applications, such as low MTBF, inability to readily and cost-effectively take multi-point<br />
measurements, and limited temperature capability. FBG based pressure sensors have no downhole electronics and so<br />
do not suffer from these limitations. Instead they offer the advantages of longer operating lifetimes at elevated<br />
temperatures, intrinsic safety, the ability to multiplex and run in hole many tens of sensors on a single optical fiber<br />
cable, and the ability to deploy sensors many tens of kilometers away from a surface electronic readout unit without<br />
the need for signal amplification.<br />
This has been recognized by, for instance, Weatherford, who added several optical pressure and temperature gauges<br />
to a DTS installation in offshore Brunei, as reported by Pallanich [3].<br />
Bani et al. [4] reported on an intelligent completions project carried out by Saudi Aramo, Baker Oil Tools and<br />
Weatherford International in which numerous FBG-based P/T gauges deployed demonstrated excellent long-term<br />
field performance, providing real-time zonal P/T for production monitoring and well diagnostics.<br />
Another example is a joint project between Smart Fibres and Shell, who have collaborated since 2003 to develop a<br />
low-cost, multi-point FBG P/T gauge technology known as SmartCell®. The objective of the project was to realize<br />
an FBG-based P/T sensing system with a price/performance ratio that represented a value proposition for the<br />
majority of operating wells, even in brownfield reserves with declining production.<br />
The development culminated in the downhole deployment pictured in (Fig. 2) of a string of 9 SmartCell P/T gauges<br />
in the Natih field of Shell operating unit Petroleum Development Oman [5-6].<br />
Figure 2: Installation of SmartCell Multi-Point FBG P/T System (Shell / Smart Fibres) [5-6].
Fiber Bragg Grating Sensors: Recent Advancements, Industrial Applications and Market Exploitation, 2011, 197-217 197<br />
Fiber Bragg Grating Sensors for Railway Systems<br />
Hwa-yaw Tam*, Shun-yee Liu, Siu-lau Ho and Tin-kin Ho<br />
Andrea Cusano, Antonello Cutolo and Jacques Albert (Eds)<br />
All rights reserved - © 2011 <strong>Bentham</strong> <strong>Science</strong> Publishers Ltd.<br />
CHAPTER 11<br />
Photonics Research Centre, Department of Electrical Engineering, The Hong Kong Polytechnic University, Hung<br />
Hom, Kowloon, Hong Kong SAR, China<br />
Abstract: Fiber Bragg Grating (FBG) sensor technology has been attracting substantial industrial interests for the<br />
last decade. FBG sensors have seen increasing acceptance and widespread use for structural sensing and health<br />
monitoring applications in composites, civil engineering, aerospace, marine, oil & gas, and smart structures. One<br />
transportation system that has been benefited tremendously from this technology is railways, where it is of the<br />
utmost importance to understand the structural and operating conditions of rails as well as that of freight and<br />
passenger service cars to ensure safe and reliable operation. Fiber-optic sensors, mostly in the form of FBGs,<br />
offer various important characteristics, such as EMI/RFI immunity, multiplexing capability, and very long-range<br />
interrogation (up to 230 km between FBGs and measurement unit), over the conventional electrical sensors for<br />
the distinctive operational conditions in railways. FBG sensors are unique from other types of fiber-optic sensors<br />
as the measured information is wavelength-encoded, which provides self-referencing and renders their signals<br />
less susceptible to intensity fluctuations. In addition, FBGs are reflective sensors that can be interrogated from<br />
either end, providing redundancy to FBG sensing networks. These two unique features are particularly important<br />
for the railway industry where safe and reliable operations are the major concerns. Furthermore, FBGs are very<br />
versatile and transducers based on FBGs can be designed to measure a wide range of parameters such as<br />
acceleration and inclination. Consequently, a single interrogator can deal with a large number of FBG sensors to<br />
measure a multitude of parameters at different locations that spans over a large area.<br />
FBG is the most promising, cost-effective and distributed sensor technology that provides an ideal platform to<br />
monitor the condition and structural health of tracks, carriages and underframe equipment in railway systems. In<br />
the last few years, a number of field trial railway projects using FBG sensors for axle counting, track and train<br />
vibration measurements, monitoring of bogie conditions, structural health monitoring of train bodies, and<br />
interaction between overhead contact lines and current collectors (pantograph) were successfully conducted by a<br />
few research institutions. These studies demonstrated the superiority of FBG sensors over conventional sensors in<br />
many crucial aspects. However, major barriers, such as lack of proprietary and custom specifications, packaging<br />
and reliability standards, insufficient field experience, have yet to be resolved before major railway operators are<br />
to embrace the FBG sensor technology.<br />
INTRODUCTION<br />
The rail industry is enjoying its biggest development boom worldwide in recent years. Fuelled by growing trade and<br />
rising environmental concerns on road transportation, the United States invested nearly US$ 10 billion in 2008 and<br />
allocated US$ 8 billion for high-speed trains in 2009. Indian Railways will invest about US$ 50 billion under the<br />
11 th Five Year Plan to modernize its railway system. As part and parcel of China’s rapid economic rise to become a<br />
modern nation, the scale of railway investment in China is gargantuan. In 2009, China spent US$ 50 billion on its<br />
high-speed rail system with a top speed of up to 350 km/hr. By 2020, China will have added more than 25,000 km of<br />
high-speed tracks and spend up to US$ 300 billion. Undoubtedly, the improvement of safety, reliability and<br />
productivity will continue to be the most important directive for the railway industry. This can be achieved through<br />
advances in on-board computers, on-board train condition monitoring systems, and wireless data transmission from<br />
wayside monitoring systems. There is also an increasing demand for better system reliability, availability,<br />
maintainability and safety from the communities. A smart condition monitoring system allows real-time and<br />
continuous monitoring of the structural and operational conditions of trains [1], overhead contact lines [2-3], as well<br />
as monitoring of the structural health of rail tracks and the location, speed and weight of passing trains of the entire<br />
rail systems. Ultimately, the inclusion of train location, speed restrictions, and train, track and overhead contact line<br />
conditions to the ‘intelligent systems’ will herald a safer railway industry with reduced maintenance cost, optimized<br />
*Address correspondence to this author Professor Hwayaw Tam: Photonics Research Centre, The Hong Kong Polytechnic University, Hung<br />
Hom, Kowloon, Hong Kong SAR, China; Tel: +852 27666175; Fax: +852 23301544; Email: eehytam@polyu.edu.hk
198 Fiber Bragg Grating Sensors Tam et al.<br />
performance and capacity. Therefore, the need of a smart condition monitoring system is imminent as indicated by<br />
the increase in railway/underground accidents/incidences around the world. Smart condition monitoring systems for<br />
the railway industry would require extensive sensor networks with large number (1,000s’) of multifunctional sensors<br />
for the measurements of temperature, strain/stress, vibration, acceleration, etc. Fiber Bragg grating sensors, in<br />
comparison to electrical sensors and other types of fiber-optic sensors, offer many advantages that are particularly<br />
well suited to railway transportation systems. These include immunity to EMI/RFI, long life-time (>20 years), and<br />
massive multiplexing capability – hundreds of sensing points along a single strand of optical fiber with length up to<br />
230 km [4], fast measurement speed, self-referencing and inherent redundancy feature.<br />
Quasi-distributed fiber-optic sensor based on fiber Bragg gratings (FBGs) is an excellent candidate for the<br />
realization of smart condition monitoring systems for the railway industry. There are more established distributed<br />
fiber-optic sensors based on Raman or Brillouin optical time-domain reflectometry but they are less suitable for<br />
condition and structural monitoring of railways which demand fast measurement time and high spatial resolution.<br />
In this <strong>chapter</strong>, the important characteristics of distributed photonic sensors, potential applications for the railway<br />
industry and some field trials will be described. Some of the FBG sensor-based monitoring systems are fully<br />
operational and in present service use – are providing valuable information about stresses experienced during<br />
service, both static and dynamic, under different operational conditions. The sensors also provide information on the<br />
loading and traffic status of the passenger cars; temperature-induced stresses and deformations on rails and<br />
carriages; temperatures in and around axles and wheel brakes; dynamic axle vibrations due to corrosion and bearing<br />
wear; and many other parameters relevant to railroad health monitoring and integrity.<br />
DISTRIBUTED FIBER-OPTIC SENSOR TECHNOLOGIES<br />
In distributed photonic sensor systems, a single fiber is used as the sensing element to substitute for thousands of<br />
conventional point sensors. The concept of distributed sensing was initially promoted in the late 70’s by optical<br />
fibers based on Rayleigh back scattering mechanism and through the technique of optical time-domain reflectometry<br />
(OTDR) [5] for locating faults in optical fibers, up to 250 km with a resolution of ~3 meters.<br />
Figure 1: Principle of optical time-domain reflectometer for distributed measurement along an optical fiber.<br />
(Fig. 1) shows the basic configuration of an OTDR in which a short pulse (~10 ns) of light is launched into a test<br />
fiber. The back-scattered spectrum due to Rayleigh, Brillouin and Raman is also shown. The position of the<br />
measured quantity is computed via the time of flight of the backscattered light pulse propagating in the fiber. In the<br />
late 80’s, a number of sensor configurations were proposed, mostly originated from the Raman and Brillouin<br />
scatterings in optical fibers [6-11]. Raman distributed sensing is based on the spontaneous scattering process<br />
generated by thermally-activated acoustic waves. Information about temperature is retrieved from the comparison<br />
between the intensities backscattered into the Stokes and the Anti-Stokes waves. Raman distributed sensor is now<br />
very mature and commercial units have typical performance of 1 K temperature accuracy with 1 m spatial resolution<br />
for fiber lengths up to 10 km. The measurement time varies from 1 minute to 10 minutes, depending on the required<br />
accuracy and spatial resolution. On the other hand, Brillouin OTDR does not base on intensity measurement, but the
Fiber Bragg Grating Sensors for Railway Systems Fiber Bragg Grating Sensors 199<br />
frequency shift of the Brillouin scattered wave. The Brillouin frequency shift depends on both temperature and<br />
strain, and the backscattered power depends solely on temperature. Brillouin OTDR can thus be used to measure<br />
temperature and strain simultaneously.<br />
Commercial Brillouin OTDR has the capability of measuring distributed strain and temperature with a resolution of<br />
20 µε and 1 K respectively, and with 1 m spatial resolution over a distance of 30 km in 1 minute. Distributed<br />
photonic sensors based on Raman and Brillouin OTDRs offer many advantages over conventional sensors in<br />
applications where a large number of sensing points is required and the environment is hazardous. In addition,<br />
optical fibers are non-conductive, non-corrosive, unaffected by EMI and RFI, low loss and small size. A common<br />
application of Raman OTDR is in the measurement and identification of hot spots along power transmission lines.<br />
Brillouin OTDRs are being employed in locating leaks in oil-pipe lines. However, the Raman and Brillouin<br />
distributed sensing systems require long measurement time and generally exhibit spatial resolution of the order of<br />
meter. Consequently, they are not suitable for applications where fast response time is needed or the required<br />
sensing regions are small.<br />
On the contrary, fiber Bragg grating sensors [12-13] can be interrogated at very high-speed of up to 2.5 Msamples/s<br />
[14]. FBGs are very small – short length of single-mode fiber (down to 0.1 mm) with periodic refractive-index<br />
variation in its 9-μm core, as shown in (Fig. 2a). FBGs can be created to reflect narrow-bands of spectrum (typically<br />
218 Fiber Bragg Grating Sensors: Recent Advancements, Industrial Applications and Market Exploitation, 2011, 218-237<br />
Fiber Bragg Grating Sensors in Nuclear Environments<br />
Francis Berghmans 1,2,* and Andrei Gusarov 2<br />
Andrea Cusano, Antonello Cutolo and Jacques Albert (Eds)<br />
All rights reserved - © 2011 <strong>Bentham</strong> <strong>Science</strong> Publishers Ltd.<br />
CHAPTER 12<br />
1 Department of Applied Physics and Photonics, Vrije Universiteit Brussel, Pleinlaan 2, 1050 Brussels, Belgium and<br />
2 SCK·CEN Belgian Nuclear Research Center, Boeretang 200, 2400 Mol, Belgium<br />
Abstract: Fiber Bragg Grating sensors are evaluated for applications in environments where the presence of<br />
highly energetic radiation is a concern, such as space and nuclear industry. To assess the feasibility of using this<br />
sensor technology in these particular application environments it is essential to evaluate the behavior of Fiber<br />
Bragg Gratings when exposed to different types of ionizing radiation. We, therefore, review the effects of<br />
ionizing radiation on several types of Fiber Bragg Gratings.<br />
INTRODUCTION<br />
Nuclear radiation effects on optical materials and photonic devices have been studied since several decades. In the<br />
early days one of the main concerns was to determine whether advanced technologies could withstand military and<br />
tactical environments such as surveillance satellite missions and nuclear weapon explosions. Today many other<br />
applications associated with presence of highly energetic radiation can benefit from the enhanced functionalities<br />
offered by photonic technologies for communication and sensing. Examples of these application fields include<br />
space, healthcare, civil nuclear industry and high energy physics experiments. Future thermonuclear fusion plasma<br />
reactors will also rely on optical communication and sensing techniques.<br />
Electronic and photonic components are well known to suffer from exposure to nuclear radiation [1]. The radiation<br />
interacts with the materials and alters their characteristics. This most often modifies the performance and affects the<br />
reliability of the device. Resulting device failures and system malfunctions may have dramatic consequences on<br />
safety and carry significant financial repercussions. One has to bear in mind that human intervention to replace<br />
components or to repair systems in radiation environments is mostly impossible due to the radioactivity levels<br />
involved or for reasons of accessibility. It is indeed almost impossible to repair onboard equipment once a satellite<br />
has been put into orbit. One can hence understand the importance of carefully investigating how radiation affects the<br />
operation and the reliability of photonic components intended for use in these harsh environments and of ensuring<br />
that devices and systems will survive the entire mission.<br />
Several years ago optical fibers carried a relatively bad reputation in terms of resistance to ionizing radiation. The<br />
fibers were indeed known to darken rapidly during exposure with substantial levels of so called radiation induced<br />
attenuation (RIA) as a result. Modern fiber fabrication technologies now allow obtaining fibers with low to moderate<br />
levels of RIA, depending on the wavelength range. This together with the undeniable and well-known advantages of<br />
fiber optic technology and the increased availability of compact, efficient and lower cost devices promoted renewed<br />
interest in the applications of optical fiber based systems in radiation environments.<br />
A fiber Bragg grating (FBG) is a typical example of a versatile photonic component that can be applied in both<br />
optical communication and sensing systems. FBGs are for example being considered as sensor elements for<br />
structural health monitoring of spacecrafts and as dosimeter in nuclear facilities. Both types of applications put<br />
entirely different demands on the FBGs: for the first application the gratings should be insensitive to radiation<br />
whereas for the second they should essentially be as sensitive as possible.<br />
The response of different types FBGs to various kinds and intensities of highly energetic radiation has therefore been<br />
the subject of many studies in the last decade. This <strong>chapter</strong> intends to review the main results obtained so far. Since<br />
radiation environments bring along considerably different environmental constraints from those conventionally<br />
*Address correspondence to this author Professor Francis Berghmans: Department of Applied Physics and Photonics, Vrije Universiteit<br />
Brussels, Pleinlaan 2, 1050 Brussels, Belgium; Tel: +32 2 629 34 53; Fax: +32 2 629 34 50; Email: fberghma@vub.ac.be
Fiber Bragg Grating Sensors in Nuclear Environments Fiber Bragg Grating Sensors 219<br />
encountered we chose to introduce the subject with an overview of different related application fields. Each of these<br />
fields is characterized by the presence of particular radiation types and intensities. These together with the basic<br />
radiation-matter interactions and the typical quantities and units involved are recalled as well. This should help the<br />
reader to understand how radiation induced ionization and displacement processes can significantly alter material<br />
and device characteristics. We then deal with radiation effects on fiber Bragg gratings by elaborating on a<br />
substantial amount of studies reported in open literature.<br />
For descriptions of the basic features of FBGs, such as spectral characteristics, fabrication methods and sensor<br />
properties we refer to other <strong>chapter</strong>s in this book.<br />
AN OVERVIEW OF RADIATION ENVIRONMENTS<br />
This overview deals with three types of radiation environments: space, civil nuclear industry and future<br />
thermonuclear fusion reactors. The choice for these environments stems from the different radiation effect studies on<br />
FBGs that will be reported later in the text. These studies were indeed conducted with the intention to assess the<br />
feasibility of applications in one of these three environments.<br />
In this section and further down the <strong>chapter</strong> we will use different units to quantify and compare the intensity and the<br />
amount of radiation that can be encountered by a FBG in the different radiation environments. For particle radiation<br />
the flux is given in number of particles per unit area and per unit time. The particle fluence is then given in number<br />
of particles per unit area. For purely ionizing radiation such as gamma rays, the total ionizing dose is expressed in<br />
“Gray” (Gy) which corresponds to 1 J of energy absorbed in 1 kg of material and is the S.I.-unit of dose.<br />
Alternatively we can use “rads” with 1 Gy ≡ 100 rad. The gamma dose-rate is given in Gy per hour or rad per hour.<br />
The meaning of these quantities will also be described in the section dealing with the basics on radiation-matter<br />
interactions.<br />
Space<br />
The main sources of energetic particles that affect space borne devices can be classified in (1) protons and electrons<br />
trapped in the Van Allen belts, (2) heavy ions trapped in the magnetosphere, (3) cosmic ray protons and heavy ions,<br />
and (4) protons and heavy ions from solar flares. The levels of all of these sources are affected by the activity of the<br />
sun. Particle energies range from keV to GeV and beyond [2].<br />
The Van Allen belts correspond to two torus shaped zones of charged particles which are trapped by our planet’s<br />
magnetic field (Fig. 1). The inner belt reaches about 1.5 earth radii far and is essentially fed by protons with energies<br />
in the 10-100 MeV range and electrons with energies from say 1 to 5 MeV. The protons can penetrate spacecrafts<br />
and damage instruments. They are also very dangerous for astronauts. The outer belt has maximum radiation flux<br />
values between 4 and 5 earth radii and contains essentially 0.1-20 MeV electrons. This belt can be hazardous to<br />
communication satellites [4].<br />
Figure 1: Simplified drawing of the Van Allen radiation belts [3].
220 Fiber Bragg Grating Sensors Berghmans and Gusarov<br />
Cosmic rays include all particles traveling through the Cosmos with very high energies. The cosmic rays can have a<br />
solar, galactic and extragalactic origin. Almost 90% of all the incoming cosmic ray particles are protons, about 9%<br />
are helium nuclei (alpha particles) and about 1% are electrons [5].<br />
Solar flares correspond to releases of magnetic energy built up in the sun’s atmosphere. With this energy release<br />
charged particles and nuclei are accelerated in the solar atmosphere. Solar flare protons, together with electrons and<br />
alphas in smaller quantities, are emitted by the sun in bursts during solar storms. The flux varies with the solar cycle.<br />
The energy spectra of solar protons are likely to be softer than those associated with trapped protons, but a<br />
spacecraft may nevertheless be exposed to considerable total fluence levels. The degree of exposure to such particles<br />
is highly dependent on orbital parameters. The Earth’s magnetic field exhibits a shielding effect in the equatorial<br />
regions, but allows the proton flux to be tunneled in towards the magnetic poles.<br />
Space agencies such as the National Aeronautics and Space Administration (NASA) and the European Space<br />
Agency (ESA) have defined guidelines for testing the radiation hardness of onboard equipment in terms of exposure<br />
levels and radiation intensities. ESA has for example specified the space environment in standard ECSS-E-10-04A<br />
[6]. This standard is intended to assist in the consistent application of space environment engineering to space<br />
products through specification of required or recommended methods, data and models to the problem of ensuring<br />
best performance, problem-avoidance or survivability of a product in the space environment.<br />
To assess the radiation dose experienced by a FBG in a spacecraft one has to consider the spacecraft orbit as well as the<br />
shielding. Typical orders of magnitude used in space radiation testing are a total ionizing dose of 10 kGy (≡1 Mrad)<br />
and a proton fluence of 10 12 cm -2 s -1 for a 10 year mission. Proton energy also has to be considered and is typically a few<br />
tens of MeV.<br />
In terms of applications FBGs have been considered for structural integrity monitoring of spacecrafts, leak detection<br />
on fuel tanks, sensor systems in nuclear propulsion systems, wavelength division multiplexing (WDM) element in<br />
intra- and inter-satellite communication systems [7-9].<br />
Nuclear Fission and Nuclear Power Plants<br />
Civil nuclear industry essentially refers to the entire nuclear fuel cycle. The two main types of radiation which needs<br />
to be taken into account in this application field are gamma-rays emitted by fuel elements and by the surrounding<br />
structures which have become radioactive following the exposure to neutrons, as well as the neutrons themselves.<br />
The latter are however only to be considered for applications inside a reactor core during reactor operation.<br />
The actual flux/dose-rate and fluence/dose levels will strongly depend on the particular application. One can expect<br />
large differences in radiation fluxes whether one considers operation in a hot-cell or in the core of an operating<br />
reactor. Table 1 lists typical orders of magnitude for various applications.<br />
Table 1: Examples of radiation conditions that can be encountered for various applications in civil nuclear environments.<br />
Φ n = neutron flux; Φ γ = gamma dose-rate; D γ = gamma total dose [1, 10-11].<br />
Application Environment<br />
In reactor core n ≤ 10 14 cm -2 s -1 , ≤ 10 7 Gy/h<br />
Containment building D ≤ 5·10 5 Gy (normal operation) / D ≤ 1.5·10 6 Gy (accident)<br />
Reactor maintenance ≈ 10 Gy/h, D ≈ 10 4 Gy<br />
Hot-cell ≈ 0-10 4 Gy/h, D ≈ 0-10 6 Gy<br />
Decontamination ≈ 10 -2 -10 Gy/h, D ≈ 10-10 3 Gy<br />
Spent fuel manipulation ≈ 10 2 -10 3 Gy/h, D ≈ 10 6 -10 7 Gy<br />
FBGs have been considered for various sensing tasks in nuclear industry. Examples include structural health<br />
monitoring of containment buildings, in reactor core temperature and mechanical stress measurements, and longterm<br />
monitoring of underground nuclear waste storage facilities [12-13].
238 Fiber Bragg Grating Sensors: Recent Advancements, Industrial Applications and Market Exploitation, 2011, 238-269<br />
Andrea Cusano, Antonello Cutolo and Jacques Albert (Eds)<br />
All rights reserved - © 2011 <strong>Bentham</strong> <strong>Science</strong> Publishers Ltd.<br />
CHAPTER 13<br />
Fiber Bragg Grating Evanescent Wave Sensors for Chemical and Biological<br />
Applications<br />
Andrea Cusano 1,* , Domenico Paladino 1 , Antonello Cutolo 1 , Agostino Iadicicco 2 and<br />
Stefania Campopiano 2<br />
1 Optoelectronic Division – Engineering Department, University of Sannio, Corso Garibaldi 107, 82100 Benevento,<br />
Italy and 2 Department of Technology, University of Naples “Parthenope”, Centro Direzionale di Napoli Isola C4,<br />
80143 Napoli, Italy<br />
Abstract: While Fiber Bragg Grating (FBG) sensors continue to act as valuable sensing platforms specially when<br />
physical parameters have to be measured in single or multipoint configuration, great efforts have been focused in<br />
the last decade to convey the great advantages of FBG technology towards the development of new devices and<br />
components employable in modern chemical and biological applications. Since the first attempt carried out in<br />
1996 by Meltz et al., many sensing schemes have been proposed based on the evanescent wave effect arising<br />
from the surrounding refractive index sensitization of the final device by tailoring either the grating structure or<br />
the host fiber. In this <strong>chapter</strong>, we review the main advances in the area of FBG evanescent wave sensors, with<br />
particular emphasis on principles of operation, technological developments, overall performances and discussing<br />
in details perspectives and challenges that lie ahead.<br />
INTRODUCTION<br />
Among the large number of fiber optic sensors configurations, Fiber Bragg Grating (FBG) based sensors, more than<br />
any other particular sensor type, have become widely known and popular within and out the photonics community<br />
and seen a rise in their utilization and commercial growth. Given the capability of FBGs to measure a multitude of<br />
parameters such as strain, temperature, pressure, and many others coupled with their flexibility of design to be used<br />
as single or multipoint sensing arrays and their relative low cost, make them ideal devices to be adopted for a<br />
multitude of sensing applications and implemented in different industrial fields [1-5].<br />
With numerous life sciences applications in mind, the photonic science and technology have recently been evolved<br />
into an important interdisciplinary field – bio-photonics [6]. One highly attractive area in this large field, the optical<br />
biosensor, is experiencing a new surge in activity, seeking to exploit novel optical structures and bio-coating<br />
materials and techniques, driven by the demand for high performance analytical tools capable of detecting and<br />
discriminating among large classes of bio-molecules. Through rapid advances in this area, the highly biosensitive/selective<br />
optical sensors appear set to become viable and preferred alternatives to traditional “solution<br />
based” assay biosensors for applications in genomics, proteomics and drug discovery research and development, as<br />
well as in food industry, homeland security and environmental monitoring applications.<br />
Over the last decade, FBGs have provided the basis for families of optical sensors for applications in aerospace,<br />
transportation, energy and civil engineering, just to name a few. Nevertheless, there has also been an increasing<br />
flurry of activity aimed at implementing optical biochemical sensors by exploring the grating’s response to the<br />
Surrounding-medium Refractive Index (SRI) or of thin specific and functionalized overlays. In fact, besides the<br />
direct influence of temperature and strain onto the optical length of the fiber grating, there is the possibility to<br />
modify the effective refractive index of the guided mode via the evanescent wave interaction [7]. This feature<br />
combined with the intrinsic multi-measurand capability in single and multipoint sensor configuration open the way<br />
to develop high performances FBGs evanescent wave sensors as valuable technological platforms for chemical and<br />
biological applications.<br />
Among fiber gratings written perpendicularly to the fiber axis, only Long-Period Fiber Gratings (LPFGs) are readily,<br />
intrinsically sensitive to the SRI via the coupled light from core to cladding penetrating the surrounding medium. As a<br />
core-to-core mode coupling device, the light in a FBG is well screened by the cladding, effectively precluding strong<br />
*Address correspondence to this author Professor Andrea Cusano: Optoelectronic Division – Engineering Department, University of Sannio,<br />
Corso Garibaldi 107, 82100 Benevento, Italy; Tel: +39 0824305835; Fax: +39 0824305846; Email: a.cusano@unisannio.it
Fiber Bragg Grating Evanescent Wave Sensors for Chemical and Biological Applications Fiber Bragg Grating Sensors 239<br />
interaction with the surrounding medium. However, FBGs can be SRI-sensitized by tailoring either the grating<br />
structure or the host fiber, creating the basis for the development of possible technological platforms for chemical and<br />
biological sensing applications. Furthermore, despite high SRI sensitivity in comparison with FBGs, LPFGs exhibit<br />
several disadvantages as deployable devices. LPFGs possess much higher temperature and bending cross-sensitivities<br />
and, thus, can be severely influenced by their environmental conditions. Another disadvantage of the LPFG is that its<br />
spectral response can be measured only in transmission, and often with poor resolution due to the broad (typically<br />
tens of nanometers) transmission loss-type resonances. To reduce the bandwidth of the LPFG significantly, the device<br />
length has to be increased substantially, compromising its use as a localized or point sensor device.<br />
Here, we review the main advances in the area of fiber grating evanescent wave sensors focusing the attention on<br />
short period FBGs. To this aim, we can envision most of the works proposed in literature as divided in four main<br />
classes employing the evanescent wave interaction as transducing principle:<br />
- FBGs written in unconventional optical fibers (D-shaped optical fibers and Micro-structured Optical<br />
Fibers (MOFs)) ;<br />
- Uniform Thinned FBGs (ThFBGs);<br />
- Tilted FBGs (TFBGs);<br />
- Micro-structuring of conventional FBGs by post processing techniques.<br />
In the following these classes are separately illustrated with particular emphasis on their principle of operation,<br />
technological development, overall performances and discussing in details perspectives and challenges that lie ahead.<br />
FBGS WITHIN UNCONVENTIONAL OPTICAL FIBERS<br />
From the historical point of view, the first attempt to realize a FBG based evanescent wave sensor relies on the use<br />
of D-shaped optical fibers [8]. Following this first experimental evidence, many efforts have been mainly devoted to<br />
optimize the structure through the use of side polished optical fibers or surface relief gratings within D-fibers. More<br />
recently, a new approach has been envisaged via grating writing within MOFs, providing the basis for modern optofluidics<br />
components and devices.<br />
FBGs within D-Shaped Optical Fibers<br />
The first attempt to use FBGs as chemical transducers via evanescent wave interaction was demonstrated in 1996 by<br />
Meltz et al. [8]. Unlike LPFGs, FBGs are intrinsically insensitive to the SRI, since the light coupling takes place<br />
only between well-bound core modes that are well screened from the influence of the SRI by the cladding. To<br />
address this issue, the basic idea proposed by Meltz et al. was to use fiber gratings written in D-fibers sensitized to<br />
the SRI through post processing cladding stripping.<br />
Figure 1: Schematic of a D-shaped fiber as supplied by KVH Industries Inc. (Source: Brigham Young University).<br />
As reference, (Fig. 1) illustrates the cross section of a D-shaped fiber supplied by the KVH Industries, Inc. The two<br />
primary features of the D-fiber are the elliptical core, which maintains the polarization, and the proximity of the<br />
fiber core to the flat surface above the core. This proximity allows access to the light in the core by a weak cladding
240 Fiber Bragg Grating Sensors Cusano et al.<br />
stripping above the core area. Because of the D-shape of the fiber, the mechanical integrity and approximately the<br />
same dimensions of the fiber are preserved.<br />
As few microns of cladding are removed from the flat side of the host fiber, the core mode is influenced by the<br />
external medium through evanescent wave interaction. As main consequence, the shift of the Bragg wavelengths<br />
related to the two polarization states occurs. The Bragg lines of both the fast and slow eigenmodes, indeed, are blueshifted<br />
when the silica cladding layer is removed and replaced with water or methanol films (lower than the core<br />
refractive index). Changes in the fiber birefringence were observed because the perpendicular and parallel modes<br />
decay into the cladding at different rates. By using a tunable laser with a narrowband Bragg grating filter or threegrating<br />
Fabry-Pérot interferometer, refractive index variations of 5 by 10 -6 could be detected. Temperature<br />
compensation methods were also discussed including the use of an isolated reference grating and the simultaneous<br />
combination of birefringence and Bragg line wavelength shift measurements.<br />
Successively, planar side polishing [9] was demonstrated as effective technique to develop reliable chemical<br />
transducers based on FBGs written in D-shaped fibers with normal birefringence and depressed cladding [10].<br />
In this experiment, standard single mode optical fiber was embedded in a glass block and successively polished<br />
down to residual cladding thickness less than 2 m (see Fig. 2). Finally, Bragg gratings were written in the central<br />
part of the polished fiber allowing multipoint sensor array being addressed by wavelength multiplexing of sensors<br />
heads in series along a single optical fiber. In follow up works, the sensing performances of the proposed<br />
configuration have been in details numerically and experimentally investigated revealing that [11]:<br />
- due to the asymmetric removal of the cladding, the dependence of the effective index of the<br />
fundamental mode on the surroundings is affected by light polarization, in particular, TM polarization<br />
exhibits a higher efficiency when compared to the TE polarization state;<br />
- the effective refractive index of the fundamental mode increases non-linearly approaching its<br />
maximum for SRI values close to the core refractive index (cut-off of the fundamental mode);<br />
- maximum sensitivity is obtained in case of cladding layer completely removed on the flat side of the<br />
structure leading to the maximum evanescent wave interaction;<br />
- in the case of coated structures with low refractive index overlays, sensing performances decrease as<br />
the coating is thinned compared with the penetration depth of the evanescent wave;<br />
- due to the dependence of the penetration depth on the optical wavelengths, sensing performances can<br />
be improved if longer wavelengths are used.<br />
Analyte n A<br />
l = 2n eff [n ]<br />
A<br />
B<br />
Figure 2: Schematic of a side polished FBG chemical sensor. (Source: IPHT Jena)<br />
Bragg Wavelength [nm]<br />
839.0<br />
838.8<br />
838.6<br />
838.4<br />
838.2<br />
838.0<br />
Figure 3: SRI sensor characteristic at 838 nm. (Source: IPHT Jena)<br />
Fiber Bragg grating in core<br />
of side-polished fiber<br />
837.8<br />
1.30 1.35 1.40<br />
1.45<br />
DAO<br />
n Ethanol<br />
A H O<br />
OZ91<br />
2<br />
LMO<br />
26%NaCI/H O OZ98<br />
2<br />
SPO<br />
L<br />
Embedded side-polished<br />
single mode fiber<br />
OZ91, OZ98 -<br />
Octane number of Petrol<br />
products<br />
SPO - Spindle oil<br />
LMO - Light machine oil<br />
DAO - De-asphalted oil of<br />
Furfural extraction product
270 Fiber Bragg Grating Sensors: Recent Advancements, Industrial Applications and Market Exploitation, 2011, 270-291<br />
Fiber Bragg Grating Sensors in Microstructured Optical Fibers<br />
Tomasz Nasilowski*<br />
Andrea Cusano, Antonello Cutolo and Jacques Albert (Eds)<br />
All rights reserved - © 2011 <strong>Bentham</strong> <strong>Science</strong> Publishers Ltd.<br />
CHAPTER 14<br />
Department of Applied Physics and Photonics, Faculty of Engineering, Vrije Universiteit Brussel, Pleinlaan 2,<br />
building F, B-1050 Brussels, Belgium<br />
Abstract: The unusual properties of Fiber Bragg Gratings in microstructured fibers deliver significantly<br />
improved performance in some respect to traditional fibers. They are also superior in one or few characteristics<br />
and eventually unveil the novel properties of fiber Bragg grating sensors, leading to the superiority of<br />
implementations and original applications, which are briefly discussed in this <strong>chapter</strong>. Moreover, a very<br />
promising functionality of doped core microstructured fibers in perspective of material and geometrical guiding<br />
mechanism competition is clarified and encouraged. Furthermore, such fibers are greatly advantageous for Fiber<br />
Bragg Grating inscription and instead of scarifying important properties they pave the way to novel applications<br />
fulfilling the industrial requirements. Additionally, very interesting and hopeful research and development of pure<br />
silica microstructured Fiber Bragg Gratings are presented as well.<br />
INTRODUCTION<br />
One of a very important and popular type of fiber sensors rely on a fiber Bragg grating (FBG) [1-2]. FBG sensors<br />
are commonly used in civil engineering (bridges, dams, railways, mines etc.). They consist 70% to 80% of the<br />
optical sensors in structural health monitoring.<br />
Some of the reasons for this common use of FBG sensors [3-5] is based in the ability to measure physical changes<br />
such as multi-axis and transverse strain, vibration, displacement, and also moisture, ice, and in general, refractive<br />
index changes of the sensor neighborhood. Additionally, because the temperature and stress directly affect the<br />
reflectivity spectrum of FBGs, they can provide unique measurements availability, such as detecting changes in<br />
stress in buildings, bridges, airplane wings or other bodies; depth measurements in rivers, reservoirs for flood<br />
control; temperature and pressure measurements in deep oil wells, dislocations of railways, non circularity of train<br />
wheels, cracks in mines etc. They also offer exceptional resolution, large dynamics, low weight, absolute<br />
measurements and modest cost per channel. Furthermore, due to their nature FBG sensors can be either time- or<br />
wavelength-multiplexed, which allows for quasi-distributed sensing – an essential benefit for structural health<br />
monitoring [6].<br />
The majority of FBG sensors are inscribed in conventional glass optical fibers. A new class of microstructured fibers<br />
(MOFs) or photonic crystal fibers (PCFs), also called holey fibers (HF) [7-8], became in recent years a subject of<br />
extensive research. Generation of supercontinuum, improvement in efficiency of fiber lasers, very low bending<br />
losses and compensation of chromatic dispersion are some of the numerous applications of MOFs already<br />
demonstrated in literature [7-8]. Endlessly single mode light propagation for very broad wavelength range (from UV<br />
up to IR) was the first paradigm shift unveiled by new generation fibers. Moreover, polarization properties of MOFs<br />
are also very interesting. For example, it has been shown that modal birefringence in MOFs may exceed 10 -3 , which<br />
is one order of magnitude higher than birefringence in classical highly birefringent (HB) fibers [7-9]. Furthermore<br />
single-polarization operation can be achieved as well in photonic crystal fibers [10-12].<br />
MOFs for sensing purposes cover various types of sensors including interferometric and polarimetric sensors of<br />
different physical parameters (temperature, hydrostatic pressure, elongation, force, bending, etc.) as well as<br />
evanescent field sensors for monitoring specific chemical compounds in gas and liquids and their refractive index<br />
changes. Many of these interesting sensing applications are presented in Refs. [13-14].<br />
The combination of a microstructured fiber with FBG is an architecture which joins two interesting ideas. On one hand<br />
*Address correspondence to this author Professor Tomasz Nasilowski: Department of Applied Physics and Photonics, Faculty of<br />
Engineering, Vrije Universiteit Brussel, Pleinlaan 2, building F, B-1050 Brussels, Belgium; Tel: +32 2 629 35 67; Fax: +32 2 629 34 50; Email:<br />
tnasilowski@tona.vub.ac.be
Fiber Bragg Grating Sensors in Microstructured Optical Fibers Fiber Bragg Grating Sensors 271<br />
we have the Bragg grating, a device which has been thoroughly studied [15-17] and successfully employed in<br />
numerous applications. On the other hand the photonic crystal fiber with its many, recently discovered possibilities.<br />
To rather big extend we can consider such components as three dimensional photonic crystal structures.<br />
Fiber Bragg gratings in conventional fibers have played a major role in many applications, and therefore in recent<br />
years there have been several attempts to incorporate a Bragg grating in MOF through different techniques.<br />
One of the most attractive fiber sensing applications steams from the use of highly birefringent MOF with a<br />
dedicated design that allows inscribing FBG in the fiber core. Such 3D microstructure can serve as pressure or<br />
transverse force transducers with a large sensitivity, while exhibiting a very low sensitivity to temperature drifts<br />
and/or longitudinal strain if adequately designed. Therefore, Bragg gratings in PCF may offer a viable alternative to<br />
traditional optical fiber sensors that require temperature compensation mechanisms and that are not intrinsically<br />
capable of distinguishing stress and temperature [18-20].<br />
The unusual properties of MOFs deliver significantly improved performance in respect to traditional fibers. They are<br />
superior in several characteristics and also unveil the novel properties of fibers. All this leads to the superiority of<br />
implementations and original applications of FBGs in MOFs.<br />
The intention of this <strong>chapter</strong> is to show that the most realistic applications of microstructured fiber Bragg<br />
gratings for the next years steam from the doped core MOFs, which are enough UV photosensitive to use<br />
standard FBG inscription setups [21]. Moreover, such fibers give additional degrees of freedom (shape and level<br />
of doped core) over pure silica MOFs in designing sensor properties, especially in case of birefringent MOFs. The<br />
most important advantage seams to be the selectivity of the measured parameters, like temperature insensitivity.<br />
In longer time scale, the point-by-point inscription technology [22] should prove a great potential in writing<br />
FBGs in low UV sensitive MOFs. However, presently this technology is still far from being mature for<br />
industrial implementation.<br />
FBG in MOFs are becoming the new basic component which leads to the novel and large scale development<br />
of fiber sensor industrial applications. As it is shown in this <strong>chapter</strong>, already now, in many cases such sensors<br />
can be fabricated on industrial scale with necessary reliability and repeatability for sensing applications.<br />
The following paragraphs give a short description of guiding mechanisms in microstructured fibers followed by the<br />
explanation of propagation mechanism competition in case of doped MOFs. Next, there is a brief presentation of<br />
different technologies for FBGs inscription in MOFs. After that the main results and achievements of FBG in solid<br />
bandgap fibers (SBF) is mentioned. Other sections demonstrate the breakthroughs of FBGs in MOFs for sensing<br />
applications of gases and liquids refractive index measurements, as well as monitoring the mechanical stresses with<br />
use of FBGs in low and high birefringent fibers, which are often temperature insensitive.<br />
There are still many unanswered questions and non established limitations of FBGs in MOFs that have to be studied<br />
within next years.<br />
GUIDING MECHANISM IN MICROSTRUCTURED FIBER<br />
The spite of high maturity of classical fibers, they have limits in designing their properties. This might sometimes<br />
limit their applications or make them less competitive e.g. with electronic devices. This situation was often taking<br />
place in case of sensing purposes and for that reason fiber sensors were mostly employed in niche applications.<br />
Microstructured fibers consist of a periodic lattice (or another type of structure) of air holes that run along the length<br />
of the fiber and that allow confining and guiding light along the fiber (Fig. 1). A defect in the centre of the lattice<br />
serves as fiber core. Adapting the particular features of the microstructure, such as the air filling fraction and the<br />
lattice period, allows obtaining optical fibers with very particular properties in terms of dispersion, mode-field<br />
confinement, endlessly single mode operation, unusual polarization properties etc. [7-8]. The optical and mechanical<br />
properties of MOFs can also be controlled by changing the size, shape and location of the air holes.
272 Fiber Bragg Grating Sensors Tomasz Nasilowski<br />
The unusual properties of MOF deliver significantly improved performance in some respect to traditional fibers.<br />
They are also superior in one or few characteristics and eventually unveil the novel properties of fibers. All this<br />
leads to the superiority of implementations and original applications of MOFs.<br />
Classical optical fibers are today finding wide use in areas covering telecommunications, sensor technology,<br />
spectroscopy, medicine etc. [23]. Their operation usually relies on light being guided by the difference in material<br />
properties sometime called index guiding, and often mistaken with total internal reflection phenomena due to its<br />
intuitive property taken from geometrical optics. Precisely, light in classical fibers is guided in the core region filled<br />
by the material with higher refractive index and surrounded by the cladding region filled by the material with lower<br />
refractive index. Some of the modes allowed by Maxwell equations or wave equation [24] in the core are not<br />
allowed in the cladding region. In other words, some stationary solutions of Maxwell equations in the higher<br />
refractive index material do not exist in the lower refractive index material. For this reason, if such modes are<br />
excited in the core region they have to propagate along the core down the length of the fiber since their existence is<br />
forbidden in the cladding. Because the propagation in classical fibers is steamed by the difference in the core and the<br />
cladding materials such guiding mechanism can be called material guiding.<br />
In case of microstructured fibers the difference between core and cladding is not in different material properties, but<br />
in different geometries of the same material. Due to this difference in geometry some of the modes are allowed in<br />
the core and are not allowed in the microstructured cladding. If these modes are excited in the core they are trapped<br />
over there and they can propagate only along the core. This is intuitively illustrated on the (Fig. 1a). Whenever the<br />
modes excited in the core are similar to the cladding modes they are not propagating because of coupling to the<br />
cladding modes, as shown on (Fig. 1b). Since the propagation is founded by the difference in geometries such<br />
guiding mechanism can be called geometrical guiding.<br />
The geometrical guiding explains the propagation in both types of microstructured fibers, namely index guiding and<br />
photonic bandgap (PBG) PCFs.<br />
Typical index guiding PCF [25-26] consist of solid glass core and microstructured hollow cladding very often<br />
surrounded again by larger solid glass region (Fig. 1). In this case the effective refractive index of the mode guided<br />
in the solid core is higher from the index of the microstructured cladding modes. For that reason this type of fibers<br />
are compared with classical, material guided, fibers. However, the great distinction of very high dispersion<br />
properties, due to the scaled with wavelength geometry and, additionally, the selectivity of cladding modes in<br />
contrast with almost continuum of modes in large and solid cladding of classical fiber are remarkable differences,<br />
which can not be neglected and are basis for most of the extraordinary properties of PCFs.<br />
n 1 eff<br />
n cl eff<br />
neff<br />
(a) (b)<br />
Figure 1: Schematic drawing of a typical microstructured fiber presents guided (a) and not guided (b) modes. If mode excited in<br />
the core has effective refractive index (n 1 eff) different from effective index of cladding modes (n cl eff) it is trapped by the<br />
surrounding cladding structure and has to propagate along the core. In case the excited mode in the core has similar effective<br />
refractive index (neff) to the one of the cladding modes it becomes the leaky mode and is coupled out from the core.<br />
n cl eff
292 Fiber Bragg Grating Sensors: Recent Advancements, Industrial Applications and Market Exploitation, 2011, 292-312<br />
Polymer Fiber Bragg Gratings<br />
David John Webb 1,* and Kyriacos Kalli 2<br />
Andrea Cusano, Antonello Cutolo and Jacques Albert (Eds)<br />
All rights reserved - © 2011 <strong>Bentham</strong> <strong>Science</strong> Publishers Ltd.<br />
CHAPTER 15<br />
1 Photonics Research Group, Aston University, United Kingdom and 2 Nanophotonics Research Laboratory, Cyprus<br />
University of Technology, Cyprus<br />
Abstract: This <strong>chapter</strong> deals with gratings recorded in polymeric optical fibers (POFs); predominantly those<br />
based on poly (methyl methacrylate) (PMMA). We summarize the different mechanical and optical properties of<br />
POFs which are relevant to the application of POF Bragg gratings and discuss the existing literature on the<br />
subject of the UV photosensitivity of PMMA. The current state of the art in POF grating inscription is presented<br />
and we survey some of the emerging applications for these devices.<br />
INTRODUCTION<br />
Work on the UV photosensitivity of poly (methyl methacrylate) (PMMA) dates back to the early 1970s, well before<br />
the discovery of photosensitivity in silica fibers. However, single mode polymer optical fiber (POF) only became<br />
available in the 1990s and it was not until the end of that decade that the first POF Bragg grating (POFBG) was<br />
reported. Compared to the current state of the art with silica FBGs, the field of POFBGs is still very immature.<br />
Considerable research challenges exist in photosensitive fiber manufacture, developing new fiber materials,<br />
understanding the mechanisms behind the photosensitivity and improving the reproducibility of grating inscription.<br />
Nevertheless, applications research is now starting to emerge, seeking to exploit the rather different properties of<br />
POF compared to silica fiber.<br />
This <strong>chapter</strong> summarizes the current state of the art in the field. We have attempted to provide useful information to<br />
help those starting to work with POF, with regard to handling the fiber and the practicalities of inscribing gratings,<br />
and have also highlighted some of the areas where further research is particularly needed, for example in<br />
understanding and improving photosensitivity. Finally, we summarize the properties of gratings produced to date<br />
and survey some of the applications that are starting to arise.<br />
TECHNOLOGICAL DRIVERS<br />
In this section we discuss the motivation for work on POF. Within the telecommunications industry, POF has a<br />
reputation for being low cost so this might seem like the first area we should explore but unfortunately, the features<br />
of POF that provide this attraction to network installers do not carry over to grating based sensing. In short distance<br />
moderate bandwidth applications, the typical POF is 1mm in diameter, most of which is taken up by the core [1].<br />
Even at this diameter, POF is highly flexible and tolerant of sharp bends, thereby being easy to install. Furthermore,<br />
the fiber can be illuminated using low cost, broad emission area sources, such as light emitting diodes. Finally, fiber<br />
connection is straightforward as the positional tolerance required is very relaxed and the fiber can be easily<br />
“cleaved” using a sharp knife. The low cost of POF in network systems is therefore not so much related to the actual<br />
fiber cost but to the ease of handling and installation coupled with the low cost of ancillary components.<br />
Where grating based sensors are concerned, the fiber will usually need to be single mode, which imposes very tight<br />
positional tolerances on connections between fibers. In addition, single transverse mode sources are usually<br />
required. These factors are all likely to preclude any cost advantage over silica fiber and consequently we have to<br />
look elsewhere for the motivation to develop this technology.<br />
Fortunately, the physical and chemical properties of polymeric materials are rather different to silica and it is in this<br />
area that there appear to be advantages. We will explore the main differences below; to do so we will often make use<br />
of the properties of PMMA as a representative polymer. It should be borne in mind though that whilst PMMA may<br />
be the most widely used polymer for optical fibers, there are many other polymers available that may well be better<br />
for certain applications.<br />
*Address correspondence to this author Dr. David J. Webb: Photonics Research Group, Aston University, Aston Triangle, Birmingham, B4<br />
7ET, United Kingdom; Tel: +44 121 204 3541; Email: d.j.webb@aston.ac.uk; kkalli@cytanet.com.cy
Polymer Fiber Bragg Gratings Fiber Bragg Grating Sensors 293<br />
Mechanical Properties<br />
Let’s start with Young’s modulus: for silica fiber it is 73 GPa [2], while a typical value for bulk PMMA is 3.3 GPa<br />
[3]. The Young’s modulus for various PMMA based fibers has been measured yielding values of: 1.6-2.1 GPa [4],<br />
2.75 GPa [5], 2.8 GPa [6] and 2.8-3.4 GPa [7]. The variations in Young’s modulus for POF observed here are<br />
unsurprising given the various co-polymers used in the fiber cores, the different processing conditions likely to have<br />
been used, the various molecular weight distributions that could exist and the addition of plasticizers that may have<br />
taken place. Note that when we talk about Young’s modulus for POF we are implicitly discussing stress and strain<br />
along the fiber axis. Following the drawing process, there is thought to be some alignment of the polymer molecules<br />
along the axis, which will probably lead to anisotropic mechanical properties.<br />
In many FBG sensing applications, the difference in the values of Young’s modulus for silica and POF will be<br />
unimportant; where it does matter is when the FBG sensor is being used to monitor a material that itself has a low<br />
Young’s modulus. The potential problem here is that a stiff silica fiber can act to locally stiffen the material,<br />
reducing the strain experienced in the region of the sensor in response to a given stress, whereas POF can have much<br />
less of an effect [8].<br />
We next consider other tensile properties of the materials. Silica fiber has a typical failure strain of 5-10% [9] and<br />
below this value exhibits good elastic behavior. Of course, extreme care must be taken during grating fabrication not<br />
to introduce any scratches onto the fiber surface, which have been shown to considerably reduce the strength [10].<br />
For PMMA and polymers in general, the situation is more complex. The good news is that extremely high failure<br />
strains are achievable. Despite the failure strain of bulk PMMA being as low as 4-5.5% (greater for higher molecular<br />
weights) [11], for fibers a value has been reported in excess of 100% [12], though it should be noted that to achieve<br />
these values it is important that the fiber is drawn under low tension. High tension drawing appears to lead to<br />
significant molecular alignment along the fiber axis which considerably reduces the failure strain, as shown in (Fig.<br />
1). Fiber drawn under high tension can be annealed to obtain the properties associated with low-tension drawing.<br />
Annealing at 95°C for a few days has the effect of: reducing Young’s modulus, yield point and tensile strength and<br />
increasing ductility (failure strain) [4]. The yield strain (limit of quasi-elastic behavior) is reported to be around 6%<br />
for PMMA [4-5].<br />
True stress (MPa)<br />
300<br />
250<br />
200<br />
150<br />
100<br />
50<br />
0<br />
00:00:00:00 00:00:00:00 00:00:00 00:00:00<br />
Increasing drawing<br />
temperature<br />
0 10 20 30 40 50 60 70 80<br />
True strain (%)<br />
Figure 1: Tensile properties of PMMA fiber drawn under different tensions. The photographs show the morphology of the fiber<br />
break [12].<br />
The bad news is that as a visco-elastic material, the tensile properties of polymers are complicated, displaying both<br />
hysteresis and a dependence on the timescales involved. For example, in one study, the yield strain of POF was<br />
00:00:00
294 Fiber Bragg Grating Sensors Webb and Kalli<br />
studied as a function of the rate of application of strain [4]. As strain rate was varied from 0.1 to 0.5 per minute, the<br />
yield strength increased from 76 to 85 MPa, with a concomitant increase in the tensile strength. The authors<br />
conclude that at low strain rates PMMA is able to conform to the applied load better than at high rates, when the<br />
internal viscosity leads to a brittle failure.<br />
PMMA based POF can display hysteresis when the applied strain is cycled. (Fig. 2) shows the effect of repeatedly<br />
stretching a POF by the same strain increment, starting each cycle from the point where the stress has dropped to<br />
zero in the previous cycle [5]. In this work, the strain was applied at a rate of 0.5%/minute. Given time in the<br />
unloaded state, the POF would relax back towards its initial value. One study reports that after strains as high as 3%<br />
the POF will return to its original length when the strain applied over a short period of time, of the order of a minute<br />
[13]. On the other hand, when the strained fiber was held for 10 hours prior to release, the relaxation was much<br />
slower. After a strain of 1%, the fiber relaxed back to essentially its original length after 10 hours, while from a<br />
strain of 4%, the length had only relaxed half way to its original value in that time [14]. This study also noted a<br />
relaxation of the fiber when under constant strain, revealed as a decrease in the stress needed to cause that strain.<br />
16<br />
12<br />
8<br />
4<br />
0<br />
Stress (MPa)<br />
Strain (%)<br />
0 0.2 0.4 0.6 0.8 1.0<br />
Figure 2: First, second, third and tenth cycles from repetitive tensile testing of PMMA fiber (after Ref. [5]).<br />
Finally, even in its quasi-elastic region, the stress-strain curve for PMMA fiber is not linear [7]. In fact the nonlinearity<br />
is approximately an order of magnitude larger than for silica fiber, and as a consequence can be important<br />
for strains above 1%, as opposed to 3% for silica fiber. This same study looked at the behavior of PMMA fiber<br />
under strain rates from 0.01 to 3/minute and noted – as mentioned earlier – that higher rates increased yield strain<br />
and yield stress but not the value of Young’s modulus.<br />
Chemical Properties<br />
The drawing of PMMA based fiber takes place at around 200°C, as opposed around 2000°C for silica. Many organic<br />
substances can survive the former temperature but certainly not the latter. Furthermore, as an organic material itself,<br />
PMMA is readily susceptible to processing using a range of organic chemistry techniques. The ability to incorporate<br />
specific organic molecules, either directly as part of the host polymer or added as copolymers or dopants, offers<br />
many possibilities for modifying the fiber, for example to enhance non-linear properties [15], to enable optical<br />
amplification [16], to provide sensitivity to specific chemical – or biochemical – species [17] or to increase<br />
photosensitivity [18].<br />
Biocompatibility<br />
Both silica and PMMA are considered to be biocompatible in many situations. An advantage for polymer fibers<br />
arises due to their mechanical properties discussed above: POF can have a much greater failure strain than silica,<br />
leading to less chance of fiber breakage in the body. Perhaps more importantly, when silica fiber breaks there is a<br />
high chance that a sharp edge or point will result that could cause damage to surrounding tissue; the much softer<br />
polymer will not pose this kind of threat.
Fiber Bragg Grating Sensors: Recent Advancements, Industrial Applications and Market Exploitation, 2011, 313-320 313<br />
Andrea Cusano, Antonello Cutolo and Jacques Albert (Eds)<br />
All rights reserved - © 2011 <strong>Bentham</strong> <strong>Science</strong> Publishers Ltd.<br />
CHAPTER 16<br />
Fiber Bragg Grating Sensors: Market Overview and New Perspectives<br />
Jeff Wayne Miller 1 and Alexis Méndez 2,*<br />
1 Micron Optics Inc., 1852 Century Place, Atlanta, GA 30345, USA and 2 MCH Engineering LLC, 1728 Clinton<br />
Avenue, Alameda, CA 94501, USA<br />
Abstract: Over the last few years, optical fiber sensors have seen increased acceptance and widespread use.<br />
Among the multitude of sensor types, Fiber Bragg Grating (FBG) based sensors—more than any other particular<br />
sensor type—have become widely known and popular. FBGs have an intrinsic capability to measure a variety of<br />
parameters along a single fiber, such as: strain, temperature, pressure, chemical and biological agents, and many<br />
others. These multi-point sensing arrays of many relative low cost FBGs, provide great flexibility of design and<br />
make them ideal devices to be adopted for a multitude of different sensing applications and implemented in<br />
different fields and industries. However, some technical hurdles and market barriers need to be overcome in order<br />
for this technology—and fiber sensors in general—to gain more commercial momentum and achieve faster and<br />
more pronounced market growth. Other relevant factors are the need for industry standards on FBGs and FBGbased<br />
sensors, adequate and reliable packaging designs, as well as training and education of prospective<br />
customers and end-users.<br />
INTRODUCTION<br />
The fiber optics field has undergone a tremendous growth and advancement over the past 40 years. Initially<br />
conceived as a medium to carry light and images for medical endoscopic applications, optical fibers were later<br />
proposed in the mid 1960’s as an information-carrying medium for telecommunication applications. The outstanding<br />
success of this concept is embodied in the millions of miles of telecommunications fiber that have spanned the earth<br />
and the seas, utterly transforming the means by which we communicate. This has all been documented with awe<br />
over the past several decades. Among the reasons why optical fibers are such an attractive communication medium<br />
are their low loss, high bandwidth, EMI immunity, small size, lightweight, safety, relatively low cost, low<br />
maintenance, etc.<br />
As optical fibers cemented their position in the telecommunications industry—and its associated technology and<br />
commercial markets matured—parallel efforts were carried out by a number of different groups around the world to<br />
exploit some of the key fiber features and utilize them in sensing applications.<br />
Initially, fiber sensors were lab curiosities and simple proof-of-concept demonstrations. However, nowadays, optical<br />
fibers are making an impact and serious commercial inroads in other fields besides communications such as in<br />
industrial sensing, bio-medical laser delivery systems, military gyro sensors, as well as automotive lighting &<br />
control—to name just a few—and spanned applications as diverse as oil well downhole pressure sensors to intraaortic<br />
catheters. This transition has taken the better part of 20 years and reached the point where fiber sensors enjoy<br />
increased acceptance as well as a widespread use for structural sensing and monitoring applications in civil<br />
engineering, aerospace, marine, oil & gas, composites, smart structures, bio-medical devices, electric power industry<br />
and many others [1-2]. Optical fiber sensor operation and instrumentation have become well understood and<br />
developed, and a variety of commercial discrete sensors based on Fabry-Perot (FP) cavities and Fiber Bragg Gratings<br />
(FBGs), as well as distributed sensors based on Raman and Brillouin scattering methods, are readily available along<br />
with pertinent interrogation instruments. Among all of these, FBG based sensors—more than any other particular<br />
sensor type—have become widely known, and seen a rise in their utilization and commercial growth.<br />
FIBER BRAGG GRATINGS AS SENSORS<br />
Since their fortuitous discovery by Ken Hill back in 1978 [3] and subsequent development by researchers at the<br />
Communications Research Centre, United technologies, 3M and several others [4], intra-core fiber gratings have been<br />
*Address correspondence to this author Dr. Alexis Méndez: MCH Engineering LLC, 1728 Clinton Avenue, Alameda, CA 94501, USA; Tel:<br />
+1 (510) 521-1069; Fax: +1 (510) 521-5079; Email: alexis.mendez@mchengineering.com
314 Fiber Bragg Grating Sensors Miller and Méndez<br />
used extensively in the telecommunication industry for dense wavelength division multiplexing, dispersion<br />
compensation, laser stabilization, and erbium amplifier gain flattening, mostly at the 1550 nm within the C-band<br />
wavelength range. However, one of the primary FBG benefits is the ability to be multiplexed, allowing multiple<br />
sensors and multiple parameters to be measured along a single fiber. These, multi-point, FBG sensing arrays provide<br />
great measurement flexibility and make them ideal devices to be adopted for a multitude of different sensing<br />
applications and implemented in different fields and industries.<br />
FBG-based sensors have been developed for a wide variety of sensing applications (see Fig. 1) including monitoring<br />
of civil structures (highways, bridges, buildings, dams, etc.), smart manufacturing and non-destructive testing<br />
(composites, laminates, etc.), remote sensing (oil wells, power cables, pipelines, space stations, etc.), smart<br />
structures (airplane wings, ship hulls, buildings, sports equipment, etc.), as well as traditional strain, pressure and<br />
temperature sensing. To date, there are a diverse number of commercial FBG-based sensors designed and packaged<br />
to measure a variety of different mechanical, electrical and chemical parameters, as shown in (Fig. 2).<br />
Figure 1: FBG sensors are used in a variety of industrial applications.<br />
Figure 2: Diverse types of commercial FBG-based sensors.<br />
Accelerometer Displacement meter<br />
Strain meter Pressure meter<br />
Thermometer<br />
The main advantage of fiber gratings for mechanical sensing is that these devices perform a direct transformation of<br />
the sensed parameter into optical wavelength—independent of light levels, connector or fiber losses, or other FBGs<br />
operating at distinct wavelengths. For instance, when compared to one of the most common and popular basic<br />
electronic sensors—the foil strain gage—the relevant advantages of FBG-based sensors become evident:<br />
• Totally passive no resistive heating or local power needed<br />
• Small size can be embedded or laminated<br />
Incline meter<br />
• Narrowband with wide wavelength operating range can be multiplexed
Fiber Bragg Grating Sensors: Market Overview and New Perspectives Fiber Bragg Grating Sensors 315<br />
• Non-conductive immune to electromagnetic interference<br />
• Environmentally more stable glass compared to copper<br />
• Low fiber loss at 1550 nm remote sensing<br />
Table 1: Pros and cons of FBG sensors.<br />
Table 1 summarizes several of the many benefits to intra-core FBGs used as sensing elements. However, there are<br />
also some limitations and cons associated with gratings. Most fundamentally, it is the fact that they are<br />
simultaneously sensitive to strain, temperature, pressure and radiation effects. Hence, adequate temperature<br />
compensation is always essential in the design and commercial offering of reliable and repeatable physical sensors.<br />
Although the theory and use of FBGs dates back to the late 1980s, the actual commercial transition did not happen<br />
until the mid-1990s, when it was strongly driven by the vast communications needs at the time coupled with the<br />
rampant demand for components brought on by the so-called telecommunications bubble, which saw a tremendous<br />
explosion on the number of companies and research groups engaged with the design, fabrication, packaging and use<br />
of gratings. The significant milestones and timeline evolution of the FBG industry over the past 30 years is<br />
illustrated in (Fig. 3).<br />
Figure 3: Historical evolution of fiber Bragg gratings.<br />
Soon after the telecommunications bubble collapse (circa 2002), there was a significant shift by many players in the<br />
industry from communications to sensing applications. At the time, this was a prudent and strategic move on the part<br />
of FBG manufacturers to keep exploiting the technical and manufacturing infrastructure they had available and ride<br />
the telecomm crisis until a possible comeback. At the present time, with the dust from the telecomm “bubble”<br />
already settled, it is clear that the original expectations and market potential originally envisioned for FBG<br />
components in the communications sector has not materialized and that the industry is operating at more realistic<br />
levels that match the trends at pre-bubble years.<br />
Nevertheless, the sensing sector benefited tremendously from this shift and resulted in an increase in activity and<br />
demand of FBG-based sensors. However, the impact of the frenzy of mergers and acquisition at the peak of the
A<br />
Fiber Bragg Grating Sensors: Recent Advancements, Industrial Applications and Market Exploitation, 2011, 321-322 321<br />
Accelerometer 134-159<br />
B<br />
Birefringence 110-133<br />
Bulk and integrated optics 73-91<br />
C<br />
Composite structures 160-173<br />
Coupled mode theory 33-47<br />
E<br />
Electricity transmission 174-185<br />
F<br />
Femtosecond laser 9-32<br />
Fossil fuel power generation 174-185<br />
G<br />
Grating inscription within polymeric fibers 276-295<br />
Guiding mechanisms in optical fibers 255-275<br />
I<br />
Interferometric interrogation 73-91<br />
Interferometric technique 9-32<br />
Inverse scattering methods 33-47<br />
L<br />
Laser sensing 73-91<br />
Light polarization 110-133<br />
Liquid pressure sensing 134-159<br />
Load sensing 134-159<br />
Local heat treatment 48-72<br />
M<br />
Market barriers 296-303<br />
Microstructured fiber Bragg grating 48-72<br />
Multi-functionality 186-206<br />
Multiplexing 186-206<br />
N<br />
Nuclear 174-185<br />
INDEX<br />
Andrea Cusano, Antonello Cutolo and Jacques Albert (Eds)<br />
All rights reserved - © 2011 <strong>Bentham</strong> <strong>Science</strong> Publishers Ltd.
322 Fiber Bragg Grating Sensors Cusano et al.<br />
O<br />
Oil and gas applications 174-185<br />
P<br />
Passive sensing 186-206<br />
Peculiar properties of polymers 276-295<br />
Phase mask 9-32<br />
Photonic bandgap 48-72<br />
Photosensitivity 1-8<br />
Physical sensing 255-275<br />
Polymeric materials 276-295<br />
R<br />
Radiation effects on fiber Bragg gratings 207-224<br />
Radiation environments 207-224<br />
Radiation-matter interaction 207-224<br />
Refractive index sensing 255-275<br />
S<br />
Sensing 1-8<br />
Sensors market 296-303<br />
Space structures 160-173<br />
Spatial division multiplexing 92-109<br />
Standardization 296-303<br />
Strain sensing 134-159<br />
Structural health monitoring 160-173<br />
T<br />
Telecommunications 1-8<br />
Temperature sensing 134-159<br />
Tidal energy 174-185<br />
Tilt sensing 134-159<br />
Tilted fiber Bragg gratings 225-254<br />
Time division multiplexing 92-109<br />
Transfer matrix method 33-47<br />
Transverse strain sensing 110-133<br />
U<br />
Ultra-long reach sensing 186-206<br />
Unconventional optical fibers 225-254<br />
W<br />
Wavelength division multiplexing 92-109<br />
Wavelength encoding 186-206<br />
Wet chemical etching 225-254<br />
Wind energy 174-185