See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/372134628

Design, Fabrication, and Characterization of Filament Based Fiber Bragg Grating

Sensors

Thesis · June 2023

CITATIONS READS

0 156

1 author:

Abdullah Rahnama

25 PUBLICATIONS 61 CITATIONS

SEE PROFILE

All content following this page was uploaded by Abdullah Rahnama on 06 July 2023.

The user has requested enhancement of the downloaded file.

DESIGN, FABRICATION, AND CHARACTERIZATION OF
FILAMENT BASED FIBER BRAGG GRATING SENSORS

by

Abdullah Rahnama

A thesis submitted in conformity with the requirements

for the degree of Doctor of Philosophy
Graduate Department of the Edward S. Rogers Sr. Department of
Electrical and Computer Engineering
University of Toronto

© Copyright by 2023 by Abdullah Rahnama

 Design, Fabrication, and Characterization of Filament Based Fiber
Bragg Grating Sensors

Abdullah Rahnama

Doctor of Philosophy

Graduate Department of the Edward S. Rogers Sr. Department

of Electrical and Computer Engineering
University of Toronto

2023

Abstract

In this thesis, femtosecond laser writing is harnessed towards advancing capabilities in the
fabrication of novel fiber Bragg grating based sensors. Femtosecond laser pulses inscribed low-
and high-contrast filament grating structures in single-mode fibers. The transmitted, reflected and
radiated diffraction from these gratings have been tailored for various applications by optimizing
the focal beam stretching, laser exposure, and grating period.

In this work, the low-contrast laser filaments have been harnessed to exploit their properties for
fabrication of all-fiber spectrometers and displacement sensors. The unique geometry of the
filament shape as an optical scattering element is an enhancement of the first-order radiative
emission of fiber-guided light into cladding modes. The opportunity exploited in this thesis is
drawn from filament based FBGs which permits strong and directional radiation modes to escape
from the fiber. The above advantages have been resulted in the designing and fabricating of all-
fiber spectrometers. Furthermore, the thin geometry of laser filaments facilitates a precise off-
centered positioning of filament arrays in the fiber core. This flexibility permits strain-optic
responses for azimuthally resolved displacement sensing in single-mode fiber. Therefore, the
special shape of the laser written filaments resulted in the designing and fabricating of opto-
mechanical displacement sensors based on off-axis filament FBGs.

ii
For the case of high-contrast filaments, the laser-processing was further advanced to high energies
sufficient to open arrays of hollow filament tracks inside the core waveguiding. The ordered Bragg
gratings presented an usually large refractive index contrast of 0.4 into the fiber core, opening a
new means for creating strong and compact photonic-stop band devices. The filament geometry
underpinned a strong capillarity effect for wetting with various liquids which demonstrated high-
resolution refractive index sensing through an intact fiber cladding. In another application,
Nematic liquid crystal was introduced into the nano-hole grating arrays. Capillary alignment has
resulted in a strong polarization extinction ratio within a broad Bragg resonance which enabled
design of all-fiber, dynamically tunable, polarization filters.

The present work is promising to transform how fibers shape the flow of light and sense the local
environment from applications in biomedical probes through to large area communication
networks.

iii
 Acknowledgments
Firstly, I want to devote my deepest gratitude to my wife, Elaheh, for her unconditional support,
kindness, and a tremendous help. You helped me during each step of this journey and I believe
that you made me strong and helped me keep going during the hardships. Thank You! I would like
to thank my mother, twin brother, and in-laws for all their support and love during my Ph.D
journey.

I want to thank my supervisor, Prof. Peter Herman, for giving me the opportunity to work in his
research group and for providing extensive help and guidance on my research projects. I also want
to thank him to give me the freedom to work on various projects and present my results in various
conferences. I would also like to thank Prof. Amr Helmy, Prof. Stewart Aitchison, and Prof. Joyce
Poon for being on my defense committee.

During my time at the University of Toronto, I was privileged to meet and work with an amazing
group of people who I call my friends and I am greatly appreciative of them. First of all, I would
like to thank my older brother and friend Prof. Keivan Mahmoud Aghdami who I enjoyed and
learned a lot from working with him in the lab. I would like to thank Young (Steve) Hwan Kim,
the greatest lab mate I ever had for his friendship, help, and support in and outside of the lab.
Thanks Steve! Next, I would love to thank my friend Gligor Djogo for his endless support and
friendship. I enjoyed chatting with him during working in the lab or during our lunchtime about
science, sport, and politics which made the lab work very memorable. Thank you, Gligor! I would
love to thank Dr. Can Kerse and Dr. Ehsan Alimohammadian for their valuable friendships and
support. Can and Ehsan, enjoying a cup of coffee with you two and Gligor after our launches while
talking about different topics was one of the greatest memories that I will always remember from
my time at UofT. I want to thank Hossein Mahlooji for his insight and friendship. I would like to
thank my other friends and colleagues at UofT, Polina Zavyalova, Amir Rahimnouri, Michael D.
Singh, Dr. Stephen Ho, Dr. Armin Sedighian Rasouli, Dr. Moein Shayegannia, and Yueqi Wang
for their valuable friendship and multiple scientific discussions. I am particularly thankful to Dr.
Jianzhao Li for his friendship, mentorship, amazing support, and valuable advice during my Ph.D.

During the last few years, I had the unique opportunity of working closely with different research
groups on multiple projects. I want to thank Prof. Tigran Galstian and Dr. Tigran Dadalyan from
the Department of Physics at Université Laval, and Prof. Fae Azhari from the Department of
iv
Mechanical and Industrial Engineering at the University of Toronto for all their help and supports
during our collaborative projects.

Lastly, I would like to acknowledge and thank the scholarships and research funds I received
during my Ph.D. studies from the Edward S. Rogers Sr. Department of Electrical and Computer
Engineering, the Natural Science and Engineering Research Council of Canada (NSERC), The
International Society for Optics and Photonics (SPIE), and Mitacs.

v
 Dedication

Dedicated to the precious memories of my father (the late Dr. Majid Rahnama) who was my big
supporter in my life and always encouraging me to follow my dreams. Alas, no words can express
my love, feelings, … for him. May God bless him.

vi
 Table of Contents

Abstract ........................................................................................................................................... ii

Acknowledgments.......................................................................................................................... iv

Dedication ...................................................................................................................................... vi

Table of Contents .......................................................................................................................... vii

List of Publications ..........................................................................................................................x

List of Conference Proceedings ..................................................................................................... xi

List of Figures .............................................................................................................................. xiv

List of Tables ............................................................................................................................... xix

List of Abbreviations .....................................................................................................................xx

Introduction .................................................................................................................................1

1.1 Fiber Optic Sensing..............................................................................................................1

1.1 Fiber Sensing Devices..........................................................................................................3

1.1.1 Fiber Spectrometers .................................................................................................4

1.1.2 Fiber Optic Displacement Sensors ...........................................................................5

1.1.3 Fiber Optic Refractive Index sensors .......................................................................6

1.1.4 In-fiber Polarization Controlling Devices ................................................................8

1.1.5 Global Challenges and Opportunities in FBG Design and Fabrication ...................9

1.2 Thesis Objectives ...............................................................................................................10

1.3 Chapter by Chapter Outline ...............................................................................................12

Background ...............................................................................................................................14

2.1 Ultrafast Laser Micro/nano Machining in Transparent Materials .....................................14

2.1.1 Material Interaction with Ultrafast Lasers .............................................................16

vii
 2.1.2 Filament formation by Ultrashort Laser Pulses .....................................................18

2.1.3 Glass and Fiber Modification with Filaments ........................................................20

2.2 Laser Fabrication of Fiber Bragg Gratings ........................................................................21

2.2.1 FBG Theory ...........................................................................................................21

2.2.2 UV and Interferometric Fabrication of FBGs ........................................................22

2.2.3 Direct Writing of FBGs .........................................................................................23

Experimental Setup of Femtosecond Laser Fabrication System and FBG Characterization....25

3.1 Femtosecond Laser System................................................................................................25

3.1.1 Beam Delivery .......................................................................................................26

3.1.2 Motion and Alignment System ..............................................................................27

3.2 Fiber Preparation ................................................................................................................28

3.3 Optical Fiber Characterization ...........................................................................................29

Filament FBG Processing and Characterization .......................................................................30

4.1 Filament FBG Processing ..................................................................................................30

4.2 Filament Morphology ........................................................................................................33

4.2.1 Optical Microscopy ................................................................................................33

4.2.2 Scanning Electron Microscopy ..............................................................................34

4.3 Chemical Etching ...............................................................................................................36

4.4 Optical Characterization ....................................................................................................37

Low-Contrast Filament Based FBG Devices ............................................................................39

5.1 All-fiber Spectrometer .......................................................................................................40

5.2 Opto-mechanical Motion Sensor .......................................................................................48

High-Contrast Hollow Filament Based FBG Devices ..............................................................61

6.1 Opto-fluidic Sensor ............................................................................................................61

viii
6.2 In‐Fiber Switchable Polarization Filter ..............................................................................71

6.3 Discussion and Significance ..............................................................................................83

Conclusion and Future Works ...................................................................................................86

7.1 Conclusion .........................................................................................................................86

7.2 Future Work .......................................................................................................................89

References .................................................................................................................................91

ix
 List of Publications

[1] A. Rahnama, M. Mahlooji, G. Djogo, F. Azhari, P. R. Herman "Off-Axis Filament Based

Fiber Bragg Gratings for Azimuthally Resolved Displacement Sensing" Optics Express,
30(3), 4189, 2022. Link

[2] K. M. Aghdami*, A. Rahnama*, E. Ertorer and P. R. Herman "Laser Nano-Filament

Explosion for Enabling Open-Grating Sensing in Optical Fiber" Nature Communications,
12, 6344, 2021 (*Equal Contribution). Link

[3] A. Rahnama, T. Dadalyan, K. M. Aghdami, T. Galstian, P. R. Herman "In-Fiber Switchable

Polarization Filter Based on Liquid Crystal Filled Hollow-Filament Bragg Gratings"
Advanced Optical Materials, 19(9), 2100054, 2021. Link

[4] A. Rahnama, K. M. Aghdami, Y. H. Kim, P. R. Herman "Ultra‐Compact Lens‐Less

“Spectrometer‐in‐Fiber” Based on Chirped Filament‐Array Gratings" Advanced Photonics
Research, 1(2), 2000026, 2020. (Featured on the Front Cover) Link

x
 List of Conference Proceedings

[1] A. Rahnama, T. Dadalyan, K. M. Aghdami, T. Galstian, P. R. Herman " All-fiber,

dynamically tunable, polarization filter based on liquid crystal filled nano-channels" Proc.
SPIE PC12012, Advanced Fabrication Technologies for Micro/Nano Optics and
Photonics XV, PC120120K, 2022. Link

[2] A. Rahnama, M. Mahlooji, G. Djogo, F. Azhari, P. R. Herman " Harnessing filament

arrays for azimuthally resolved displacement sensing in fiber Bragg gratings". Proc. SPIE
PC11991, Frontiers in Ultrafast Optics: Biomedical, Scientific, and Industrial
Applications XXII, PC119910L, 2022 Link

[3] A. Rahnama, T. Dadalyan, K. M. Aghdami, T. Galstian, P. R. Herman "In-fiber

switchable polarizer by capillary-aligned liquid crystals", in Optics Of Liquid Crystals,
2021 (Invited Talk).

[4] A. Rahnama, T. Dadalyan, K. M. Aghdami, T. Galstian, P. R. Herman "In-fiber

Polarization Control Using Nematic Liquid Crystal in Nano-Capillary Bragg Grating
Array" in Conference on Lasers and Electro-Optics, OSA Technical Digest (Optical
Society of America), paper SM1P.3, 2021. Link

[5] H. Mahlooji, A. Rahnama, G. Djogo, F. Azhari and P. R. Herman, "Off-Axis Filament

Based Fiber Bragg Gratings for Azimuthally Resolved Displacement Sensing," in 2021
Conference on Lasers and Electro-Optics Europe & European Quantum Electronics
Conference (CLEO/Europe-EQEC), 2021, 1-1. Link

[6] A. Rahnama, K. M. Aghdami, Y. H. Kim, P. R. Herman "2D filament grating array:

enabling an efficient, high-resolution lens-less all-fiber spectrometer," Proc. SPIE 11676,
Frontiers in Ultrafast Optics: Biomedical, Scientific, and Industrial Applications XXI,
116760Y, 2021. Link

[7] P. Zavyalova, A. Rahnama, E. Alimohammadian, J. Li, S. Sivanandam, P. R. Herman

"From filaments to light-sheets: tailoring the spectrum of fiber Bragg gratings with

xi
 femtosecond lasers", Proc. SPIE 11676, Frontiers in Ultrafast Optics: Biomedical,
Scientific, and Industrial Applications XXI, 116760X, 2021. Link

[8] O. B. Vorobyev, Y. H. Kim, J. Li, M. Bakaic, N. Burgwin, A. Rahnama, P.R. Herman

"Femtosecond laser welding of silica glass fiber for robust Bragg grating sensing in high
temperature environment," Proc. SPIE 11676, Frontiers in Ultrafast Optics: Biomedical,
Scientific, and Industrial Applications XXI, 1167618, 2021. Link

[9] A. Rahnama, Y. H. Kim, K. M. Aghdami, and P. R. Herman "Visible-light, All-fiber

Spectrometer Based on Radiative Emission From a Chirped Filament Grating Array" in
OSA Advanced Photonics Congress (AP) 2020, OSA Technical Digest (Optical Society of
America), SoTh1H.4, 2020. Link

[10] A. Rahnama, E. Alimohammadian, P. Zavyalova, J. Li, K. M. Aghdami, and P. R. Herman

"Spatial Light Shaping of Uniform Femtosecond Laser Sheets for Fabricating Fiber Bragg
Gratings" in OSA Advanced Photonics Congress (AP) 2020, OSA Technical Digest
(Optical Society of America), ITh2B.4, 2020. Link

[11] P. R. Herman, K. M. Aghdami, E. Ertorer, S. Ho, J. Li, and A. Rahnama "Femtosecond

Laser Nano-Filament Explosion: Opening Fiber Bragg Gratings for Opto-Fluidic
Sensing," in Conference on Lasers and Electro-Optics, OSA Technical Digest (Optical
Society of America), STh1H.2, 2020. Link

[12] A. Rahnama, K. M. Aghdami, Y. H. Kim, E. Ertorer, and P. R. Herman "Ultra-compact

'Spectrometer-in-Fiber' based on chirped filament-array gratings", Proc. SPIE 11270,
Frontiers in Ultrafast Optics: Biomedical, Scientific, and Industrial Applications XX,
112700W, 2020. Link

[13] P. R. Herman, E. Alimohammadian, K. M. Aghdami, E. Ertorer, Y. H. Kim, J. Li, and A.

Rahnama "Harnessing femtosecond laser filaments for nano-structuring of "Lab-in-Fiber"
sensors and "Spectrometer-in-Fiber" microsystems ", Proc. SPIE 11292, Advanced
Fabrication Technologies for Micro/Nano Optics and Photonics XIII, 1129202, 2020
(Invited Talk). Link

xii
[14] K. M. Aghdami, E. Ertorer, A. Rahnama, P. R. Herman "Femtosecond Laser Opening of
Hollow-Filament Arrays: the Fiber Bragg Grating Opto-fluidic Sensor," in Photonics
North (PN), Quebec City, QC, Canada, 2019, pp. 1-2. Link

xiii
 List of Figures

Figure 1.1: Schematical arrangement of a distributed FBG system for remote sensing of various
parameters. Image is reproduced, with permission from ref. [14]. © 2019 MDPI. ........................ 1

Figure 1.2: Schematic of different microfluidic and optical elements in fiber for a variety of lab-
in-fiber and lab-on-fiber sensing approaches. Images are reproduced, with permission from ref.
[13,15]. © 2019 Elsevier, and © 2016 Wiley-VCH GmbH. .......................................................... 2

Figure 1.3: A schematic of different optical fiber spectrometers with (a) an index matching prism
[35] and (b) cylindrical lenses for focusing of the diffracted light on the CCD sensor [36]. © IET
1997, and © Optica 2019. ............................................................................................................... 4

Figure 1.4: Schematic diagram of TFBG (a) and off-axis FBGs (b) for optomechanical sensing.
Images are reproduced, with permission from ref. [60,64]. © 2011 Elsevier, and © 2018 Optica. 6

Figure 1.5: (a) The longitudinal cross section of nano-holes formed by Bessel-beam. (b) Top view
of a glass substrate showing nano-holes formed with different pulses energies having different
diameters Images are reproduced, with permission from ref. [90]. © 2019 Optica. ..................... 7

Figure 1.6: Schematic diagram of a photonics crystal hollow core fiber filled with liquid crystal
for in-fiber polarization control applications. Image is reproduced, with permission from ref. [94].
© 2019 IEEE. .................................................................................................................................. 8

Figure 2.1: Schematic of laser micromachining in transparent material for two different pulse
durations and their effects on the HAZ. (a) Long pulsed or CW laser beams would ablate the
surface and generate a large HAZ. (b) Ultrashort laser pulses could otherwise modify internal
volume of glass without the damaging the glass surface while the HAZ may be reduced to below
the size of the focused beam . Image is reproduced, with permission from ref. [109]. © 2014 MDPI.
....................................................................................................................................................... 15

Figure 2.2: Schematic diagram for different nonlinear laser processes for light absorption in
transparent glass: (a) Multi-photon ionization. (b) Tunnel ionization. (c) Collisional ionization.
Images are reproduced, with permission from ref. [118]. © 2020 Springer. ................................ 16

Figure 2.3: Different optical means for the formation of filaments in glass such as (a) Kerr lensing,
(b) Bessel beams, and (c) aberration generated at an interface, also called surface aberration.
Images are reproduced, with permission from ref. [118,128,131]. © 2020 Springer, and © 2020
MDPI, and © 2014 Optica. ........................................................................................................... 19

Figure 2.4: (a) Fabrication of filaments in two-dimensional arrays, defining various types of
volume Bragg grating structures in bulk glass. (b) Tightly packing of filaments in fiber core further
allows fabrication of FBGs with unique radiation and reflection properties. Images are reproduced,
with permission from ref. [134,31]. © 2013, and 2018 Optica. ................................................... 20

xiv
Figure 2.5: Shematic diagram of FBGs’ working principle. Grating pitch, Ʌ, determines the
reflected wavelength appearing in reflection spectra or as a notch in the transmission spectrum
transmission spectra. Image is reproduced, with permission from ref. [136]. © 2017 MDPI. ..... 21

Figure 2.6: Interference-based inscription techniques would form spatial interference patterns
inside the fiber core. These methods are divided into two main designs: (a) interference of two
beam of light or (b) different orders of a phase mask . Images are reproduced, with permission
from ref. [138,30]. © 2017, and 2011 Optica. .............................................................................. 22

Figure 2.7: Schematic diagrams of different femtosecond direct laser writing geometries of
creating FBGs such as point-by-point (PbP), line-by-line (LbL), and plane-by-plane (Pl-b-pl).
Image is reproduced, with permission from ref. [141]. © 2020 IEEE. ......................................... 23

Figure 3.1: Schematic of experimental setup for point-by-point fabrication of FBGs in single-
mode optical fiber by ultrashort laser pulses [142]. ...................................................................... 25

Figure 3.2: Photos of the experimental femtosecond laser fabrication setup. (a) Laser beam
delivery system containing mirrors, lenses, irises and translational stages (b) Femtosecond laser
system with a beam expander (c) Back reflection alignment diagnostic setup including a CCD
camera and imaging lens. Green lines depict the laser path. ........................................................ 26

Figure 4.1: Schematic of the femtosecond laser focusing arrangement for filament-by-filament
FBG writing into a single-mode fiber. Long and uniform filament shape of light was generated by
surface aberration from pre-focusing the laser beam through the fused silica plate of 2 mm
thickness [100]. ............................................................................................................................. 31

Figure 4.2: Optical microscopy of the filament array showing the lateral and central alignment of
the array to the fiber core in the cross section (a) and top (c) view. The side view (b) of the fiber
shows formation of filament tracks at various focal depths. (d) Time sequence (left to right) of
evaporating isopropanol with menisci level (orange arrow) moving downward inside views of the
nanoholes recorded over ~10s [31,104]. ....................................................................................... 33

Figure 4.3: Schematical view of a nanohole embedded optical fiber with angled facet (a). Under
increasing magnification, SEM images of the angled fiber facet (b–d) unveil open nanoholes
having formed continuously and uniformly (~200 nm diameter) through the core (white oval zone)
and cladding without a breakthrough on the tight 1.072 µm periodic spacing [104]. .................. 35

Figure 4.4: The point‐by‐point fabrication was combined with post-chemical etching to engineer
strong photonic stopbands directly inside of the compact and flexible fiber. The hole diameter
expanded from 200 nm (a) to 350 nm (b) after 4 min of chemical etching (5% HF) [104]. ....... 36

Figure 4.5: (a) Optical characterization setup for recording reflection and transmission spectra of
different FBGs. Example recorded spectra for uniform (b) and π-shifted (c) FBGs. ................... 37

Figure 5.1: (a) Schematic of all-fiber compact lens-less spectrometer. (b) Side view image of the
FBG (fiber horizontal and blurred) with the radiated spectrum in focus, showing spectrally narrow
and multi-wavelength lines sharply focused by the chirping and filament array geometry onto a
screen. The grating created two near-identical radiation patterns on opposite azimuths of the fiber,
recorded on a CCD sensor left (a) or on a paper screen with a camera (b) [100]......................... 41
xv
Figure 5.2: (a) Top view of multicolored radiation patterns projected onto a curved screen from
the fourth-order FBG, revealing the favorable azimuthal narrowing of a single filament. (b) The
reflection and transmission spectra recorded from a second-order FBG of 1 mm length [100]. .. 42

Figure 5.3: Chirped FBG showing (a) the filament spacing design for spectral focusing to point
F, (b) with grating pitch plotted versus pitch number, and (c) the resulting reflection and
transmission spectral responses of second-order chirped FBG [100]. .......................................... 43

Figure 5.4: Chirped FBG first-order radiation patterns of blue, green, and red light sources
recorded (a) onto a CCD camera (1 mm × 1 mm frames) and (b) converted to spectral lines
(colored) that reveal a source bandwidth limited resolution (black line). (c) Assembly of FBG
recordings spanning most of the visible spectrum [100]. ............................................................. 45

Figure 5.5: Full linewidth (3 dB) of the various sources measured by the chirped FBG
spectrometer (squares) and comparison with the source bandwidth as measured by a high-
resolution OSA (triangles) [100]. ................................................................................................. 46

Figure 5.6: Schematic of a cantilever-based FBG displacement sensor with two orthogonal sets of
off-axis filaments (displacements ξx and ξy) crossed through the fiber core [101]. ..................... 49

Figure 5.7: Reflection spectra (a) recorded for uniform FBGs of 2000 filaments (Epulse = 1 µJ,
Λ = 536 nm) at different centering offsets (ξx = 0, 1, 2, 3 µm) show the decay of the reflection
strength with increasing offset. The optical microscope image (b) shows a side view of filaments
stretching more than 40 µm to fully pass through the fiber core. Top view images show single
filament-arrays with offsets ξx = -1 µm (c) and ξx = +2 µm (d) from the fiber axis [101]. .......... 50

Figure 5.8: (a) Schematic diagram of the optical characterization arrangement for probing the
FBG reflection responses under cantilever actuation. (b) Close up view of the fiber at the V-groove
clamp, imaging scattering of red waveguiding light from the filaments of the FBG. (c) Schematic
of the fiber cantilever hosting a ℓ = 1 mm long grating located near the support of a L = 30 mm
long cantilevered fiber (d) Cross-sectional view of two filaments arrays representing gratings
FBG1 and FBG2 offset by ξ1 and ξ2, respectively. The fiber was rotated by θ from the horizontal
neutral axis. The waveguide core is shown non-concentric with the theoretical central axis. (e)
Reflection spectra recorded from a first-order FBG (ξx ≅ 2 µm, Λ = 536 nm, ℓ = 1 mm) under bent
(δ = 2.25 mm) and straight conditions, indicating a stress-induced wavelength shift of the Bragg
resonance (inset). Note: Technical drawing of select components in the SolidWorks drawing were
acquired from THORLABS, Inc [101]. ........................................................................................ 52

Figure 5.9: (a) Temporal response of Bragg resonance wavelengths following two first-order
FBGs with parallel and offset filaments in horizontal position (ξx1 = +2 µm, ξx2 = -1 µm, ΛΛ = 536
nm, ℓ = 1 mm, θ = 0°) and driven under square-wave cantilever agitation (δ = 2.25 mm).
Compressive and tensile stresses induced on opposite sides of the neutral axis led opposing signs
in the respective wavelength shifts, ΔλB1 and ΔλB2. (b) Wavelength shifts from the double grating
sensor showing the azimuthal cantilever response for a tip displacement of δ = 2.25 mm and
sinusoidal representations of the data (red, blue lines) [101]. ...................................................... 54

Figure 5.10: (a) Bragg wavelength shifts of a first-order FBG (ξx ≅ 2 µm, Λ = 536 nm, ℓ = 1 mm)
for different displacements (δ = 0 to 2.25 mm) and azimuths (θ = 0° to 180°). (b) The displacement
response, ΔλΒ/δ, from (a) presented on polar plot (blue dots) for varying rotation angle (θ) and the
xvi
corresponding simulation data for a grating offset of ξ = 2 (red) and the variance (grey) over grating
offsets of ξ = 1.5 to 2.5 µm, which is consistent with the ± 0.5 µm laser writing precision. (c) Bragg
wavelength shift of FBGs at different offsets (ξx ≅ 0, 1, 2 µm) recorded over a 2π azimuth (green,
red, blue dots) for a fixed tip displacement of δ = 2.25 mm and comparison with sinusoidal
responses (solid lines) best representing the data at corrected offset values, ξsim. ....................... 57

Figure 6.1: Schematical arrangement of optical aberration plate and optical fiber used for
generating femtosecond laser filaments, and opening high-aspect ratio nanoholes through the fiber
cladding and core waveguide [104]. ............................................................................................. 62

Figure 6.2: Reflection spectra (a) recorded from a uniform FBG of 600 nanoholes
(Epulse = 4.5 µJ, Ʌ = 1072 nm, 2 mm plate) forming strong (3 dB) and wide (~2 nm) stopbands
when filled with low (nH = 1) or high (nH = 1.661) refractive index materials. The colored bands
mark the 3 dB bandwidth. Refractive index matching oil (nH = 1.448) dramatically weakens the
stopband. Reflection spectra (solid lines) of unpolarized light (b–e) were recorded from a π-shifted
FBG of 1200 nanoholes for four different examples of RI values (nH by color from legend a) and
comparison with simulated spectra for P-polarized light (dashed lines). Laser-induced
birefringence of δλ = 210 pm is resolved for the air-filled case (b inset). Gaussian line fits of the π-
defects (blue and red lines in b inset) provide ±10 pm spectral precision [104]. ......................... 64

Figure 6.3: Reflection spectra recorded from uniform FBGs (Epulse = 4.5 µJ, Ʌ = 1072 nm, 600
holes) filled with low (a nH= 1), medium (b nH = 1.44), and high (c nH = 1.661) refractive index
materials, revealing the influence of increasing nanohole diameter due to varying chemical etching
time (0–6 min). Peak reflection strength (d) and 3-dB linewidth (e) of the Bragg resonances plotted
as a function of refractive index for different chemical etching times (following legend a). Bragg
resonance wavelength (shifts on right axis) plotted versus refractive index (f) for different chemical
etching times. A global fit of the data by EME simulation is presented (dashed lines) for hole
diameters of 220–700 nm (see legend). The slopes of data provide a strongly increasing refractive
index sensitivity (inset) for increasing RI and increasing hole diameter (color-coded lines). Shaded
zones mark axis scale changes [104]. ........................................................................................... 67

Figure 6.4: The complete set of the reflection spectra recorded from π-shifted FBGs having 1200
nano-hole filaments (Epulse = 4.5 µJ, Ʌ = 1072 nm) and their widely varying response to liquids
with varying RI values (color coded and as labelled in (a)) after undergoing 0 (a), 2 (b), 4 (c) and
6 (d) minutes of chemical etching time. The unetched nano-holes having smallest diameter (a)
provided the strongest reflection resonances for a majority of the liquids evaluated, while an
increasing hole diameter (b to d) provided an increasing Bragg wavelength shift for improving
refractive index sensitivity, but with weakening overall reflection strength. The π-shift resonance
for laser-formed and chemically etched holes appears as narrow (50 to 370 pm) transmission
windows that sharpened the sensing resolution of Bragg resonance. However, a birefringent
broadening of the π-defect resonance plays out differently in the spectra according to the refractive
index contrast and hole diameter. In the sequence of increasing hole diameter (a-d) for air filled
holes (negative refractive index contrast, n = -0.45), the π-defect displays a moderately rising
birefringence of δλ = 210 to 370 pm for chemical etching times rising from 0 to 6 min.
Birefringence broadening of δλ  20 pm is also strongly evident in select cases of positive
refractive index contrast (i.e., n = 0.1 at 2 and 4 min etch for Oil6 and Oil7 in (b) or (c)).
Otherwise, the birefringence is unresolved for a majority of the liquid solvents (1.31 < n H< 1.67)
and etching time (0 – 6 min) presented here [104]. ...................................................................... 69
xvii
Figure 6.5: Schematic of an in-fiber Bragg grating for polarization switching based on NLC-filled
nano-filament cavity arrays. At room temperature, NLCs favored homogeneous alignment as
depicted in the cross-sectional view (a) of the nano-holes opened to ≈200 nm diameter by
femtosecond laser nano-explosion. The nano-hole FBGs (b) were loaded with NLC by the
capillarity effect and the stopbands were spectrally tested for thermo-optic and electro-rotation
responses using a thermal heater and capacitive cell. When tuned to the FBG stopband for vertical
polarization (0o, z axis, n∥ = 1.661), only horizontal polarized light (90°, y axis, n⊥ = 1.511)
transmits through the grating [105]............................................................................................... 73

Figure 6.6: (a) Schematic arrangement for optical characterization of the reflection and
transmission polarization responses in the LC filled FBG. (b) Transmission spectra recorded
through a second-order FBG of 1000 nano-holes (Λ = 1.072 µm) without LC filling, comparing
the two states of linear polarization aligning parallel (0°) and perpendicular (90°) with respect to
filament orientation. The air-filled capillaries induce a small birefringent Bragg shift of Δλ = 0.25
nm [105]. ....................................................................................................................................... 75

Figure 6.7: Reflection and transmission spectra recorded from a uniform FBG (Λ = 1072 nm,
1000 holes) filled with NLC for (a) perpendicular (90°) and (b) parallel (0°) polarization states.
The respective stopbands have shifted apart according to the perpendicular (a; green) and parallel
(b; amber) polarization alignment with the ordinary (n⊥ = 1.511) and extraordinary (n∥ = 1.661)
refractive indices of the NLC. The spectra (c) of polarization extinction ratios (parallel (a) divided
by perpendicular (b)) for transmission (PERT, green line) and reflection (PERR, orange line)
present distinct bands of high extinction of the parallel (PERT, amber) or perpendicular (green)
polarized light, which extends to short wavelengths (PERT, pink) due to the high cladding mode
loss (b) seen for the parallel polarization state. A large component of the rejected parallel polarized
light is reflected, appearing symmetrically, further provides in the PERT spectrum with a negative
value (amber and pink bands). The recorded transmission (d) and reflection (e) spectra (solid lines)
are seen to follow approximately with the stopband and sidelobe positions predicted in the EME
simulated spectra (dashed lines) [105]. ......................................................................................... 77

Figure 6.8: A sequence of (a) transmission spectra recorded through a second-order FBG (1000
nanoholes) filled with NLC for incident linearly polarized light aligned at varying angles of 0° to
90° with respect to filament orientation. Polar plot of the (b) transmittance measured at 1555 nm
wavelength from (a) (black squares) and comparison with the ideal linear polarizer response (red
circles) from Malus’ law [105]. .................................................................................................... 79

Figure 6.9: The thermo-optic response of 2nd order FBG (1000 nanoholes) filled with NLC
showing a sequence of transmission spectra (a) recorded for parallel polarization over temperature
range of 20 to 35 °C. The ratio of spectra (b) recorded at 26, 30 and 35 °C with respect to the room
temperature (20 °C) spectrum demonstrates a dynamically tunable polarization extinction
response. The induced thermo-optic attenuation for different temperature changes is shown in (b)
for three different values (ΔT = 6, 10, 15 °C) [105]. .................................................................... 80

xviii
 List of Tables

Table 1: Refractive index (RI) of liquids applied inside of nanohole FBGs. .............................. 63

xix
 List of Abbreviations

1D one dimensional

2D two dimensional

3D three dimensional

CCD charge-coupled device

CI collisional ionization

CW continuous wave

FBG fiber Bragg grating

FDLW femtosecond direct laser writing

FEA finite element analysis

FWHM full width at half maximum

HAZ heat affected zone

IR Infrared

LC liquid crystal

MCF multi core fiber

MFD mode field diameter

MMF multi mode fiber

MPI multiphoton ionization

NA numerical aperture

xx
OSA optical spectrum analyzer

OTDR optical time-domain reflectometry

PDL polarization dependent loss

PER polarization extinction ratio

RI refractive Index

RIU refractive index unit

SEM scanning electron microscopy

SMF single mode fiber

SNR signal to noise ratio

TI tunnel ionization

UV ultraviolet

VBG volume Bragg grating

VIS visible

xxi
“I do not know what I may appear to the world, but to myself I seem to have been only like a boy
playing on the seashore, and diverting myself in now and then finding a smoother pebble or a
prettier shell than ordinary, whilst the great ocean of truth lay all undiscovered before me.”

Sir Isaac Newton

xxii
 Introduction
1.1 Fiber Optic Sensing
During last few decades, optical fibers have found enormous applications in our daily lives. One
of the most important applications of optical fibers is in long-distance communications where low
attenuation of infrared (IR) wavelength light through the fused silica waveguide offers high
reliability and high capacity [1]. Moreover, the compact and flexible format of optical fiber affords
highly dense packaging of network designs serving broadly over our information highway [2]
through to facilitating efficient interconnection in data centers [3].

Beyond telecommunications, optical fibers have found applications serving in various types of
optical sensing. Their compact shape and optical properties further facilitate sensing of the local
environment that can reach over short to long distances and probe into harsh environments
(Figure 1.1) [4-6].

Sensing structures may be embedded into optical fiber having different forms such as a Fabry–
Perot interferometer, [7,8] fiber Bragg grating (FBG) [4,9] or an optical time-domain reflectometer
(OTDR) [10]. Optical fiber sensors offer unique advantages over their electronic or electro-
mechanical counterparts because of their compact size, flexibility, immunity to electrical and

Figure 1.1: Schematical arrangement of a distributed FBG system for

remote sensing of various parameters. Image is reproduced, with permission
from ref. [14]. © 2019 MDPI.

1
magnetic field, inert and biocompatible properties, ease of system multiplexing, and overall ability
to withstand high temperature or pressure [6,11-13]. Further, the discrete sensing elements can be
distributed and multiplexed over large areas as shown the parallel fibers of FBG arrays as shown
in in Figure 1.1 to support remote and large area sensing of multiple parameters [14].

In the design of a fiber optic sensor, the grating or interferometric sensing elements may be
embedded by positioning micro or nano structures either directly inside of the fiber core waveguide
to engineer a Lab-in-fiber sensor as shown in Figure 1.2(a,b), or on the fiber cladding surface for
Lab-around-fiber sensing (Figure 1.2(a,b)), or on the fiber end-facet for Lab-on-tip sensing
(Figure 1.2(a)) [13,15]. The physical structures may be applied by means of chemical etching [16],
mechanical polishing, thermal tapering [17], laser modification [18,20], or ion milling [21]. Such
structures can be engineered to directly or indirectly interact with the waveguiding light in a such
a way to probe both internal and external fiber environmental properties such as temperature,
pressure, vibration, and refractive index. Among these methods, laser machining of fiber has

(a)

(b)

Figure 1.2: Schematic of different microfluidic and optical elements in fiber for a
variety of lab-in-fiber and lab-on-fiber sensing approaches. Images are reproduced,
with permission from ref. [13,15]. © 2019 Elsevier, and © 2016 Wiley-VCH GmbH.

2
offered a highly flexible means for micro- and nano-structuring of the mechanical and optical
properties of fibers and enabling a facile tailoring and tuning of spectral responses for various
sensing applications [19].

1.1 Fiber Sensing Devices

Among many different fiber sensor types, FBGs have become one of the most favored devices that
have found a large commercial market owing to strong environmentally resonant responses
available in highly localized fiber positions. FBG based sensors further benefit from the ease of
external laser fabrication through the cladding core. Currently, FBGs can be tailored to a precise
spectral design to serve as notch or dispersive filters that are used as fiber laser
mirrors [22], wavelength-division multiplexers [23], dispersion compensators in
telecommunication networks [24], thermo-mechanical sensors [25], and other functions in
aerospace, energy, defense, and medical applications [4,11].

FBGs are typically fabricated by various means of laser modification that are dictated by the
specific application requirement. Interference-based FBG fabrication often centers on splitting a
laser beam with a phase mask [26] that yields reproducible fabrication of FBGs for commercial
markets. However, the phase mask typically requires time-consuming exposures and are limited
in flexibility for tailoring or tuning of the spectral features due to the fixed phase mask design. For
this reason, point-by-point writing of FBGs, as introduced with excimer laser exposure [27],
facilitate high flexibility in varying the grating design. The point-by-point technique was extended
to femtosecond laser exposure [28] to take advantage of driving strong photosensitivity responses
regardless of the fiber material, and thus provide uniform refractive index modification across the
cladding and core without a further need of hydrogen loading. The beam-shaping geometry for
femtosecond laser fabrication has been extended to line-by-line writing [29], sweeping phase mask
exposure [30], and filament array writing [31], representing a varied means for tuning the shape
of the volume grating to inhibit or enhance radiative and cladding mode effects.

The next sub-sections (1.1.1 – 1.1.4) review the background and identify key challenges in four
highlighted areas of fiber optic sensing targeted for study in this work: fiber spectrometers, fiber
optic motion sensors, optofluidics fiber sensors, and in‐fiber polarization controlling devices.

3
1.1.1 Fiber Spectrometers
Traditional FBGs rely on transversely uniform refractive index structures (planar) that inhibit
diffraction to cladding or radiation modes [32,33] and thus trap light efficiently inside of the
waveguide. Tilting of the Bragg grating breaks this symmetry to activate coupling into cladding
and radiation modes, enabling the sensing of the outer cladding environment that is attractive for
biomedical and mechanical sensing applications [34]. Wagener et al. [35] harnessed both grating
tilting and chirping to report the first demonstration of a fiber-based grating spectrometer that
radiated and externally focused the waveguiding light into a high-resolution spectrum external to
the fiber. A resolving linewidth of ~120 pm was demonstrated in a narrow (14 nm) band at around
1550 nm wavelength in the telecommunication band (Figure 1.3(a)). This lens-less spectrometer
required prism coupling, creating a moderately large fiber spectrometer of 160 mm focal length.

Qin et al. [36] extended the fiber spectrometer to a high spectral resolution of 50 pm linewidth,
while covering a 115 nm band centered near 1550 nm wavelength (Figure 1.3(b)). The grating
planes were tilted but not chirped, and thus required two external lenses to collect and focus light,
resulting in a large device size exceeding 15 cm. Also, the 45° grating tilt lead only to p-polarized
radiation due to the near-Brewster angle orientation of the grating planes for low-index contrast
gratings.

(a) (b)

Figure 1.3: A schematic of different optical fiber spectrometers with (a) an index matching prism
[35] and (b) cylindrical lenses for focusing of the diffracted light on the CCD sensor [36]. © IET 1997,
and © Optica 2019.

Alternative forms of refractive index modification with non-planar geometry have been
demonstrated by different point-by-point laser processes that offer advantages for tuning and
optimizing the radiative modes of the gratings such as required in all-fiber spectrometers.
Waltermann et al. [37] used a femtosecond laser to form ellipsoidal refractive index structures and

4
demonstrated a chirped fiber spectrometer, offering spectral focusing and a short wavelength
extension to 810 nm.

1.1.2 Fiber Optic Displacement Sensors

In mechanical displacement sensing, optical fibers have seen widespread applications from
directing small probes to monitoring large civil structures [38,39]. These optical sensors offer
unique advantages over their electronic or electro-mechanical counterparts owing to their compact
size, flexibility, electrical and magnetic field immunity, inert and biocompatible properties, ease
of system multiplexing, and overall ability to withstand high temperature or pressure.

To overcome an inherently weak photoelastic response in fiber, displacement sensing has generally
required fibers modified with a multicore (MCF) [40-44], eccentric core [45-47], surface core [48],
or structured cladding [49,50] that reposition the core-guided light away from the neutral axis.
Otherwise, sensing elements must be positioned asymmetrically to an off-axis position within the
core or cladding of traditional fiber types, such as multimode (MMF) [51], multi-clad [52], and
single-mode fiber (SMF) [53,54]. Various methods of micro-structuring in the fiber cladding have
enabled displacement sensing through formation of tapers [55,56], tuned mass elements [57,58],
and Fabry-Perot cavities [59]. However, these approaches rely on a significantly reshaping the
fiber at the expense of structural integrity.

Alternatively, optical sensor elements have been favorably positioned in the central core
waveguide of traditional SMF through a much less invasive process, for example, by laser
inscription of a FBG generated through a periodic refractive index modulation. In this direction,
tilted FBGs (Figure 1.4 (a)) [60-62] or off-axis (Figure 1.4(b)) FBGs [53,54,63,64] have been used
for core-to-cladding coupling to boost the displacement response, but at the cost of low return light
recoupling. On the other hand, FBGs positioned away from the central axis, but at the SMF core,
have provided lowest-loss probing of photoelastic responses by directly measuring the Bragg
wavelength shift [65], the back-reflected power [64], or the higher-order mode spectral features
[66]. Inscribing FBGs further outside the core has provided a beneficial linear rise in sensitivity,
but at a high cost of an exponential drop in sensing signal [67]. Previously, direct writing with
focused ultrashort laser pulses had induced fiber gratings with phase elements localized over small
cross-sectional zones of ∼1 × 2 µm2 area [65] or larger [64,66], over which a varying photoelastic

5
response may be optically detected. Further down-scaling of grating cross section to below 1 µm
allows fabricating shorter devices while having high sensitivity.

(a) (b)

Figure 1.4: Schematic diagram of TFBG (a) and off-axis FBGs (b) for optomechanical sensing.
Images are reproduced, with permission from ref. [60,64]. © 2011 Elsevier, and © 2018 Optica.

1.1.3 Fiber Optic Refractive Index sensors

FBGs are one favored device in localized optical fiber sensing owing to sharp and environmentally
responsive resonances. However, the FBG typically probes the local environment indirectly
through the thick cladding, without benefiting from the open photonic bandgap structure such as
available in silicon photonic or another planar optical circuit technologies [68-71]. An open
periodic structure with high refractive index (IR) contrast offers strong photonic bandgap
responses, which can be spectrally and spatially narrowed by optical defects to provide high-Q
micro-resonators for highly localized sensing. Such micro-resonators have been shown to enable
optical trapping [72], label-free sensing [73] or single nano-particle detection [74] on a chip, but
have been limited in fiber by a solid core FBG buried deeply in the cladding. Various means of
chemical, laser, and mechanical machining or thinning of the cladding have facilitated evanescent
sensing at the surface of the FBG [75-78]. An open-structured photonic crystal has otherwise been
challenging to fabricate transversely into the fiber core. Ion milling or laser machining at the core
of a thinned fiber has provided FBG sensors with nano-structured surface relief [79,80], blind holes
[81], and through holes [82,83]. However, the cladding processing renders such fibers
mechanically fragile and lacking robustness for practical application.

6
As a solution, femtosecond lasers provide an alternative direction for opening high-aspect-ratio
holes in transparent materials such as glasses by Kerr lensing [84]. Yang et al. accessed the core
waveguide through a narrow cladding channel to open an array of micro-holes by chemical etching
of self-focused femtosecond laser tracks [84]. Third-order Bragg stopbands provided a
microfluidic RI-sensing response, but were restricted to the low sensitivity of 5 nm/RIU and
narrow RI sensing window from 1.32–1.41 owing to a large hole diameter (1.35 µm) exceeding
the optical wavelength. Therefore, the scaling from micro-scale in Yang et al. work [84] to
nanoscale remain the main challenge to unlock a key requirement for building strong and
controllable photonics crystal stop bands for refractive index and other in fiber sensing
applications.

A promising direction to reach internally into the core waveguide of an optical fiber is by laser
micro-explosion of nanocapillaries from “filament-shaped” beams. Rather than relying on unstable
Kerr lensing effects, nano-channels (hollow-filaments) with high aspect ratios have been opened
in bulk glass with femtosecond lasers pre-shaped into an optical Bessel beam [85-88]. Zhang et al.
[89,90] have used this approach to open an array of nano-holes through a laser written waveguide,
providing a Bragg grating waveguide inside of a bulk glass plate as shown in Figure 1.5. Extension
into fiber requires overcoming of the optical aberrations from the cylindrical shaped fiber cladding.

Figure 1.5: (a) The longitudinal cross section of nano-holes formed by Bessel-beam. (b) Top view of
a glass substrate showing nano-holes formed with different pulses energies having different diameters
Images are reproduced, with permission from ref. [90]. © 2019 Optica.

7
1.1.4 In-fiber Polarization Controlling Devices
Another area for in-fiber sensing applications is embedding of nematic liquid crystals (NLC) inside
of core waveguide of optical fibers to create dynamically reconfigurable components and devices
[91-94]. The anisotropic molecules of NLCs are favored for their ease of natural self-alignment
that results in high optical birefringence, typically on the order of Δn ≅ 0.15 at 25 °C [95]. The
dynamic control may be obtained by an external stimulus such as electric or magnetic fields that
reorient the birefringence director. The birefringence can also be tuned with temperature, for
example, with Δn decreasing when approaching the phase transition temperature, e.g., ≈34.5 °C
for the 5CB [95].

However, there are challenges for opening this direction. The evanescent field-based devices have
offered a tuneable attenuation, but without polarization control. The filling of the micro-holes in
photonic crystal fibers with NLCs [91,96,97] have enabled a high modal overlap of the guided
light with the NLC, but at the cost of dramatically altering or entirely removing the stopband
structure for waveguiding of the light (Figure 1.6 shows an example of these devices). Similarly,
the presence of appreciable NLC volume in the core of standard silica optical fiber disturbs the
modal properties as the typical values of extraordinary (ne = 1.7) and ordinary (no = 1.5) refractive
indices for bulk NLC materials are noticeably higher than the refractive indices of the cladding
and the core of widely deployed single-mode telecommunication fibers. For this reason, the laser
structuring of nano-scale ripples [98] in combination with laser machining of buried microchannels

Figure 1.6: Schematic diagram of a photonics crystal hollow core fiber filled
with liquid crystal for in-fiber polarization control applications. Image is
reproduced, with permission from ref. [94]. © 2019 IEEE.

8
[99] will not meet the requirement to align NLCs and provide polarization control without
disturbing the optical properties of the fiber waveguide.

Harnessing the NLC birefringence properties directly in the core thus requires a new means for
nano-structuring of the internal volume of the fiber core in a way that does not impose dramatic
changes in the fundamental modal properties or lead to high radiative loss.

1.1.5 Global Challenges and Opportunities in FBG Design and

Fabrication
Higher-resolution and more efficient gratings are thus required to enable all-fiber spectrometers
with more compact designs, less polarization sensitivities, and broader spectral coverages.
Opening such fiber devices further to the visible spectrum has remained a major challenge.
Therefore, femtosecond lasers have been most promising to generate appropriate geometric
structures such as ellipsoidal and filaments which can in turn direct the radiation into favorable
directions or patterns into .

In the case of motion sensing with FBGs, reshaping of phase elements into narrow structures is
needed for creating ultra-thin planar gratings with femtosecond lasers. Such gratings are needed
to provide strong and narrow Bragg reflections over short lengths and even when occupying a
minimal cross-sectional area of the fiber core. FBGs with these fine structures open an opportunity
in displacement sensing, where the photoelastic response can be measured at a single plane, offset
from the neutral axis.

As mentioned previously (Section 1.1.3), one promising direction for in-fiber refractive index
sensing and polarization filtering is reaching internally into the core waveguide of an optical fiber
by laser micromachining. While there are already existing demonstrations of micro fluidic
structures in the fiber, the reported holes could only be opened after destruction of large parts of
the fiber cladding and using time-consuming and difficult multi-step processes. The scaling from
micro-scale into nano-scale holes remains a key challenge which unlocks a key requirement for
building strong and controllable photonics crystal stop bands.

9
1.2 Thesis Objectives

Ultrashort-pulsed lasers have previously provided a facile means for generating periodic arrays of
long and uniform filaments into the optical fiber core that form into FBGs with strong and narrow
spectral stop bands [31]. In this work, femtosecond laser writing has been selected as a flexible
tool to open new capabilities towards advancing capabilities in photonics fabrication of novel
optical fiber Bragg grating based sensors and devices. The thesis is composed of four multi-
disciplinary project areas where low- and high-contrast (hollow) filament FBGs have been
fabricated and studied for serving in various sensing and telecommunication applications.

In the first area, filament-shaped Bragg grating structures [31] were applied as an alternative
geometry for controlling the optical radiation patterns scattered out of the fiber cladding by the
waveguiding light in the fiber [100]. The elongated filament geometry served as a long coherent
radiating antenna that collectively enabled strong and directional radiation modes to escape from
the fiber core and cladding into a narrow azimuthal angle. The chirping of the Bragg grating array
further provided focusing onto a compact external sensor, permitting a compact fiber spectrometer
design that bypasses the bulky and expensive gratings and focusing optics required in traditional
spectrometer designs.

In the second area, the unique ultra-thin geometry of filament arrays with submicron diameter was
harnessed for optical fiber displacement sensing applications [101]. Filament based FBGs
provided strong and spectrally narrow Bragg resonances in SMF [31,100] over short lengths and
even when occupying a minimal cross-sectional area of the fiber core. Therefore, filament-by-
filament FBG writing enabled the measurements of photoelastic response at a single plane, offset
from the neutral axis. Moreover, the geometry favors a highly contrasting azimuthal sensitivity
that reaches null signal when displacement is parallel with the filament lines. Lastly, filament
gratings were overlaid with different offsets and azimuthal angles, for example, orthogonally
crossing, to harvest more precise and complete displacement data based on vector sensing
capabilities [102,103].

Third area entails the study of formation of nano-capillaries in the fiber. The laser interaction was
further scaled up to drive a controlled filament explosion in the fiber cross-section, resulting in

10
densely packed arrays of uniform nanoholes [104,105]. The isolated, blind or through holes were
patterned with controllable positioning to selectively pierce the core and/or cladding. The process
was noncontact and less invasive than any other prior art, which leaves the fiber cladding fully
intact. The hollow-filament tracks provided high aspect ratio nano-holes (~200 nm diameter) that
extended fully through the core and cladding zones. The result is a facile capability in writing
photonic structures with high contrast in refractive index along the fiber core. In this way, FBGs
with strong second order stop bands have been fabricated. The novel utility of the nano-hole
grating array is the strong capillary action by the nano-channels in bringing the outside cladding
environment directly into the core waveguide. The holes served as nano-fluidic channels for
drawing isotropic liquids such as solvents and oils from the surrounding cladding environment
directly into the fiber core without inducing high losses or de-activating waveguiding. FBGs with
strong resonances and sharply resolved π-shifts were fabricated for providing high capillarity for
wetting with various solvents and oils, demonstrating high-resolution refractive index (RI) sensing
from n = 1 – 1.67 through an intact fiber cladding [104].

In the fourth area, infiltration of anisotropic liquids such as liquid crystals (LC) to the laser formed
nano-capillaries and their polarization response were studied. Hollow filament grating presented a
strong capillarity effect to draw LC into homogeneous alignment with the cylindrical walls and
present a high birefringent response in the second-order Bragg stopband. The LC provides a strong
polarization extinction ratio with only a moderate insertion loss arising between the two
polarization states of the shifted stopbands. The Bragg grating would follow the thermo-optic
response of the LC to further offer dynamic tuning and switching of the polarization extinction
response. The flexible laser writing enables tailoring of the Bragg spectral and polarization
responses in the telecommunication C-band. The active polarization discrimination capability of
LCs is targeted for to study the dynamic polarization control in optical fibers.

As a results, the above four areas of study collectively offer development of advanced new fiber-
based sensors and devices for both local and remote sensing to potentially improve diagnostic
capabilities at the heart of optical communication networks and data centers or in medicine,
astronomy, and manufacturing that take advantage of the flexible and compact format of the optical
fiber. A brief description of each chapter is summarized in the upcoming section.

11
1.3 Chapter by Chapter Outline
This thesis study presents design, fabrication, and characterization of filament-based FBGs and the
further potential for filament-FBG applications in spectroscopy, displacement , and refractive
index sensing are organized into the following chapters:

Chapter 2 (Background)

This chapter provides a brief description on the interaction of ultra-short laser pulses with
transparent materials and their application for fabricating various optical devices in bulk glass and
fiber. The interactions are directed towards generation of filament structures in transparent glass
to serve as FBG devices. The chapter closes with a review of models and the current fabrication
methods of FBGs.

Chapter 3 (Experimental Methods)

In this chapter, the experimental arrangement for the femtosecond laser and beam delivery systems
applied in fabrication of FBGs are presented along with the characterization tools and methods.

Chapter 4 (Filament FBG Processing and Characterization)

This chapter presents the procedures for shaping and aligning of light filaments into optical fiber
that bypass the astigmatic focusing of the cylindrical fiber shape. Two methods focus on the
formation of filament arrays with low and high contrast in RI, the latter representing formation of
nano-capillaries. The methodology for optimizing laser pulse energy, focusing geometry, and the
grating design are presented. The morphology of the low and high contrast filaments are presented
under optical and electron microscopy. Procedures for obtaining optical spectral responses of the
FBGs are presented together with representative examples.

Chapter 5 and 6(Filament Based FBG Devices)

In these two chapter, the procedures from Chapter 4 were optimized for design, fabrication and
characterization of exemplary in-fiber sensors and devices in to the four different application
directions. The results were divided in two sections, staring with low refractive-index contrast
filaments, before addressing the high contrast filaments. In the first chapter (Section 5), low

12
contrasts filaments (n = 10-3) have been tailored for in-fiber spectroscopy (Section 5.1 ) and
displacement sensing (Section 5.2). The formation of high contrast hollow filaments (n = 0.4)
are presented in chapter 6 and target applications in refractive index sensing (Section 6.1) and
polarization filtering (Section 6.2).

Chapter 7 (Conclusion and Future Works)

This chapter summarizes the outcomes of the thesis and points to future research paths that have
opened from the results of this study.

13
 Background

This chapter starts with a brief description on the ultrafast laser interaction physics dominating in
the modification of transparent materials (Section 2.1). The focus is on how ultrafast laser pulses
interact with the material in the regime of nonlinear optics that can permanently modify the
properties of the material. General types of material modification in bulk glasses and optical fibers
will be described and then expanded into the case of lasers forming an extended depth-of-focus
such as by laser filamentation and the Bessel beam (Section 2.1.1).

The second part of the chapter (Section 2.2) focusses on the laser interactions that are typically
applied to optical fibers for generating FBGs. Different laser inscription methods (Section 2.2.2
and 2.2.3) have been reviewed and the state-of-art in FBG devices have been summarized.

2.1 Ultrafast Laser Micro/nano Machining in Transparent

Materials
The current fabrication method of integrated photonics devices is mostly based on the existing
semiconductor fabrication methods which are limited to 2D structuring of planar waveguides
[106]. These semiconductor lithography fabrication techniques are expensive and require
cleanroom facilities. On the other hand, femtosecond direct laser writing (FDLW) provides an
alternative direction for 3D structuring inside the transparent material without the need for a
cleanroom environment [107,108]. The fast and single-step fabrication of optical components is a
serial writing method that can dramatically reduce the fabrication costs. Transparent materials such
as glasses are well suited to be a host platform for fabrication of integrated photonic components
due to their broadband visible and telecommunication bands transparency, low absorption, high-
temperature resistance, stiffness, and low cost.

Use of nonlinear effects to confine the interaction into the focal volume deposit a controlled
amount of laser energy into a small volume. Therefore, focusing of ultrafast laser pulses inside
transparent materials permits fine internal 3D processing of the material while also controlling the
size of the heat-affected zone (HAZ). As an example, Figure 2.1 shows inscribing the internal

14
volume of transparent material with ultrashort pulses in contrast with interaction with a CW
(continuous wave) or long pulsed laser. In the case of long pulses, the HAZ can build up to extend
far outside of the size of the focused beam and result in large thermal gradients and other stressing
effects that generate microcracks or initiate ablation of the surface [109].

In the short-pulsed case (Figure 2.1(b)), the sub-diffraction limit modified region allows fine micro
or nano-scale structuring deep in the material while the surface remains intact. Therefore, the
internal modification of transparent materials by the short-pulsed laser presented in Figure 2.1(b)
has presented a noncontact, high-resolution method of 3D internal structuring that has facilitated
the mass production of components to meet the growth and increased demand for the inscription
of novel optical components.

As an example, the physical and chemical interactions induced by such lasers have further allowed
development and fabrication of various 3D photonics components such as waveguides [110],
Bragg gratings [111], resonators [112,113], modulators [114,115] and couplers [116,117] inside
of bulk glass and fiber. For the present thesis, the focus of application has centered on FBG devices
as described in Section 2.2.

(b)
(a)

Figure 2.1: Schematic of laser micromachining in transparent material for two

different pulse durations and their effects on the HAZ. (a) Long pulsed or CW laser
beams would ablate the surface and generate a large HAZ. (b) Ultrashort laser pulses
could otherwise modify internal volume of glass without the damaging the glass
surface while the HAZ may be reduced to below the size of the focused beam .
Image is reproduced, with permission from ref. [109]. © 2014 MDPI.

15
2.1.1 Material Interaction with Ultrafast Lasers
This section briefly describes the physics behind material modification with ultrashort laser pulses.
A comprehensive review on this topic that assesses different regimes of ultrafast laser-glass
interaction are written by Stoian [118], Ams et al. [119], and Mao et al. [120].

Transparent dielectric materials like glasses have a large bandgap energy in the order of 8-9 eV
which is much larger than the photon energy in the visible and telecommunication spectra
(0.7 – 3 eV). As briefly mentioned in the previous section, the transparency and low absorption of
the glasses at these wavelengths make these materials highly desirable for optical applications.
Ultrashort pulsed lasers at high intensity focusing then directly excite electron across the wide
bandgap by multi-photon absorption, leading to permanent changes in the local band structure.

The energy of ultrashort laser pulses can be controllably deposited into the material through the
excitation of bound electrons from the valence to the conduction band. This process is called
photoionization and is responsible for changing the properties of dielectric materials. The
photoionization is described by the electronic band theory of solids. The following processes are
key pathways usually expected for absorption: multiphoton, tunnel, and collisional ionization as
shown in Figure 2.2.

Multiphoton ionization (MPI) occurs in the presence of a sufficiently strong laser field. The laser
field induces a nonlinear absorption where several photons can be absorbed simultaneously to

Figure 2.2: Schematic diagram for different nonlinear laser processes for light absorption in
transparent glass: (a) Multi-photon ionization. (b) Tunnel ionization. (c) Collisional ionization.
Images are reproduced, with permission from ref. [118]. © 2020 Springer.

16
provide enough energy for the electron to be excited to the conduction band (Figure 2.2(a)).
Alternatively, tunnel ionization (TI) is a parallel process that generally will dominate over MPI at
larger laser field intensities and longer wavelengths. In TI ionization process, the laser field bends
the potential barrier of an electron in an atom and enables tunnelling through the Coulomb barrier
(Figure 2.2(b)). In the collisional ionization (CI), the electrons that were excited into the
conduction band by MPI and TI receive additional energy directly by Coulomb force from the
oscillating field that is sufficient to excite the lower band electrons to the conduction band through
high energy collisions (Figure 2.2(c)) [121].

The above-mentioned processes collectively lead to formation of an electron plasma in the solid
which induce time-varying interaction dynamics by temporary modification of the absorption
coefficient and refractive index of the material. After a relaxation time, the excited material
undergoes a permanent structural modification when the laser energy exceeds the modification
threshold [122,123]. For the case of fused silica, these modifications have been classified into three
separate categories. In the Type I category, the laser energy is barely above the material threshold
which causes melting and re-solidification of material, and leads to permanent isotropic refractive
index change in the material [124]. Type II modification with a slightly larger laser energy results
in the further formation of nano-gratings structures in the glass that assembles over many exposure
pulses [125,126]. In this latter case, the material undergoes a refractive index change with
birefringence. Finally, type III modification occurs at much higher energies where micro-
explosions are reported to form voids and hollow structures deep inside the material [127]. In
addition to the laser energy, the final size of the voids and the modification zone also depend on
the other laser parameters such as the laser wavelength, repetition rate, pulse duration, and
scanning speed [118].

In this thesis, the laser pulse energy and the focusing geometry have been optimized in such a way
to generate modification zones in glasses with a near-uniform filament shape beam. As a single
exposure, these represent type I and type III modifications. Filaments with Low contrast RI have
been applied in this study to generate FBGs for applications in displacement sensing and all-fiber
spectroscopy (section 5.1 and 5.2 , respectively) [100,101]. Moreover, larger values of laser pulse
energy were applied to micro-explode open hollow filaments (type III). FBG devices based on
these nano-capillaries enabled fabrication of in-fiber refractive index sensors and in-fiber
polarization filtering devices (section 6.1 and 6.2) [104,105].

17
2.1.2 Filament formation by Ultrashort Laser Pulses
In this section, various methods of filament formation of laser pulses in the transparent glass are
briefly reviewed. Long zones of laser modification with cylindrical symmetry, called filaments,
have been inscribed by ultrashort laser pulses through various effects such as Kerr lensing, Axicon
optics, or surface aberration [31,118,128].

In the first method, an appropriate energy level can balance self focusing by Kerr effect against
defocusing by laser plasma generation and forms a stable zone of filament modification with long
and narrow geometry extending much longer than the confocal beam parameter. Kerr lensing is a
third-order NL interaction with polarization described by the third order susceptibility, χ(3). With
high-intensity Gaussian laser beam shape, the Kerr effect leads to self focusing due to the induced
nonlinear refractive index of the material. The following equation describes the dependence of the
material’s refractive index on laser intensity:

n(I) = n0 + n2 I, (2-1)

where n(I) is the refractive index, n0 is the linear refractive index, n2 is the nonlinear refractive
index coefficient, and I is the intensity. The nonlinear refractive index is negligible for CW or low
energy pulses, but for high intensity pulses will play the key role in the self focusing.

Due to the Gaussian intensity profile of a short laser pulse, the spatial center of the pulse faces
higher refractive index passing through crystal or glass. Therefore, the spatial varying refractive
index profile of the material acts as a lens to focus the ultrashort beam. On the other hand, the
electron plasma formed by various ionization processes as described in Section 2.1.1 would cause
the defocusing of the laser pulse owing to the lowered refractive index of such media. Under high
intensity, the laser beam can be driven through a series of Kerr lens focusing and plasma
defocusing events which lead to a non-uniform filament track in the material. Alternatively,
experimental conditions can permit a uniform filament focal line to form as depicted in Figure
2.3(a).

In the second method (Figure 2.3(b)), high aspect ratio filaments have been formed with axicon
optics to generate non diffracting Bessel beams [128-130]. Inducing conical phase fronts on
Gaussian beams results in the formation of Gaussian-Bessel beams. The phase front is modified

18
by using axicon lenses or spatial light modulators. As an example, Figure 2.3(b) shows the
formation of nondiffracting beams over a certain propagation length (Zb).

The third method of forming long and uniform filaments is based on harnessing the surface
aberration of a glass plate. Due to the refractive index mismatch at the glass surface, different rays
of a tightly focused beam would enter the second medium with different refraction angles which
are depicted in Figure 2.3(c). This would result in the elongation of the focal spot of a Gaussian
beam compared to focusing into the air [131]. Therefore, the focal stretching spreads the laser
energy into a reservoir, which inhibits the traditional beam distortions by Kerr lensing and plasma
defocusing, while retaining the benefits of nonlinear optical modification feeding into a sub
diffraction limited spot size extending over long lengths.

In this thesis, surface aberration of a glass plate has been used for fabrication of various filament-
based FBGs. More details of the method are described in section 4.1.

(a)

(c)
(b)

Figure 2.3: Different optical means for the formation of filaments in glass such as (a) Kerr
lensing, (b) Bessel beams, and (c) aberration generated at an interface, also called surface
aberration. Images are reproduced, with permission from ref. [118,128,131]. © 2020 Springer,
and © 2020 MDPI, and © 2014 Optica.

19
2.1.3 Glass and Fiber Modification with Filaments
The unique cylindrical geometry of the laser-formed filament tracks open doors for various
micromachining applications inside bulk glasses and optical fibers. Ahmed et al. used arrays of
laser filaments for cleaving display [132] and Li et al. [133] applied filament tracks for non-contact
scribing of optical glasses. 2D and 3D processing of glasses with filaments further allow the
fabrication of volume Bragg gratings (VBGs). Mikutis et al. [134] used Gauss-Bessel laser beams
to form a high-efficiency VBG inside of fused silica glass plate with a 2D array of filaments
(Figure 2.4(a)).

In another frontier, Ahmed et al. harnessed filaments for inscribing FBGs in SMF fiber for strain
sensing [135]. Ertorer et al. [31] used an improved fabrication method for inscribing filaments
inside of fiber while increasing filament uniformity. In their method, a combination of index
matching oil and a glass plate was used to form long and uniform filaments in the fiber core (Figure
2.4(b)). Further details on this method are provided in Section 4 .

(a)

(b)

Figure 2.4: (a) Fabrication of filaments in two-dimensional arrays, defining various types of
volume Bragg grating structures in bulk glass. (b) Tightly packing of filaments in fiber core
further allows fabrication of FBGs with unique radiation and reflection properties. Images are
reproduced, with permission from ref. [134,31]. © 2013, and 2018 Optica.

20
2.2 Laser Fabrication of Fiber Bragg Gratings
2.2.1 FBG Theory
Fiber Bragg gratings (FBGs) are optical devices where the refractive index of the waveguiding core
of the fiber have been periodic or aperiodic modulated. This refractive index modulation induces a
wavelength-dependent reflection, transmission, and radiation for the guiding light in the fiber core.
Figure 2.5 shows the working principle of FBGs where spectrally broadband light is injected into
the fiber but only a narrow band of the spectrum is reflected from the Bragg array [136]. The
reflected light appears only near the Bragg resonance wavelength , B , defined by:

B = 2 ne Ʌ. (2-2)

Here, ne is effective index of the mode propagating in the fiber core and Ʌ is the grating pitch
(Figure 2.5).

The unique opportunity of fabricating FBGs with different bandwidths and central Bragg
resonances have allowed researches to use these devices in different applications such as fiber laser
mirrors [22], wavelength-division multiplexers [23], and dispersion compensators [25].
Furthermore, the wavelength dependency of the Bragg resonance on the environmental variables

Figure 2.5: Shematic diagram of FBGs’ working principle. Grating pitch, Ʌ,

determines the reflected wavelength appearing in reflection spectra or as a notch in the
transmission spectrum transmission spectra. Image is reproduced, with permission from
ref. [136]. © 2017 MDPI.

21
such as temperature, vibration, and strain made FBGs a great device for various thermo-mechanical
sensing applications [4,11,25].

2.2.2 UV and Interferometric Fabrication of FBGs

During the last three decades, FBGs have been fabricated by different techniques based on
inducing periodic modulation of the refractive index by light interactions in the material. These
methods are separated into two main categories of interferometric and direct writing techniques.

Interference-based fabrication techniques such as phase masks are well developed today for large-
scale FBG production, but typically require time-consuming exposures with lasers and are further
limited in flexibility for tailoring or tuning of spectral features. Interference-based FBG writing
with UV light can generally be divided into two groups.

In the first one, the UV beam is split into two beams by a beam splitter or a grating (Figure 2.6(a)).
The beams are spatially overlapped inside of the fiber core. The spatial interference pattern
forming inside the core of photosensitive fibers is used to induce the permanent change in
refractive index while the grating period (Ʌ) is controlled externally by the angle formed between
the two beams [137,138].

The second method was introduced by Hill et al. where a phase mask was used to directly form
the interference patterns in the fiber core (as shown in Figure 2.6(b)). In this transverse writing

(a) (b)

Figure 2.6: Interference-based inscription techniques would form spatial interference patterns inside the
fiber core. These methods are divided into two main designs: (a) interference of two beam of light or (b)
different orders of a phase mask . Images are reproduced, with permission from ref. [138,30]. © 2017, and
2011 Optica.

22
method the fine tuning of the central Bragg resonance is achieved by applying different strains to
the fiber during fabrication [30].

2.2.3 Direct Writing of FBGs

Direct writing of FBGs [27,28] was first demonstrated with excimer lasers (such as KrF) due to
challenges of interferometric writing techniques. In this method, the laser beam was tightly focused
inside the fiber using an objective lens with high NA and was scanned along the fiber to
periodically modulate the index of refraction in the core (Figure 2.7). The direct writing of FBGs
allowed significant control on the location of each grating element. Therefore, FBGs with various
designs such as chirped, apodized and π-shifts could be inscribed to tailor the radiation,
transmission and reflection properties of FBGs [139]. On the other hand, the direct writing
technique has challenges such as slower processing time comparing to interferometric FBG writing
methods. On the other hand, interferometric techniques such as phase-mask offer higher finesse
gratings with lower insertion loss.

Figure 2.7: Schematic diagrams of different femtosecond direct laser

writing geometries of creating FBGs such as point-by-point (PbP), line-
by-line (LbL), and plane-by-plane (Pl-b-pl). Image is reproduced, with
permission from ref. [141]. © 2020 IEEE.

23
Direct writing methods have been extended to femtosecond laser exposure by Martinez et al. [28]
to take advantage of a more uniform material photosensitivity to ultrashort duration laser light, and
thus provide uniform core/cladding refractive index modification without a further need of
enhancing responses by hydrogen loading.

The geometry of femtosecond direct laser writing has been an important research direction that
encompasses point-by-point (PbP) [139], plane-by-plane (Pl-b-Pl) [140,141], and line-by-line
(LbL) writing, [29,31] representing a varied means for tuning the shape of the grating elements to
inhibit or enhance radiative and cladding mode effects. Figure 2.7 shows these different writing
techniques with various refractive index modification geometries in the fiber core.

24
 Experimental Setup of Femtosecond Laser
Fabrication System and FBG
Characterization
This chapter introduces the experimental setup which was used for ultrafast laser micromachining
in fiber and bulk glass. The setup includes a femtosecond laser (Section 3.1) with beam delivery
(Section 3.1.1), and motion and alignment systems (Section 3.1.2). A simplified schematic of the
femtosecond laser fabrication setup is shown in Figure 3.1. The chapter also presents the optical
fiber preparation (Section 3.2) and FBG characterization methods (Section 3.3).

3.1 Femtosecond Laser System

A frequency-doubled Yb-doped fiber laser (Amplitude Systems, Satsuma) provided femtosecond
pulses of 515 nm wavelength, 250 fs pulse duration and M² < 1.2 beam quality. The frequency was
set at 800 kHz repetition rate and could be down counted to values as low as 1 Hz. The filament-
by-filament writing in this thesis was intrinsically a fast-writing processing for generating FBGs.
Each filament-shaped phase element was formed with a single laser pulse. In this way, the FBG
writing defined a single step process that could layout a full grating in time scales of seconds to

To CCD

Dichroic
Mirror
Yb-doped Fiber Laser
 = 515 nm
t  250 fs
Fabrication Lens
NA = 0.55
Beam
Expander
Optical Fiber

5-Axis Translational
Stage

Figure 3.1: Schematic of experimental setup for point-by-point fabrication of

FBGs in single-mode optical fiber by ultrashort laser pulses [142].

25
sub-second exposure times with only modest repetition rates of ~1 to 5 kHz as typical values
applied in this work. Pulse energies were varied up to 8 µJ to manage the degree of refractive index
change, to initiate micro explosion of open capillaries, or to accommodate changes in filament
length. Laser processes were often developed first in bulk glass (fused silica, Nikon, NIFS-S (S-
grade)) and then transferred to various types of single-mode optical fiber.

3.1.1 Beam Delivery

Figure 3.2 shows photographs of the ultrafast laser fabrication setup where a variety of mirrors,
lenses, and irises were used to control the beam size and direction, and forming the beam delivery
system. The green line shows the laser path in the system. The output beam from the laser
(Figure 3.2(b)) was expanded ~2 times to a spot radius of 5 mm (1/e2 intensity) using a
commercial beam expander (Linos Magnification Beam Expander, 4401-257-000-20). A
vertically aligned bread board served as the beam delivery platform (Figure 3.2(a)) for various
beam shaping and characterization purposes, before pivoted vertically down to the sample mount
and translation stages. The beam delivery system was placed on a granite arch and a floating optical

(a) To CCD
(c)

CCD
Camera

Z-axis
Translational Stage

Aspheric
Lens

(b)

Figure 3.2: Photos of the experimental femtosecond laser fabrication setup. (a) Laser beam delivery
system containing mirrors, lenses, irises and translational stages (b) Femtosecond laser system with a
beam expander (c) Back reflection alignment diagnostic setup including a CCD camera and imaging
lens. Green lines depict the laser path.

26
table to increase the stability and improve reproducibility of the experiments. An aspherical
focusing lens (0.55 NA, 4.5 mm effective focal length, New Focus, 5722-A-H) was positioned at
the end of the beam line and used to focus the beam transversely into the core and cladding zone
of fibers or bulk glass. A spot radius of 0.4 µm (1/e2 intensity) with an expected depth of the focus
of 1.4 μm was calculated to form in the silica glass.

A CCD camera (Sony, XCD-x710) was used to look at the small back reflection light from the
sample surface to assist with aligning and monitoring the focusing geometry (Figure 3.2(c)).

3.1.2 Motion and Alignment System

The fiber alignment procedure was adapted from Ertorer et al. [31]. All fiber types tested had
similar cladding diameters of ~125 µm and were mounted in parallel alignment in a custom-made
plate with alignment grooves, and sandwiched from the top by a glass plate. The thickness of the
glass plate was varied from 1 to 3 mm to introduced aberration filament length (see Section 4.1).
Index matching oil (Cargille, 50350) was applied to fill the gap between the fiber and two plates
to remove the astigmatic aberration caused by the cylindrical-shaped fiber. The stretched laser
beam forming into thin filaments were then aligned laterally to ±0.5 µm precision to bisect the
waveguide core and also shifted vertically to illuminate any portion of core and cladding cross-
section. Full control on the position of laser-formed filaments was provided by a three-axis motion
and alignment system (Aerotech Inc., Aerotech-PlanarDL-00 XY and ANT130-060-L Z) shown
in Figure 3.2(a). The linear stages provided a 3 nm resolution (minimum incremental motion) and
±100 nm bidirectional repeatability. The laser timing, stages, and fiber position were controlled by
G-code simulator provided by Aerotech Inc (NView program, HMI V2.21).

The speed of fiber motion stage was controlled in combination with down counting of the laser
repetition rate to inscribe FBGs to vary grating pitches over ranges of 364 to 1072 nm that defined
1st order to 4th order Bragg grating resonances across the visible and telecommunication bands. In
this way, the Bragg resonance wavelength, 𝐵 , was tuned by the pulse repetition frequency, 𝑓, and
motion stage speed, 𝑣, as follow [139]:

2 𝑛𝑒 𝑣
𝐵 = 𝑚𝑓
. 3-1)

27
Here, 𝑚 is the grating order, and ne is effective refractive index. As an example, the fiber scanned
at ∼0.5 and ∼1 mm/s speed formed first and second order filament gratings (centered at
telecommunication band, 𝐵 = 1500 nm), respectively, while the laser repetition rate was down
counted to 1 kHz. Second order FBGs having 1000 phase elements (Ʌ = 1.072 µm) were inscribed
in just a few seconds. For the FBG fabrication, grating periods of 364 to 1072 nm were typically
studied, using speeds in the 0.3 to 1 mm/sec range.

3.2 Fiber Preparation

The polymer jacket of the fibers where mechanically striped (Thorlabs, FTS4) in the vicinity of
the grating writing area before laser processing. The stripping was followed by wiping fibers with
optical tissues (Thorlabs, MC-5), soaked in isopropanol or acetone to remove any remaining debris
or contamination. After each experiment, the fibers were also cleaned with acetone or isopropanol
to remove the index matching oil which was used for removing the cylindrical aberration during
laser fabrication and characterization. In the case where FBGs consisted of opened capillaries, the
FBG zones were cleaned with tissues using dichloromethane (DCM) and acetone solvents to better
remove the index matching oils and any other contaminants that accumulated inside of the
capillary. In this way, the FBG capillaries returned to pristine air-filled nanoholes to receive the
next liquid, for example, in studying refractive index sensing (Section 6.1).

Bare single-mode fibers (SMF405-XP, SMF-28) containing FBGs were cleaved (FITEL, S326)
and then fusion-spliced (FITEL, S179A) from both ends to FC/PC fiber optic patch cables before
analyzing the optical properties with optical spectrum analyzer (OSA). The splicing losses were
held below ~0.05 dB.

To prepare fibers for transverse and cross-sectional imaging by optical and scanning electron
microscopy (SEM), the cleaved facet of the fiber was polished with silicon carbide paper (1200
grit), which was wetted with a few drops of distilled water. After cleaning with acetone and
isopropanol, the polished fiber facet was viewed in orthogonal alignment with an scanning electron
microscope (SEM; Hitachi, SU5000), providing high contrast views of the full fiber cross-section.

28
3.3 Optical Fiber Characterization
Various light sources, detectors, spectrum analyzers, microscopes (optical and SEM), and fiber
optic components have been used to characterize the optical properties of filament based FBGs in
the visible and telecommunication spectral bands.

An optical microscopy (Olympus, BX51) has been used to confirm formation of low- and high-
contrast filaments and their relative positioning/uniformity in the fiber core and cladding. To
improve imaging of the internal filament structure, optical fibers were frequently placed under
microscope cover slips (~100 µm thickness) while the gaps between the fiber and upper plate were
filled with index matching liquid to remove astigmatic aberration of the cylindrical fiber cladding
walls. Optical microscopy was also used for confirming and monitoring the wetting, capillary flow
and evaporation of liquids with various values of RI into and out of the laser-formed hollow
filaments.

SEM imaging (Hitachi, SU5000) had been used to verify the formation of isolated nanoholes and
further examine the cross-sectional morphology and uniformity of these structures in the fiber core.

The spectral responses of the laser-formed FBGs were characterized in the visible and
telecommunication bands with the optical probing arrangement shown in Figure 4.5(a).
Reflection and transmission spectra were excited with various broadband (Thorlabs, ASE-FL7002,
1530-1610 nm), and tunable light sources (Photonetics, TUNICS-BT, 1530-1600 nm;
NKT Photonics, SUPERK EXTREME12, SUPERK SELECT, 450-650 nm) covering the
respective bands of visible and telecommunication spectra. The reflected light from the FBG was
relayed through optical fiber circulators (Thorlabs, 6015-3-FC; Ascentta) to record spectra by
high-resolution OSAs (Anritsu, MS9740B; Yokogawa, AQ6373B). FBGs were tested post-
fabrication as well as in real-time during laser fabrication and during chemical etching, or while
filling or evaporating the open-capillary FBGs with various liquids.

External radiation patterns emitted externally through the cladding from second- and fourth-order
FBGs were recorded transversely to the fiber with a beam profiler (Spiricon, SP928) having
3.69 μm pixel pitch. The beam profiles were analyzed by BeamGage Standard software (Ophir
Photonics).

29
 Filament FBG Processing and
Characterization
Different methods of shaping laser filaments in transparent material have been covered in
Section 2.1.2. In this thesis, the elongation of focus by aberration on propagation through fused
silica glass plate of 2-3 mm was used to form long and uniform low- and high-contrast modification
of refractive index filaments. This chapter provides a description of the laser fabrication methods
of filament-based FBGs and the resulting control over the filament morphology and their assembly
into coherent grating structures in the optical fiber (Section 4.1). The grating structures were
further characterized for morphology and optical properties by using optical imaging
(Section 4.2.1) and SEM (Section 4.2.2) tools. The chemical etching of the filaments provided the
unique opportunity of opening the nano holes diameters to tune the photonic bandgap properties
as demonstrated in Section 4.3. Representative examples of FBG spectral recordings are presented
in Section 4.4.

4.1 Filament FBG Processing

Figure 4.1 shows a schematic of the laser focusing arrangement for generating a filament focal
shape in the fiber with an aspherical lens of 0.55 NA. The fabrication methods were adopted from
pervious work by Ertorer et al. [31]. Long and uniform filament beam shape with sub-micron
diameter (<1 µm) was generated by surface aberration induced by the flat surface of a fused silica
plate, positioned between the lens and optical fiber. With sufficient pulse energies, the filament
modification tracks were observed to stretch over 40 to 125 μm length, increased ~30 to ~90 fold
over the 1.4 μm depth of the focus without aberration. The surface aberration distributed the laser
energy over a large depth of focus to form a near-uniform filament shape similar in shape with a
Bessel-like beam formed by an axicon optic [129,130]. In this way, the high pulse energy did not
all converge into a single focus and thus avoided inducing distortion from Kerr lens focusing and
plasma defocusing effects. Such filament generation retained the nonlinear interaction benefits of
narrowing the modification zone to a sub-diffraction-limited size. Each filament modification
structure was formed with a single laser pulse and assembled into arrays with different grating
periods (Ʌ).

30
 Focusing Lens
0.55 NA
Laser Filament
Fs Laser 2 mm Fused
515 nm Silica

Grating

Fiber Core

Index
Matching Oil

Figure 4.1: Schematic of the femtosecond laser focusing arrangement for filament-by-filament FBG
writing into a single-mode fiber. Long and uniform filament shape of light was generated by surface
aberration from pre-focusing the laser beam through the fused silica plate of 2 mm thickness [100].

The laser pulse energy was adjusted upward from the low-contrast conditions generated at the
threshold for single-pulse modification (i.e., 350 nJ to ~2μJ) and presented a progression of
increasing optical contrast of filament tracks that corresponded with increasing refractive index
change. The physical process changed dramatically above a high energy pulse exposure
(i.e. >3 J) that was associated with the onset of nano-explosion, opening hollow-filament shapes
inside of the fiber. Single-pulse exposure of 3 and 7 μJ energy opened such capillaries with lengths
of 80 µm to 125 µm, respectively, with silica plates of 2 and 3 mm thickness, respectively. The
opening of the hollow filaments by this more aggressive process of nano-explosion, entailed
physically moving material out of the exposure zone to leave a nano-hole with uniform diameter
stretching over long lengths. With appropriate alignment, pulse energy ( >4 µJ) and plate thickness
(3 mm), the filament track is positioned to reach fully through the fiber core and cladding as
presented in the next section.

The axial focusing was optimized to avoid the burning or evaporating of the RI-matching oil
positioned between the fiber cladding and upper glass plate and minimize ablative machining and
near-surface damage at the fiber cladding surface. In this way, blind (2 mm plate) or fully open
(3 mm) holes were selectively formed in the fiber core and cladding. In the sections below

31
(Section 4.2), the filament tracks will be verified to have aligned side-by-side to ~1 µm spacing
and remained isolated without breaking through to neighboring tracks, without undergoing beam
propagation distortion or beam breakup into multiple filaments. Filament tracks were assembled
into a tightly packed linear array and precisely centered along with the fiber core for evaluation
for FBG responses. The filament-by-filament writing procedure according to the arrangement in
Figure 4.1 facilitated flexibility in writing uniform, chirped and π-shifted FBGs. The fabrication
method was also extended to writing 2D arrays for various packaging arrangements of photonic
crystal with cursory results being reported elsewhere [142].

32
4.2 Filament Morphology
4.2.1 Optical Microscopy
In this section, optical microscopy has been used as a first level evaluation to confirm the formation
of either low- or high-contrast filaments and their accurate positioning in the fiber core and
cladding. The lateral and axial positioning of the filament tracks in the fiber cross-section was also
verified by front- and back-lighting of the fiber with an optical microscope (Figure 4.2(a,b, and c)).

(a) Cross Section View (b) Side View (c) Top View

Cladding
(125 µm)
Fiber Core

Core (8 µm)
Core (8 µm)

(d) Time sequence of evaporating liquid from hollow filaments.

Figure 4.2: Optical microscopy of the filament array showing the lateral and central alignment of
the array to the fiber core in the cross section (a) and top (c) view. The side view (b) of the fiber
shows formation of filament tracks at various focal depths. (d) Time sequence (left to right) of
evaporating isopropanol with menisci level (orange arrow) moving downward inside views of the
nanoholes recorded over ~10s [31,104].

33
The optical imaging confirms the formation of single filament tracks and their positioning within
the fiber core to sub-micron precision. Such imaging assisted with calibration of the focusing
positions in three dimensions while tuning of the laser pulse energies aided in controlling the
heights and lengths of the individual tracks. Figure 4.2(a) depicts cross-sectional views of a
cleaved fiber where a filament with cylindrical shape bisected the center of the fiber core.
Figure 4.2(b) shows the side view of a fiber where filament tracks (Ʌ = 5 µm) were formed at
different heights by changing the focal depth into the fiber by the vertical motion stage (Figure
3.2(a)). The length of the filament generally increased with increasing pulse energy up to our
maximum exposure (7 J), providing lengths from 20 to 125 µm in the fiber. A top view of a fiber
with a uniform filament array (Ʌ = 1.072 µm) is shown in Figure 4.2(c). The sub-micrometer laser
writing precision allows inscribing tightly packed filament arrays with offset positions set at ±0.5
µm precision with respect to the fiber’s center axis (Section 5.2).

In the case of nanohole formation, optical microscope image and video recordings of the optical
fiber cross section (125 µm diameter) have resolved the nano-hole array formed with 200 nm
diameter and 1072 nm period (Ʌ) in standard optical communication fiber (SMF-28). As an
example, the recordings capture the filling of isopropanol into the nano-hole array (Figure 4.2(d)),
with the meniscus confirming that the nano-holes are fully open from cladding-to-cladding without
showing hole-to-hole break-through. The contrast difference distinguishes opening filament track.
Comprehensive results are presented in Section 6.1.

4.2.2 Scanning Electron Microscopy

In the case of hollow filaments, the cross-sectional morphology of the nanohole structures was
further examined by scanning electron microscope (SEM) imaging. A section of nanohole FBG
was cleaved and polished at an oblique angle (~30), as shown schematically in Figure 4.3.
Electron microscopy of the oblique fiber cross-section identifies an array of nanoholes having
formed through the core and cladding using 4.5 µJ pulse energy, and 2 mm thick aberration plate
(Figure 4.3 (b)). With an increasing magnification of the core section, the sequence of SEM images
(Figure 4.3(b–d)) confirms the formation of isolated nanoholes having a relatively uniform
cylindrical shape, and positioned on Ʌ = 1.072 μm period with an estimated hole diameter of
~200 nm. A slightly irregular hole perimeter is noted with an internal surface roughness of ~50 nm,
which is ~30× smaller than the optical probe wavelength. Hence, the holes can form fully isolated

34
tightly packed arrays having nearly uniform cross-section through the core (Figure 4.3) and most
of the cladding. The nanohole arrays are only noted to break into each at the cladding surface over
a depth of ~15 µm. Hence, nanoholes with a high-aspect ratio >500× have been demonstrated in
the fiber. The present results have not unveiled the high-stress zones expected to be forming around
the nanoholes as seen previously [143,144] in bulk glass when formed with pulses of longer
duration (1 ps). Such densified zones may form a barrier to micro-crack formation to toughen the
glass in the same way chemical treatment stressing works in display glass. The hole diameter is
anticipated to increase with increasing pulse energy as reported by Hoyo et al. [145]. For the
present work, larger diameter capillaries were opened with improved control by chemical etching
(Section 4.3) and presented in Section 6.1. The results of this section were published in ref [104].

Figure 4.3: Schematical view of a nanohole embedded optical fiber with angled facet (a). Under
increasing magnification, SEM images of the angled fiber facet (b–d) unveil open nanoholes having
formed continuously and uniformly (~200 nm diameter) through the core (white oval zone) and
cladding without a breakthrough on the tight 1.072 µm periodic spacing [104].

35
4.3 Chemical Etching
Femtosecond laser machining followed by chemical etching [146-151] was adopted here for
opening the nanohole diameter and tuning the hole diameters and further engineer the stopband
response. Chemical etching extended the diameter of the laser-formed nanoholes in a predictable
and reproducible way. The laser-formed FBGs were immersed in acetone and followed with DCM
to remove oils and debris. Fibers were submerged in 5% dilute HF acid solution for up to 7 mins.
Longer etching times were found to degrade the FBG spectrum. After etching, the fiber was
promptly immersed in distilled water, then IPA and left to air dry.

SEM images of the polished FBG facets revealed a near-linear response of increasing hole
diameter with etching time, beginning from the ~200 nm diameter (Figure 4.4 (a)) for the laser-
formed nanohole and expanding to 350 nm for the case of 4 min etching in 5% HF acid (Figure
4.5(b)). From these data points, the comparison between the experimentally recorded and
simulated FBG stopbands confirmed the control of capillary diameters from 200 nm to values as
high as 700 nm for 7 min etching time, before degradation in the structure was observed. More
details are presented in Section 6.1. The results of this section were published in ref [104].

(a) Laser formed (b) 4 min etched

~ 200 nm ~ 350 nm

1 µm 1 µm

Figure 4.4: The point‐by‐point fabrication was combined with post-chemical etching to engineer
strong photonic stopbands directly inside of the compact and flexible fiber. The hole diameter
expanded from 200 nm (a) to 350 nm (b) after 4 min of chemical etching (5% HF) [104].

36
4.4 Optical Characterization
The spectral responses of FBGs were probed optically with various broadband light sources
(Section 3.3). Reflection and transmission spectra were recorded with a telecommunication/visible
OSA and optical fiber circulators. An example schematic of the characterization setup is shown in
Figure 4.5(a).

After measuring the insertion loss through different ports of the fiber circulator, the loss-corrected
spectral recordings of the fiber device under test was recorded. As an example, Figure 4.5(a) shows
the calibrated transmission and reflection spectra of a uniform first-order low-contrast FBG
(Epulse = 0.4 µJ, Ʌ = 536 nm, 2000 filaments) in SMF 28 fiber. A strong and broad Bragg resonance
dominates the spectra as expected at 1550 nm. The filament “antenna” geometry further enabled a
relatively strong radiation and cladding mode coupling, the latter contributing to an ~7 dB loss
extending to a short wavelength from the ~1 nm Bragg resonance (transmission spectra,
Figure 4.5(a)). Otherwise, a smaller loss of (< 1 dB) seen on the longer wavelength side of the
Bragg resonance (1550–1560 nm) could be attributed to optical scattering (Rayleigh, Mie) from
irregular patterning and non-uniformity in the filament arrays.

(a)

(b) (c)

Figure 4.5: (a) Optical characterization setup for recording reflection and transmission
spectra of different FBGs. Example recorded spectra for uniform (b) and π-shifted (c) FBGs.

37
The filament-by-filament FBG inscription provided a flexibility for fabricating FBGs with
different spectral responses and designs such chirped and phase shifted ones. As an example,
Figure 4.5(c) shows the reflection spectra recorded from a second-order π-shifted hollow filament
FBG (Epulse = 4.5 µJ, Ʌ = 1072 nm, 1200 filaments) in SMF28 fiber. Two narrow transmission dips
are visible in the center of Bragg resonance (Figure 4.5(c) inset) where the laser-induced
birefringence of δλ  210 pm was resolved inside of the larger bandwidth of the Bragg stopband.

38
 Low-Contrast Filament Based FBG Devices
The thesis study was built to drive the design, fabrication, and characterization of a range of novel
FBG devices and sensors based on the uniform low-contrast and high-contrast (hollow) filaments.
In this chapter (5), low-contrast FBGs and their applications in defining two novel optical devices
based around all-fiber spectroscopy (5.1) and motion sensing (5.2) are presented. In the next
chapter (6), high-contrast filament FBGs have been harnessed for enabling in-fiber refractive index
sensing (6.1) and polarization filtering (6.2). A full description of each project area and their results
are presented in each of the sections.

At low pulse energy exposure (~350 nJ), low-contrast filament tracks (n  10-3) arrays were
previously shown to assemble in silica fiber on Ʌ = 0.536 μm period and provide strong first-order
Bragg grating responses at 1550 nm [31]. Further downscaling of the grating period to
Ʌ = 0.364 μm allows recording second-order Bragg grating reflection at the short-wavelength, 532
nm Bragg resonance, centered in the visible spectrum [100].

The unique geometry of the laser filament shape as an optical scattering element is an enhancement
of the first-order radiative emission of fiber-guided light into cladding modes [31]. The opportunity
exploited here is drawn from second-order FBGs which permits strong and directional radiation
modes to escape from the fiber core and cladding that is not supported by traditional designs of
FBGs. The above advantages have been resulted in the design and fabrication of all-fiber
spectrometers described in this chapter in the section 5.1.

Furthermore, the submicron diameter of the laser filaments facilitates a precise off-centered
positioning of filament arrays in the fiber core as well as their azimuthal rotation in the core. This
flexibility permits strain-optic responses for displacement sensing in single-mode fiber with the
possibility for azimuthally resolved sensing. The off-centering accommodates an otherwise zero
photoelastic sensitivity in traditional FBGs. The unique shape of the laser written filaments
resulted in the design and fabrication of opto-mechanical displacement sensors based on off-axis
filament FBGs (section 5.2).

39
5.1 All-fiber Spectrometer
Preface: The filament FBG writing in a single-mode fiber (SMF-28) was first
demonstrated in our group by Dr. Erden Ertorer [31] and extended here for all-fiber
spectrometer. The fiber spectrometer study was a collaborative between myself as PhD
candidate, Prof. Keivan Mahmoud Aghdami as visiting scientist, and Young Hwan Kim as
MSc student.
The Author’s contribution in this research includes jointly conducting experiments in both
design, fabrication, and characterization of the fiber spectrometer and motion sensor with
myself the author as the project main lead. The results of this project have been published
in one journal paper [100] and presented in two conferences [142,152].

Optical grating spectrometers are available with widely varying geometries and sizes to offer low
to high optical resolution and bandwidth that serve in multi-faceted applications such as optical
sensing, characterization, and material analysis [153]. When source etendue is sufficiently small,
one can significantly reduce the physical size of spectrometers, for example, by moving from free-
space to planar lightwave circuits based on Echelle diffraction gratings (EDG) [154], arrayed
waveguide gratings (AWG) [155], or planar holograms [156]. This one-dimensional (1-D)
confinement enables high resolution on-chip spectrometers that are compact, portable and
integrable into larger optical systems.

A further degree of 2D confinement is represented by a 1D volume grating when aligned along a

single-mode waveguide such as an optical fiber. Here, the input slit, collimating lenses, and grating
as found in a conventional spectrometer are laid out along a thin line. Such a highly compact fiber
spectrometer is desirable in a wide base of optical fiber applications in today's telecommunication
networks, data centers, sensing systems, and medical imaging tools. The fiber Bragg grating (FBG)
is key to this application.

The novel mode of femtosecond laser filament focusing was exploited to form a chirped FBG in
single-mode fibers designed for the visible spectrum. The ultrathin filament tracks were extended
fully through the core and near-cladding zone to present a subwavelength scattering antenna
having strongly directed radiation modes to the outside of the cladding. The filament shape
narrowed the azimuthal radiation zones which significantly brightened the spectrum. The
combination of chirping and filament shaping eliminated the need for focusing optics, enabling
the simple all-fiber spectrometer configuration, as shown in Figure 5.1. The fiber spectrometer
provided bright and sharply resolved lines, as narrow as 335 pm, extending over a majority of the

40
visible spectrum (450–650 nm). The blue response marks the state of the art in point-by-point FBG
writing, with a grating period of 364 nm. A highly compact and lens-less spectrometer was thus
presented with total volume of less than 3 mm3. The laser fabrication is extensible to other fiber
types, for broad visible–infrared spectral coverage.

In the first objective to reach the visible spectrum, second- and fourth-order FBGs were tested at
a central wavelength of λ = 532 nm and scaled to short grating periods, reaching Λ = 364 nm.
Uniform period gratings were first optimized for radiation modes satisfying the grating dispersion
equation

mλ = neff Ʌ(𝑠𝑖𝑛 𝜃𝑖 + 𝑠𝑖𝑛 𝜃𝑟 ) (5.1)

Here, m is the diffraction order, Λ is the grating pitch, θi = 90° is the incident angle, θr is internal
diffraction angle, and neff is the effective refractive index of the fiber.
External radiation patterns emitted externally from second- and fourth-order FBGs were readily
observed by eye, as shown in Figure 5.1(b) and Figure 5.2(a). A full spectrum of visible colors

(a) Focused Spectrum

CCD Sensor

Chirped Filament
Grating

Fiber Core
Side view
(b)

Optical Fiber

Figure 5.1: (a) Schematic of all-fiber compact lens-less spectrometer. (b) Side view image of
the FBG (fiber horizontal and blurred) with the radiated spectrum in focus, showing spectrally
narrow and multi-wavelength lines sharply focused by the chirping and filament array geometry
onto a screen. The grating created two near-identical radiation patterns on opposite azimuths of
the fiber, recorded on a CCD sensor left (a) or on a paper screen with a camera (b) [100].

41
was seen emerging in the ±1st, ±2nd, and ±3rd orders (weak third-order lines were not picked up by
the camera) from a fourth-order uniform FBG (Λ = 728 nm) of ≈1 mm length, formed with 1 μJ
pulse energy. The diffraction pattern was projected onto a paper screen wrapped around the fiber
at ≈20 mm radius. The supercontinuum source was filtered into eight wavelength channels (from
450 nm to 635 nm, 1 nm linewidth per channel) and seen as white light scattering from the ≈1 mm
long grating zone of the fiber. All wavelengths have been spatially dispersed and separated into
colors that followed the expected grating dispersion response (Equation 5.1). Figure 5.2(b) shows
the reflection (m = 2) and transmission spectra recorded from a second-order FBG of uniform
period (364 nm) and ≈1 mm length, fabricated with 1 μJ pulse energy. The short grating yielded a
modest reflection strength of −11 dB and ≈300 pm bandwidth (3dB), corresponding to a
moderately strong grating coupling coefficient of КAC = 0.29 mm−1 that aligns with prior
observation in longer-period FBGs [31]. A reflection dip could not be resolved in the transmission
spectrum due to intensity fluctuations in the supercontinuum source.

A key benefit of the filament scattering antenna is shown by the highly narrowed radiation patterns,
emitted symmetrically from both sides of the fiber (Figure 5.2(a)). Each radiation zone subtends
an azimuthal angle of only 1.96° (3 dB), representing a remarkable 90-fold brightening of the
diffracted radiation pattern due to azimuthal narrowing. The uniform grating in Figure 5.2(a)
projected a parallel beam of ≈1 mm width, equal to grating length, following the non-focusing
fiber spectrometer concept in a study by Qin et al. [36]. The grating period in the present work has
been scaled threefold shorter to reach the visible spectrum. Point-by-point writing now further
opens the avenue for chirping of the grating design.

(a) (b)

Figure 5.2: (a) Top view of multicolored radiation patterns projected onto a curved screen from
the fourth-order FBG, revealing the favorable azimuthal narrowing of a single filament. (b) The
reflection and transmission spectra recorded from a second-order FBG of 1 mm length [100].

42
The design of the chirped grating was aimed at forming a sharply resolved spectral line (F) at a
design distance of D = 8.8 mm from fiber cladding, as shown in Figure 5.3(a). Although it was
possible to generate stronger reflection from longer FBGs, a grating length of ≈1 mm was selected
to reduce internal reflection loss and imaging aberrations that arise at higher diffraction angles
from longer devices. The filament array (black dots) was centered in a fiber radius of R = 62.5 μm
with nonperiodic spacing, Λn, chirped to provide focusing at position F. Focusing was obtained by
forcing the optical path length (OPL) difference on radiation paths from two adjacent filaments
(Xn, Xn+1) to a common focal point (F) to equal integer multiples of wavelength, satisfying

𝑂𝑃𝐿𝑋𝑛+1 𝐹 − 𝑂𝑃𝐿𝑋𝑛𝐹 − neff Ʌ = mλ (5.2)

Here, n is the filament number, and the grating order is m = 1. The external, θr, and internal, θ′r,
diffraction angles were related by Snell's law. A chirped FBG design of 3000 filaments and center
wavelength of λ = 532 nm was obtained with a chirping rate of 90 nm mm−1, with the grating period
varying from 364 to 400 nm, as shown in Figure 5.3(b).

Figure 5.3: Chirped FBG showing (a) the filament spacing design for spectral focusing
to point F, (b) with grating pitch plotted versus pitch number, and (c) the resulting
reflection and transmission spectral responses of second-order chirped FBG [100].

43
A chirped grating of this design was fabricated with 1 μJ pulse energy and 1.143 mm length,
yielding the reflection and transmission spectra shown in Figure 5.3(c). The chirping design has
broadened the Bragg transmission and reflection bands, m = 0 and 2, respectively, to
Δλchirp > 100 nm, weakening the second-order reflection to −25 dB in contrast with −11 dB in the
case of the uniform grating in Figure 5.2(a). Figure 5.3(c) shows a grating transmittance of 90%,
suggesting that ≈10% of the fiber-coupled power was redistributed into the targeted first-order
radiation as well as into the cladding mode and scattering losses.

The chirped FGB projected opposing external beams of the ±1st orders into the expected diffraction
angles of θr = 0° and 180°. Radiation patterns are shown in Figure 5.4(a) for 451, 535, and 635 nm
wavelength sources (left to right), as recorded by the beam profiler positioned at the fiber imaging
plane, D = 8.8 mm (Figure 5.3). The geometric narrowing benefit of the long filament design is
also evident, as seen by the short 300 μm height (full width at half maximum) of the radiation
pattern in Figure 5.4(a), spanning ≈2°, as noted in the uniform grating case (Figure 5.2(a)). This
narrowing is approximately two-fold smaller than the diffraction spread expected from the mode-
field diameter (≈4 μm). Such narrowing eliminated the cylindrical focusing lens as required in a
study by Qin et al. [36]. Narrow profile, 1D cameras may thus be used here to reduce device size
and cost.

The spectral line positions (Figure 5.4(a)) aligned with the dispersion equation (Equation 5.1),
providing a moderate dispersion rate of ≈23 μm displacement per 1 nm wavelength. The intensity
profiles were significantly narrowed from 1 mm width in the uniform grating case (Figure 5.2(a))
to linewidths (3 dB) of 7.38, 25.8, and 36.9 μm for the respective 450–635 nm wavelengths,
demonstrating the focusing benefit of the chirped FBG design. The widths approach the charged
coupled device (CCD) pixel size of 3.69 μm.

To assess spectral resolution, intensity profiles on the CCD images (i.e., Figure 5.4(a)) were
integrated transversely to the spectrum and converted by the dispersion equation (Equation 5.1) to
provide the high-resolution spectra (colored lines) shown in Figure 5.4(b). The resulting spectral
linewidths of 0.3, 1.1, and 1.6 nm (left to right) are compared against the directly recorded source
spectra (black line) and found to be source bandwidth limited in resolution. Figure 5.4(c) shows a
wider spectral coverage of the FBG spectrometer with 12 different wavelength sources (450–
635 nm) assembled to cover a majority of the visible spectrum. Noting that the side bands are also

44
present in the light source spectra (Figure 5.4(b)), the fiber spectrometer provided a high signal-
to-noise ratio (> 50). The waveguide and grating dispersion properties avoid wavelength ambiguity
by multi order mixing across the visible spectrum.

A comprehensive assessment of the spectral resolution of the FBG is shown in Figure 5.5, obtained
at various wavelengths provided by the supercontinuum laser and laser diode sources. For the
supercontinuum source, the measured linewidths (blue square) were source bandwidth limited,
increasing monotonically from 0.8 at 465 nm wavelength to 1.58 nm at 635 nm wavelength, that

Figure 5.4: Chirped FBG first-order radiation patterns of blue, green, and red light
sources recorded (a) onto a CCD camera (1 mm × 1 mm frames) and (b) converted
to spectral lines (colored) that reveal a source bandwidth limited resolution (black
line). (c) Assembly of FBG recordings spanning most of the visible spectrum [100].

45
closely followed the source bandwidth (blue triangles). Laser diodes (green triangles) at 450, 520,
and 635 nm wavelength provided narrower linewidth emissions (green triangles) that unveiled the
resolution limit of the fiber spectrometer (green squares). Compared with the theoretical resolving
limit (red line), Δλt = λ/N, for N = 3000 filaments; the chirped grating reached within twofold of
the theoretical resolution, providing 335 pm linewidth or ≈1550 resolving power near the design
wavelength of 532 nm.

Free-space power measurements from the ≈1 mm-long FBG spectrometer revealed ≈1.5%
diffraction efficiency into each of the first orders for a 532 nm source. No other groups have
reported efficiency values for their externally emitting FBG spectrometers [36,157]. Several
options for improving the efficiency may be considered, which begin with strengthening the
grating which showed a high 90% transmittance in Figure 5.3(c). One can ignore the second-order
reflection (−25 dB), which is significantly weakened by the chirping design. Longer FBGs are
straightforward to fabricate, which will require a redesign of the chirping rate and the imaging
distance (D) as well as the consideration of Fresnel or total internal reflection loss and aberration
at the air–cladding interface. Using a higher laser exposure to strengthen the refractive index

Figure 5.5: Full linewidth (3 dB) of the various sources measured by the chirped
FBG spectrometer (squares) and comparison with the source bandwidth as measured
by a high-resolution OSA (triangles) [100].

46
contrast is limited here by the optical focusing resolution limit of aberrated beams at 364 nm
grating pitch.

A further advance of the FBG spectrometer encompasses 2D photonic bandgap engineering with
laser patterning of 2D filament arrays. In this way, one may tailor and concentrate the radiation
modes, providing a higher overall first-order diffraction efficiency. The −1st order grating light
may be redirected to the +1st order, thus doubling the efficiency. More generally, such 2D bandgap
engineering of FBG radiation modes is a new concept, not previously applied to single-mode
optical fibers. The flexibility and precision of femtosecond laser writing of filament arrays in high-
resolution patterns are promising to improve the spectral and geometric diversity of in-fiber optical
devices. The laser fabrication method is extensible to other fiber types and can be further scaled
from the near-ultraviolet spectrum to the infrared telecommunication bands.

In conclusion, a compact all-fiber grating spectrometer has been presented in the visible spectrum
for the first time, enabled by the single-pulse laser writing of long and uniform filament tracks.
The unique chirped filament geometry removed the need for focusing or collimating optics, while
generating favorably narrowed radiation patterns of bright and highly resolved lines of 335 pm
bandwidth. The all-fiber optical spectrometer is promising in both local and remote sensing to
improve diagnostic capabilities at the heart of optical communication networks and data centers
or in medicine, astronomy, and manufacturing that take advantage of the flexible and compact
format of the optical fiber.

47
5.2 Opto-mechanical Motion Sensor
Preface: The filament FBG writing in a single-mode fiber (SMF-28) was first demonstrated
by Dr. Erden Ertorer [31] and extended here for motion sensing. This research direction
entailed collaboration with Mechanical Engineering group lead by Professor Fae Azhari
and her PhD student Hossein Mahlooji, as well as PhD student Gligor Djogo in our group.
The Author’s contribution in this research included jointly conducting experiments in both
design, fabrication, and characterization of the fiber spectrometer and motion sensor with
myself the author as the project main lead. The results of this project have been published
in one journal paper [101] and presented in two conferences [158,159], with myself the first
author.

A reshaping of the FBG phase elements into narrow and long filaments [135] has provided a new
direction for creating ultra-thin planar gratings with femtosecond lasers. Such gratings provide
strong and spectrally narrow Bragg resonances in SMF [31,135], over short lengths and even when
occupying a minimal cross-sectional area of the fiber core. Point-by-point, or in this case filament-
by-filament writing has facilitated a flexible means to improve resolution with π-shifts or with
shifted Bragg resonances for multi-grating systems [31]. This technique opens an opportunity in
displacement sensing, where the photoelastic response can be measured at a single plane, offset
from the neutral axis. Moreover, the geometry favors a highly contrasting azimuthal sensitivity
that reaches null signal when displacement is parallel with the filament lines. Lastly, filament
gratings may be overlaid with different offsets and azimuthal angles. As an example, Figure 5.6
shows one concept of the crossed filament grating array that provides an independent control of
grating period and offsetting (x, y) on two sensing axes to harvest more precise and complete
displacement data based on vector sensing capabilities [53,66,102]. The writing process facilitates
overlaying of addition gratings to improve measurement precision or provide temperature and
axial strain compensation.

In this section, the laser filament writing technique was harnessed for the first time in displacement
sensing, to generate off-axis FBGs in a cantilever-based sensor. The femtosecond laser inscription
process was optimized for generating short length, high reflection gratings, while laterally shifted
to outer zones of the fiber core. Bragg wavelength responses to bent fiber strain were recorded
experimentally and compared with simulated values. These were also analyzed azimuthally to
verify the directional sensitivity of the filament FBGs. Overall, the filament-by-filament writing
offered a facile means for overlaying gratings to calibrate for alignment errors in the fiber

48
 Figure 5.6: Schematic of a cantilever-based FBG displacement
sensor with two orthogonal sets of off-axis filaments
(displacements ξx and ξy) crossed through the fiber core [101].

manufacturing and lock-in the grating offsets during fabrication. In this way, an improved
measurement calibration was obtained that is promising for azimuthally resolved displacement
sensing, while complementing the existing capabilities of temperature and axial strain
compensation [53].

The filament-array gratings, as presented in Figure 5.6, were fabricated following procedures
described previously in chapter 3 and 4. The filaments were assembled inside the core waveguide
into a tightly packed array with a precise offset position (ξx or ξy in Figure 5.6) determined from
the fiber’s center axis with a precision of ±0.5 µm with respect to the cladding surface. A pulse
energy of ∼1 µJ was applied to generate a relatively strong first-order Bragg resonance with
reflection peaks of -1.5 to -2 dB, targeted for a ∼1 mm grating length (ℓ). A Bragg period of
Λ = 536 nm provided a first-order resonance wavelength near λ = 1550 nm. Second and third order
gratings were also fabricated.

Single wavelength FBGs were prepared with varying offsets, from the fiber center (ξx = 0) to the
core-cladding interface (ξx = ∼4.2 µm). The FBG filaments were inspected by optical microscopy
to verify the orientation, periodicity, and the off-axis positioning of the filament gratings with
respect to the fiber core.

49
Figure 5.7(a) shows the FBG spectra recorded from single-row gratings formed with identical
exposures (Epulse = 1 µJ, ℓ = 1 mm) but in planes positioned on different offsets (ξx ≅ 0, 1, 2, 3 µm).
The spectral responses (strength, bandwidth) of these first-order gratings (Λ = 536 nm) aligned
with our prior work, peaking around -1.5 to -2 dB on reflectance for similar exposure conditions
[31,100]. However, the peak reflectance fell off according to -1.4, -1.8, -2.0, and -8.2 dB with
increasing offset of ξx ≅ 0, 1, 2, and 3 µm, respectively. This falloff aligns approximately with the
Gaussian mode field profile (-1.4, -1.8, -3.0 and -5.0 dB) in SMF-28 fiber at 1550 nm wavelength.
The difference between the recorded and calculated reflection values is attributed to the positioning
error (± 0.5 µm) of the filament array in the fiber core.

The ∼1 nm bandwidth (3 dB) of the uniform gratings in Figure 5.7(a) obscures a weak
birefringence of about ∼5×10−5 as determined from a π-shifted filament Bragg grating in previous
work [31]. However, an anticipated birefringence splitting of Δλ ∼ 50 pm was not spectrally
resolvable in the uniform filament gratings as presented in the Figure 5.7(a).

Figure 5.7(b) shows a side view image of a single-row filament grating of 1.072 µm period,
generating a near-uniform contrast of straight modification lines that pass undistorted from the
cladding to fully through the fiber core. Figures 5.7(c,d) present a top view of the fiber for single-
row gratings, demonstrating a high-precision control of offsets (±0.5 µm) with intended
displacements of ξx ≅ -1 µm and ξx ≅ +2 µm, respectively. Here, the end-views of the grating

Figure 5.7: Reflection spectra (a) recorded for uniform FBGs of 2000 filaments (Epulse = 1 µJ,
Λ = 536 nm) at different centering offsets (ξx = 0, 1, 2, 3 µm) show the decay of the reflection
strength with increasing offset. The optical microscope image (b) shows a side view of filaments
stretching more than 40 µm to fully pass through the fiber core. Top view images show single
filament-arrays with offsets ξx = -1 µm (c) and ξx = +2 µm (d) from the fiber axis [101].

50
filaments appear as a straight line of dots, sharply resolved on the 1.072 µm grating period.
However, the physical size of the filament modification cross-section is unresolved and expected
to be under 200 nm based on prior studies [100,104,105]. Moreover, the filaments in first order
FBGs were not optically resolved, but were expected to remain isolated on the Λ = 0.536 µm
grating periods as evidenced by our prior works yielding a strong AC coupling coefficient (ҚAC)
between ∼0.3 to ∼3 mm-1 [31,100].

Furthermore, pairs of gratings with different periodicities (Λx, Λy) and resonant wavelengths were
prepared, overlapping in the same fiber core zone but positioned on different planes. The grating
planes were either aligned parallel or orthogonal (i.e., Figure 5.6) with different offsets of ξx1 and
ξx2, or ξx and ξy respectively.

For displacement sensing measurements, the FBG zone of the fiber was locked into a mounting
fixture (Figure 5.6) to create a cantilever with bending stress concentrated at the FBG by the
support. The center Bragg resonance wavelength of the FBG was assessed (Figure 5.8(a)) to a
wavelength resolution of ±0.5 pm by monitoring the reflection of a broadband light source through
an optical circulator to a high-speed interrogator. Since the source was unpolarized, birefringence
responses in the filament FBGs had become blended into a single peak response. The interrogator
sampled spectra at a frequency of 17 kHz. The fiber displacement was periodically modulated on
a low-frequency (0.33 Hz) square wave profile induced transversely by a solenoid actuator with
V-grove mounting head. A duty cycle of 33% enabled unambiguous tracking of the negative or
positive wavelength shifts corresponding to compressive and tensile stresses, respectively.

For azimuthal positioning, θ, the fiber was aligned with a fiber rotator as shown in Figure 5.8(a),
prior to clamping in a V-groove fiber holder. This alignment was assisted by launching visible
light in the fiber and monitoring the two opposing radiation scattering modes projected
orthogonally from the grating plane [100,142,152]. A telemicroscope recorded visible light
scattering from filaments in the core waveguide, as shown in Figure 5.8(b), permitting a precise
azimuthal positioning (θ) of the filament grating plane with respect to the bending axis within ±1°
tolerance. FBGs of ℓ = 1 mm length were positioned to the edge of the V-groove support using an

51
axial translation stage (not shown), to within ±50 µm longitudinal precision. Fibers were displaced
by up to δ = 2.25 mm on a cantilever length of L = 30 mm (Figure 5.8(c)). This configuration

(a)

Interrogator
(Ibsen Photonics) Circulator
(Thorlabs)
V-groove mount
(Thorlabs)
Broadband source
Fiber tip
(Thorlabs)
displacement Finer Rotator
(Thorlabs)

(b) V-groove
mount
FBG

(e)

(c)

(d)

Figure 5.8: (a) Schematic diagram of the optical characterization arrangement for probing the
FBG reflection responses under cantilever actuation. (b) Close up view of the fiber at the V-groove
clamp, imaging scattering of red waveguiding light from the filaments of the FBG. (c) Schematic
of the fiber cantilever hosting a ℓ = 1 mm long grating located near the support of a L = 30 mm
long cantilevered fiber (d) Cross-sectional view of two filaments arrays representing gratings
FBG1 and FBG2 offset by ξ1 and ξ2, respectively. The fiber was rotated by θ from the horizontal
neutral axis. The waveguide core is shown non-concentric with the theoretical central axis. (e)
Reflection spectra recorded from a first-order FBG (ξx ≅ 2 µm, Λ = 536 nm, ℓ = 1 mm) under bent
(δ = 2.25 mm) and straight conditions, indicating a stress-induced wavelength shift of the Bragg
resonance (inset). Note: Technical drawing of select components in the SolidWorks drawing were
acquired from THORLABS, Inc [101].

52
provided a maximum displacement response and largest Bragg wavelength shift when the filament
array was aligned horizontally (θ = 0°).

Grating offsets of |ξx| ≲ 2 µm were selected for generating a sufficiently strong reflectance (> -10
dB) to provide a high signal to noise ratio and sensitivity on the Bragg wavelength shifts. The
positioning of the grating planes (± 0.5 µm) relative to the cladding were expected to align the
offsets (ξ1, ξ2) closely with the neutral axis of the fiber as indicated in Figure 5.8(d) due to a
homogeneous fiber. However, nonconcentric errors in fiber production (≤ 0.5 µm) were found to
shift the fiber core from the theoretical position as shown (Figure 5.8(d)), presenting a sensor
geometry with slight asymmetric displacement response on azimuthal (θ) fiber rotation.

An example of displacement sensing (Figure 5.8(e)) is presented by the ∼18 pm shift in Bragg
resonance, observed in the first-order reflection spectra recorded with the tunable laser from an
FBG (ξx ≅ 2 µm, Λ = 536 nm, ℓ = 1 mm) under fiber displacements of δ = 0 and δ = 2.5 mm. The
displacement sensing experiments were supported by a 3D opto-mechanical model (ANSYS
Workbench) based on finite element analysis (FEA) of a FBG under varying offsets, ξ x, and ξy,
azimuth angles, θ, and tip displacements, δ.

The FEA also provided the fiber displacement locally along the FBG together with the radius of
curvature. For example, a tip displacement of δ = 1 mm induced a vertical displacement of
1.648 µm and a radius of curvature of 310 mm at the furthest edge of the FBG positioned ℓ = 1
mm from the support. The FEA analysis was performed by PhD candidate Hossein Mahlooji
[101,159].

Figure 5.9(a) presents an example time history of Bragg resonance wavelength recorded for a
cantilever fiber sensor with two parallel gratings, FBG1 and FBG2, fabricated with different Bragg
resonances of λB1 = 1553.96 nm and λB2 = 1549.09 nm, and intended offsets of ξx1 ≅ +2 µm and
ξx2 ≅ -1 µm, respectively (i.e., on opposite sides of the fiber neutral axis as shown in Fig. 4(a)
inset). The gratings provided moderately strong (-1.8 and -2 dB) and narrow (∼1 nm at 3 dB)
reflection peaks as typically observed (i.e., Figure 5.8(e)) for the present laser writing conditions
(Epulse = 1 µJ, Λ = 536 nm, ℓ = 1 mm) [31]. A tip displacement of δ = 2.25 mm applied at the θ =
0° azimuth induced a negative Bragg wavelength shift of ΔλB1 = -23.7 ± 0.5 pm for FBG1
(experiencing compression) and a positive shift of ΔλB2 = +17.6 ± 0.5 pm for FBG2 (undergoing
tension). These values yielded a displacement sensitivity of 9.5 pm/mm and 7.0 pm/mm for FBG1

53
and FBG2, respectively, which correspond to curvature sensitivities of 2.9 pm/m-1 and 2.2 pm/m-
1, respectively. Aside from transient ringing (∼100 ms), the Bragg shifts tracked each other with
a relative amplitude ratio of -1.35:1, which aligns with the intended offset ratio of ξx1/ξx2 = -2 after
accounting for the ±0.5 µm uncertainty in the laser writing positions. Given a high translation
precision in the sample motion stages (Aerotech Inc, Aerotech-PlanarDL-00 XY) relative grating
offsets of ξx1 − ξx2 = 3.00 ± 0.05 µm could be precisely assessed, and thus provide a correction to
the grating offsets values of 𝜉cx1 = +1.70 µm and 𝜉cx2= -1.30 µm to account for the observed -
1.35:1 ratio on the Bragg wavelength shifts. The corrected offsets are given with respect to the
normal axis of the fiber (Figure 5.8(d)) and are independent of the nonconcentric positioning
expected for the fiber core. Similar positional calibrations were applied in all subsequent data sets.

With rotation of the fiber, Figure 5.9(b) shows the azimuthal response of the double grating sensor
for a tip displacement of δ = 2.5 mm observed around a full 2π azimuth. Angles of θ = 0 and 360
are seen to induce maximum compressive and tensile stresses at the top (FBG1, red) and bottom
(FBG2, blue) gratings, respectively, when the filament tracks were horizontal. A rotation to
θ = 180° inverted the peak tension and compression states, leading to the out-of-phase maxima in
the Bragg wavelength shifts. At θ = 90° or 270°, the compressive and tensile stresses along the top
and bottom halves of the filaments were anticipated to cancel out and generate a near zero net shift

Figure 5.9: (a) Temporal response of Bragg resonance wavelengths following two first-order
FBGs with parallel and offset filaments in horizontal position (ξx1 = +2 µm, ξx2 = -1
µm, ΛΛ = 536 nm, ℓ = 1 mm, θ = 0°) and driven under square-wave cantilever agitation (δ = 2.25
mm). Compressive and tensile stresses induced on opposite sides of the neutral axis led opposing
signs in the respective wavelength shifts, ΔλB1 and ΔλB2. (b) Wavelength shifts from the double
grating sensor showing the azimuthal cantilever response for a tip displacement of δ = 2.25 mm
and sinusoidal representations of the data (red, blue lines) [101].

54
in the Bragg wavelength. At other angles, the intermediate values of the Bragg shift were observed
to follow the sinusoidal representations (blue and red lines). However, several discrepancies arise
from mechanical limitations of the present testing apparatus and from imperfections in the fiber
manufacturing, explained below.

In Figure 5.9(b), the magnitude of the out-of-phase peak at 180° fell below the maxima observed
at the 0° and 360° positions by up to ∼29%. The fall was attributed to a residual longitudinal
displacement of the FBG position in the V-groove upon fiber rotation, varying the stress focusing
condition on the FBG. Another potential source of this error is misalignment in the applied load,
which causes a small unintentional torsional moment or a bending moment about the transverse
axis. Also, the sinusoidal curve fit for FBG1 (red line) appears shifted by ∼20° with respect to
FBG2 fit (blue line). Similar offsets were also reported in [44] and may be understood to arise
from imperfect centering of the core waveguide (≤0.5 µm, SMF-28, Corning, NY, USA) relative
to the point of intersection of the neutral axes over different azimuths (see Figure 5.8(d)). The
nonconcentric core thus prevents nulling of the strain response as otherwise expected when
displacements are applied on the 90° and 270° azimuths. These responses, only unveiled by
overlaying two FBG sensors on opposing sides of the neutral axis (Figure 5.8(d) and
Figure 5.9(b)), attest to the sensitivity of off-axis filament FBGs on fiber specifications, which can
in turn be used to detect fiber manufacturing errors such as core-clad concentricity and cladding
non-circularity.

To assess the sensing limits for small cantilever displacement, Bragg wavelength shifts were
recorded over a 180° azimuth and applied in Δθ = 20° increments while reducing the tip
displacements from δ = 2.25 mm to zero in Δδ = 250 µm increments. The results are shown in
Figure 5.10(a). A single filament array was embedded (Epulse = 1 µJ, Λ = 536 nm, ℓ = 1 mm) into
the fiber core with an intended offset of ξx ≅ 2 µm. Linear fits (solid lines) of the wavelength shift
data provided a good representation of the response. Mechanical limitations of the test apparatus
are apparent for tip displacements below δ = 500 µm, corresponding to an estimated radius of
curvature of ∼620 mm at the FBG. This limit in detecting a large bending radius compares
favorably with limits of 25 mm [66] to 100 mm [160] radii as established in related FBG sensors
based on SMF. For tip displacements above δ = 500 µm (i.e., radius < 620 mm), the Bragg shifts
observed for tensile (θ = 90° - 180°) and compressive (θ = 0° - 90°) strain are rising nearly
symmetrically from the null response condition of vertical filament position (θ = 90°). For a given

55
displacement, the Bragg shifts for each angle are well differentiated on the Δθ = 20° increments,
indicating the potential for isolating the azimuthal and displacement responses if a second grating
with orthogonal alignment was also embedded for sensing as shown in Figure 5.7. Moreover,
multiplexing of filament gratings having different offsets at the same fiber location affords wider
latitude for improving precision as well as enabling multi-dimensional temperature-compensated
displacement sensing.

A displacement response, ΔλΒ/δ, was determined from the slopes of best fit lines in Figure 5.10(a)
for each rotation angle, θ, and considering only the data above the mechanical response level of δ
> 500 µm (non-shaded zone in Figure 5.10(a)). The displacement response is presented as a polar
plot in Figure 5.10(b) (blue dots). The data are seen to fall inside of a modelling zone (grey) for
grating offsets varying from ξ = 1.5 to 2.5 µm that is consistent with the ± 0.5 µm positioning error
of FBG placement in the core. The azimuthal dependency is sinusoidal-like as noted previously
for the fixed displacement cases (δ = 2.25 mm) in Figure 5.9(b) and approximately follow the
modelling results (red curve in Figure 5.10(b)) for the intended grating offset of ξx = 2 µm.
However, the data are skewed and show a similar asymmetry (-1.23 ratio) in peak responses of
+12.3 pm/mm and -9.4 pm/mm for filaments positioned above and below the neutral axis,
respectively, for angles oriented at θ = 0° and 180°, respectively. This asymmetry arose from the
same artifact of axial displacement as noted above (Figure 5.9(b)). Mechanical improvements to
the fiber mounting are necessary to ensure that the full stress concentration can be consistently
applied over all fiber azimuths and fill out maximal responses matching with the modelling.

The modelling has assumed a concentric core waveguide, thus yielding null response for filaments
in vertical orientation (θ = 90°). The non-zero response of +0.5 pm/mm detected with the filaments
oriented near vertically (θ = 80 - 100°) arises from a non-concentric core position (Figure 5.8(d))
as was noted for the double FBG in Figure 5.9(b) for the 90°/180° filament positions. These low-
response positions thus could provide a means for calibrating the core concentricity with respect
to the neutral axes of the fiber, providing an estimated core off-centering of ∼0.5 µm for the present
grating response. With further optomechanical modelling, more precise Bragg shifts may be
extracted from the stress fields and resulting refractive index profiles imposed along the filaments.
Such modeling may also account for mechanical limitations in components such as the V-groove
clamp that together promise to improve on the precision of an azimuthally resolved FBG
displacement sensor.

56
Under the present mechanical limitations of fabrication and sensing precision, the filament gratings
have demonstrated a relative strong response to displacements that improves linearly with grating
offset position, while also showing a high sensitivity to azimuthal positioning. These merits are
highlighted in the polar plot of Figure 5.10(c) where full azimuthal angle responses (θ = 0° to 360°)
of the Bragg wavelength shifts were recorded for a fixed tip displacement of δ = 2.25 mm in
Δθ = 10° increments. Identical FBGs (Epulse = 1 µJ, Λ = 536 nm, ℓ = 1 mm) were formed with a
progression of increasing offsets, ξx ≅ 0, ξx ≅ 1, and ξx ≅ 2 µm. The data sets followed the expected
sinusoidal-like response, and improving Bragg shift response. The displacement responses were

Figure 5.10: (a) Bragg wavelength shifts of a first-order FBG (ξx ≅ 2 µm, Λ = 536 nm, ℓ = 1
mm) for different displacements (δ = 0 to 2.25 mm) and azimuths (θ = 0° to 180°). (b) The
displacement response, ΔλΒ/δ, from (a) presented on polar plot (blue dots) for varying rotation
angle (θ) and the corresponding simulation data for a grating offset of ξ = 2 (red) and the
variance (grey) over grating offsets of ξ = 1.5 to 2.5 µm, which is consistent with the ± 0.5
µm laser writing precision. (c) Bragg wavelength shift of FBGs at different offsets (ξx ≅ 0, 1,
2 µm) recorded over a 2π azimuth (green, red, blue dots) for a fixed tip displacement of
δ = 2.25 mm and comparison with sinusoidal responses (solid lines) best representing the data
at corrected offset values, ξsim.

57
closely matched with simulated data (solid lines) for the ideal cases of a concentric waveguide
core with FBG axial positioning at the maximum stressing position on the fiber clamp. However,
to account for the variance in peak amplitudes (i.e., observed ΔλB = 4.4 pm, 14.7 pm, and 32.0 pm
for expected shifts of ΔλB = 0 pm, 12.6 pm, and 25.2 pm, respectively), the grating offsets were
corrected to values of ξx ≅ 0.3 µm, ξx ≅ 1.1 µm, and ξx ≅ 2.5 µm. These corrections fell statistically
within the fabrication precision (± 0.5 µm) of the intended ξx ≅ 0, ξx ≅ 1, and ξx ≅ 2 µm values,
respectively. Also, the FBG wavelength shifts present a high peak-to-min contrast of ∼13, ∼12,
and ∼100 for the respective ξx ≅ 0, ξx ≅ 1, and ξx ≅ 2 µm offsets that points to near-concentric
positioning of the waveguide core here in contrast with the case presented in Figure 5.9(b).

In this section, the benefit of off-axis writing of 1 mm long, first-order filament gratings
(Λ = 536 nm) in SMF-28 was studied for displacement sensing. The main objective in translating
filament FBGs into the field of displacement sensing was to exploit a unique planar sensing
geometry. Unlike other laser fabrication techniques, the filament grating formed within a
submicron thick plane to optically probe within a single layer stress-zone in the fiber cross-section.
In this way, the FBGs could be precisely offset from the fiber center to scale up displacement
responses in SMF and provide strong Bragg reflections that follow the mode field. Furthermore,
the planar filament geometry imposed a highly sensitive azimuthal displacement response that
would permit both azimuthal and displacement characterization when sensing from orthogonal and
overlapping gratings (i.e., Figure 5.6 inset).

The pairing of parallel filament gratings (Figure 5.9) enabled a precise calibration of the grating
positions to ±50 nm, permitting an accurate modeling and calibration of fiber displacement from
Bragg wavelength data based on small wavelength shifts from ∼1 to 25 pm. This sensitivity
extended to evaluate the small imperfections arising in fiber production, such as core eccentricity.
The close sub-micron proximity of the filament pair also naturally compensated for temperature
drifts in the displacement sensing (i.e., Figure 5.9(a)). Such differential stress measurement may
be combined with well-known techniques [53,161] to provide a full temperature compensation to
extend the operating range of the sensor, and improve sensitivity [161] for applications such as in
accelerometers [162].

The cantilever fiber design (Figure 5.6) took advantage of stress concentration at the clamping
support to provide a three-fold increase in displacement sensitivity compared with radial bending

58
of the FBG that is more commonly studied [160]. Despite the non-uniform stress profile along the
cantilevered fiber, the short length of the FBG probes a near-constant stress. Over the 1 mm length,
there was only a 2% stress fall-off as confirmed by the opto-mechanical modelling, falling from
0.4822 MPa at the support to 0.4710 MPa at the FBG end, but decreasing significantly further to
2.2×10−4 MPa at the cantilever end. Inscribing short gratings with high reflection is a benefit of
the filament technique over traditional inscription.

The Bragg resonance wavelength shift monitoring provided more stable and reliable responses
compared to other intensity-based measurements techniques [54,60] which are susceptible to
environmental factors and fiber deformation. Moreover, FBGs in a standard single-mode fiber
eliminated the fabrication and optical sensing complications associated with using multimode and
multicore fibers [40-43]. The detection and tracking of the single Bragg resonance response of a
uniform FBG in a SMF is more practical for scaling up to distributed sensing applications,
comparing with limitations from multi-Bragg resonances in other fiber types such as MMFs [163].
In this way, multi-FBGs have been tracked in parallel in a variety of single-mode waveguide
configurations [53,64,67,102] such as was used here to calibrate and improve sensing precision.
The SMF poses an overall disadvantage of low photoelastic response, limited by the small modal
off-setting available from the neutral axis. Chen et al. [67] have pushed gratings further outside of
the core to gain moderately higher responses, but resulted in low signal-to-noise due to
extraordinarily weak reflection from the exponential mode fall-off. Alternatively, Waltermann et
al. [64] arranged FBGs near to the cladding-core waveguide interface and used bending loss to
infer fiber curvature, but with lower sensitivity than in the present work. Responses were improved
in [53] by core-cladding recoupling, but with significantly higher loss and complexity in signal
analysis that precludes a high-speed peak tracking interrogator as used in the present work. More
complex fiber sensors are otherwise required to provide both high response and low insertion loss,
for example, as shown by Lee et al. [102] in a laser-fabricated fiber circuit with multi-core
waveguides, FBGs, and 3-way directional coupler.

The present filament grating fiber sensor may be readily extended to vibration sensing applications
such as reported in [164], and provide both amplitude and azimuthal resolution with using
multiplexed grating pairs. The further utility of π-shifted filament gratings [31,104] and the
potential polarization sensitivity are promising directions to improve on the spectral resolution as
well as enable biaxial sensing, but at the expense of more complex polarization control. A further

59
advantage of the filament-by-filament writing process was a high writing speed, yielding a net two
second exposure for each FBG (1st order), in contrast with minutes or hours of exposure in the
related studies [165,166]. Finally, such filament FBGs may be distributed over short or long fiber
lengths and permit distributed sensing through wavelength-division multiplexing. The in-core
sensing will permit higher sensor density and spatial acuity in contrast with the optical scattering
limits occurring with the intensity-based FBG techniques [54,60].

60
 High-Contrast Hollow Filament Based FBG
Devices
This chapter presents the applications of high contrast (hollow) FBGs where the laser formed
hollow filament tracks passing fully through the 125 μm diameter of the all-silica fiber cladding.
As previously described in chapter 3, optimizing the focusing and laser exposure allows selectively
opening and positioning of nano-channels centered in the waveguide core. The laser formed
hollow-filament tracks provided high aspect ratio nano-holes (~200 nm diameter) that facilitated
high resolution writing of photonic grating structures with high contrast in refractive index along
the fiber core. The formation of isolated nanoholes was verified by optical microscopy and SEM
(section 4.2).

In this more aggressive laser processing, FBGs with strong second order stop bands have been
successfully fabricated. The capillaries served as nano-fluidic channels for drawing solvents, oils,
or liquid crystals from the surrounding cladding environment directly into the fiber core without
inducing high losses or de-activating waveguiding. This chapter introduces application of high
contrast (hollow) filament FBGs for refractive index sensing (Section 5.1) and for polarization
filtering applications (Section 5.2).

6.1 Opto-fluidic Sensor

Preface: The filament FBG writing in a single-mode fiber (SMF-28) was first
demonstrated by Dr. Erden Ertorer. The method has been further developed for
opening high aspect ratio nano-channels for in-fiber refractive index sensing and
polarization filtering.
The refractive index sensing project was a collaboration between myself as PhD
candidate, Prof Mahmoud Aghdami as visiting scientist. The results of this project
have been published in one journal paper [104] and presented in two conferences
[167].

A schematic representation of the nanohole array formation by laser filament nano-explosion is

depicted in Figure 6.1. The approach meets four key challenges for nano-structuring of strong
photonic stopbands in optical fiber. First, the beam delivery avoids astigmatic aberration by the
cylindrical cladding shape (Figure 6.1), resulting in the formation of nanoholes partially to fully
through the fiber cross-section, without thinning or inducing deleterious damage in the fiber

61
 Figure 6.1: Schematical arrangement of optical aberration plate and optical fiber used for generating
femtosecond laser filaments, and opening high-aspect ratio nanoholes through the fiber cladding and
core waveguide [104].

cladding structure. Second, the nanohole processing facilitates nanohole assembly on small
periodicity at optical wavelength scale, without inducing melting or significant heat-affected zone.
The resulting 0.46 contrast in RI provides strong photonic bandgap responses without high optical
scattering loss. Third, the laser direct writing affords flexible and rapid patterning to tune the
spectral response and provide micro-cavity like responses such as π-shifted FBGs. Chemical
etching facilitates photonic bandgap engineering by tuning the nanohole diameter from 200 nm to
700 nm. And fourth, the nanoholes draw significant capillary force to enable wetting by a wide
range of solvent types fully through the 125 µm diameter of the fiber. The nanohole array thus
defines an open and flexible FBG-sensing structure that can be fabricated rapidly in a single-step
procedure, and provide robust mechanical integrity together with strong optofluidic responses due
to the subwavelength hole diameter.

The validation of strong and responsive photonic stopbands is presented in the reflection spectra
of Figure 6.2(a) for nanohole arrays filled with air and a range of solvents (Table 1). Under high
RI contrast with air (nH = 1.0) or Oil10 (nH= 1.66), the nanohole array provided broad (~2 nm) and

62
 Table 1: Refractive index (RI) of liquids applied inside of nanohole FBGs [104].

strong (~3 dB) Bragg resonances with only 600 holes (i.e., 643 µm length). The evolution of the
spectral profile, reflection peak and linewidth with the increasing grating length for the case of air-
filled holes followed the growth trends of traditional FBGs, expect to develop much more rapidly
owing to a ~100-fold higher RI contrast (Δn = 0.45) over traditional FBGs (Δn ~ 10−3) [168].
Although the 200 nm hole diameter encompassed only a ~2% overlap with the modal field
(MFD ≅ 10.4 µm), the gratings provided a strong effective coupling strength of up to
ҚAC = 9.17 cm−1. The strong responses are enabled by the strong field repulsion or attraction effects
of the guided light into or out of the planar grating zone for the cases of low and high RI,
respectively. As a result, the Bragg resonance had shifted definitively from 1549.93 nm in the air
(nH = 1.0) to 1551.76 nm in Oil10 (nH = 1.661), providing an average RI sensitivity of 2.75 nm/RIU
across a wide RI-sensing range. The progression of strong-to-weak stopbands, from −2.72 dB
reflection with air to −17.28 dB with Oil5 (nH = 1.448), marks the matching condition on RI,
beyond which higher RI (i.e., Oil10, nH = 1.661, Figure 6.2(a)) regenerated strong stopbands
(−2.76 dB).

Although strong in reflection, the breadth of the stopbands under high RI contrast is limiting in
specifying the precise center Bragg wavelength. A more rewarding direction was obtained by
implanting a π-defect into a Bragg grating array of 1200 nanoholes, opening well-resolved
passbands (Figure 6.2(b–e)). In the progression for nanoholes filled with air, water,
dichloromethane and Oil7 (nH = 1.0, 1.316, 1.412, and 1.550 in Figure 6.2(b–e), respectively), the
π-defect provided the highest contrast (>10 dB) and sharpest resolution (δλB = 150 pm) pass-
bandwidth (3 dB) when tuned nearer to the refractive index matching condition, nH ≅ nG

63
(Figure 6.2(d)). For air-filled nanoholes, the resonance is separated into two ~3 dB peaks
(δλB = 210 pm, Figure 6.2(b)), unveiling a waveguide birefringence of ΔnH = 0.07. Otherwise, the
birefringence was unresolved and served only to broaden the passband (δλB = 150–400 pm) with
increasing RI contrast examined in the range 1.31 < nH < 1.67 (Table 1). The point-by-point writing

Figure 6.2: Reflection spectra (a) recorded from a uniform FBG of 600 nanoholes
(Epulse = 4.5 µJ, Ʌ = 1072 nm, 2 mm plate) forming strong (3 dB) and wide (~2 nm) stopbands when
filled with low (nH = 1) or high (nH = 1.661) refractive index materials. The colored bands mark the
3 dB bandwidth. Refractive index matching oil (nH = 1.448) dramatically weakens the stopband.
Reflection spectra (solid lines) of unpolarized light (b–e) were recorded from a π-shifted FBG of
1200 nanoholes for four different examples of RI values (nH by color from legend a) and comparison
with simulated spectra for P-polarized light (dashed lines). Laser-induced birefringence of
δλ = 210 pm is resolved for the air-filled case (b inset). Gaussian line fits of the π-defects (blue and
red lines in b inset) provide ±10 pm spectral precision [104].

64
of such high-aspect ratio nanoholes with <100 nm positional precision thus facilitated photonic
band shaping with sharp spectral features tuned to ~100 pm resolution.

Eigen mode expansion (EME, Lumerical Inc., MODE: waveguide simulator) modelling
(Figure 6.2(b–e), dashed line spectra) provided a good representation of the π-shifted grating
reflection (Figure 6.2(b–e), solid line spectra) with the hole diameter optimized to ~200 ± 10 nm
for the 1200 nanohole array design (Ʌ = 1.072 µm). The simulated spectra have reproduced the
broad stopband and sideband features as well as the narrow π-defect response and accurately
tracked the wavelength shifts for the full range of RI changes from nH = 1.0 (Figure 6.2(b)) to 1.550
(Figure 6.2(e)). However, the grating reflection fell short of the simulated peak values by 3 dB to
7 dB over the respective air (nH = 1.0) to Oil7 (nH = 1.550) cases, pointing to unaccounted losses
from Rayleigh scattering on nanohole surface roughness (~50 nm, Figure 4.3), variances of up to
±50 nm on the nanohole positioning, and first-order Bragg radiation.

To improve on the resolution for RI sensing, Gaussian-shaped functions were found to best match
the π-defect transmission dip, for example, as shown in the case of the air-filled holes
(Figure 6.2(b) inset). The spectral matching enabled center resonant wavelengths to be specified
with a precision of ±10 pm. Otherwise, the Gaussian representations furnished linewidths varying
from 180 to 360 pm (FWHM) for the unresolved birefringence cases (Figure 6.2(c–e)).

The as-formed nanoholes provided added utility in guiding chemical etchant to tune the hole
diameters and further engineer the stopband response. SEM revealed a near-linear response of
increasing hole diameter with etching time, beginning from the ~200 nm diameter for the laser-
formed nanohole and expanding to 350 nm for the case of 4 min etching in 5% HF acid
(Figure 4.4). In real-time monitoring of FBG reflection spectra (1200 holes), the stopbands in both
uniform and π-shifted gratings shifted monotonically to a shorter wavelength over a 7 min etching
time. The blue spectral shifting arises from a drop in effective RI as the lower RI of etchant (nH ≅
nH2O ≅ 1.3164) displaces the higher RI glass (nG = 1.45). Increasing hole diameter is noted to
weaken and narrow the stopbands from −3.7 to −29 dB in peak reflectivity strength and 1.5–
0.97 nm in bandwidth, over the 7 min etching time. However, the π-shifted passband (not plotted)
retained a full 9 dB contrast inside of the FBG stopband, even as the stopband weakened by >
20 dB over the full 7-min etching time. The preservation of sharply resolved π-defect resonances
attests to a highly ordered and unbroken patterning of the high-aspect ratio nanoholes, that does

65
not breakthrough on a tight packing density (1.072 µm period) even as diameters were opened up
over the 7 min etching time.

Nanohole FBGs with uniform gratings were prepared with different etching times (0, 2, 4, and
6 min) to provide a widely varying FBG response when filled with solvents or oils spanning a large
range of RI values (Table 1). The nanohole arrays as formed by the laser (0 min) typically offered
the strongest stopbands (Figure 6.3(a)). The progression to weaker stopbands (Figure 6.3(a–c))
varied from strongly to weakly for solvents having the lowest (Figure 6.3(a), nH = 1), matched
(Figure 6.3(b), nH= 1.448) and highest (Figure 6.3(c), nH = 1.6) values of RI. This progression is
noted in the plot of peak Bragg reflectance (Figure 6.3(d)) over the full range of tested RI values
(Table 1). The fall-off was most pronounced for air (nH = 1), decreasing by >25 dB over the 6 min
etching time. The fall-off was delayed in the case of higher RI oils with RI values of nH = 1.522–
1.66, where reflectivity first increased to a maximum of ~5 dB for 2–4 min etching time. In all
cases, the stopbands were weakest for the 6 min etching time. The bandwidth (3 dB) of the Bragg
stopbands was also strongly influenced by the chemical etching time, either narrowing by ~50%
or broadening by more than threefold according to the negative (nH < 1.45) or positive (nH> 1.45)
contrast of RI (Figure 6.3(e)).

The alignment of the EME-modelled spectra with the observed reflection spectra (Figure 6.2)
provided a precise, semi-empirical estimate of the nanohole diameter, for example, yielding
500 ± 20 nm diameter for 4 min etching time. In this way, hole diameters of 220, 300, 500, and
700 nm were assigned with variances of ~±10 nm to FBGs opened with 0, 2, 4, and 6 min etching
time, respectively. The spectral corroboration identifies a peak value of FBG reflectance
(Figure 6.3(d)) arising on the first quarter wavelength resonance of the nanohole diameter (i.e.,
λ/4nH), for example, encompassing hole diameters of 220–300 nm for solvents varying from nH= 1
for air to nH = 1.41 for dichloromethane. The steep fall-off of reflection thus aligns with an anti-
resonance on doubling of the hole diameter to λ/2nH, corresponding to diameters of 500–700 nm
for air to high index oil (nH = 1 to 1.60). Hence, the diameter of nanoholes transitions from a first-
order resonance for the strongest stop reflection band (0–2 min etching) to the first anti-resonance
(6 min) over the ranges of laser formation and chemical etching tested here.

66
An increasing hole diameter further provided stronger spectral shifts of the stopbands
(Figure 6.3(a–c)), moving in reversed directions as expected depending on the positive (nH > nG)
or negative (nH < nG) contrast of solvent RI with respect to the glass index. The largest hole
diameter (~700 nm for 6 min) and highest RI oils offered the highest wavelength sensitivity, with
the Bragg center wavelength shifting by up to +40 nm for nH= 1.60 solvent over the 0 to 6 min
etching time. In contrast, a smaller and negative wavelength shift of −2.2 nm was noted
(Figure 6.3(a)) for the case of the air-filled holes over the same 6 min etching time. Under index
matching (Figure 6.3(b), nH = 1.45), the Bragg resonance did not shift.

Figure 6.3: Reflection spectra recorded from uniform FBGs (Epulse = 4.5 µJ, Ʌ = 1072 nm, 600 holes) filled
with low (a nH= 1), medium (b nH = 1.44), and high (c nH = 1.661) refractive index materials, revealing the
influence of increasing nanohole diameter due to varying chemical etching time (0–6 min). Peak reflection
strength (d) and 3-dB linewidth (e) of the Bragg resonances plotted as a function of refractive index for
different chemical etching times (following legend a). Bragg resonance wavelength (shifts on right axis)
plotted versus refractive index (f) for different chemical etching times. A global fit of the data by EME
simulation is presented (dashed lines) for hole diameters of 220–700 nm (see legend). The slopes of data
provide a strongly increasing refractive index sensitivity (inset) for increasing RI and increasing hole
diameter (color-coded lines). Shaded zones mark axis scale changes [104].

67
In order to provide the highest resolution RIU sensing, π-shifted FBGs of 1200 hole arrays were
modified with similar etching times (0, 2, 4, and 6 min). EME modelling offered close matching
of the spectra strong bands, side lobes, and π-shift passbands (Figure 6.2(4b–e)), yielding similar
values of effective hole diameters, corresponding to 220, 300, 500, and 700 nm for the respective
0, 2, 4, and 6 min etching times. The π-defect passbands (Figure 6.4) were spectrally fitted to
Gaussian line shapes, varying from 200 to 400 nm linewidth. A narrow birefringent splitting
(~200 pm) of the π-defect was occasionally resolved in the spectra in cases with the strongest
stopbands and largest contrast in the RI.

With spectral line fitting of the π-defects (Figure 6.4), center Bragg wavelengths could be
determined to ±10 pm precision, enabling RI sensing to a high-resolution of 10−5 RIU. When
plotted against the RI (Figure 6.3(f), solid fonts), the π-resonance shifts demonstrated an
impressive RI response of FBG stopbands, shown globally over an extraordinary range of RI
values (nH = 1–1.66) and nanohole diameters (220–700 nm). The EME modelling (Figure 6.3(f),
dashed line) followed each data set to ±200 pm (rms) spectral precision, relying only on one value
of optimized hole diameter across the full RI testing range. Discrepancies in the air (±500 pm) and
between methanol and water may arise from surface tensions effects that require further study.
Sharply forming π-defect resonances were identified in all cases except the two highest indexes
(nH = 1.661 and 1.670) and large diameter (6 min) condition (open triangle Figure 6.3(f)), where
stopbands became overly broad and mixed with sidebands.

The narrow π-shifted stopbands provided strong optical responses to sense the local environment,
as demonstrated in the wide 1548–1590 nm shift in Bragg wavelengths (Figure 6.3(f)). The slopes
of these responses yielded a widely ranging response in RIU sensitivity (Figure 6.3(f), inset) that
reached as high as 600 nm/RIU for the case of highest RI (nH= 1.66) and largest hole diameter
(700 nm). This RIU sensitivity is comparable to the best FBG-based demonstrations to date (i.e.,
945 nm/RIU in ref. [169]). Moreover, spectral line fitting (±10 pm) offered a high-resolution
determination of the RI to ±10−4 precision.

68
The chemical etching of the filaments demonstrated an attractive progression of improving RIU
sensitivity response (Figure 6.3(f), inset), but at the cost of forming broader and weakened
stopbands, as the hole diameter was increased from 200 nm to 700 nm (i.e., see Figure 6.3). This
range of subwavelength hole diameter (i.e., below λ/n = 1.072 µm) was beneficial in circumventing

Figure 6.4: The complete set of the reflection spectra recorded from π-shifted FBGs having 1200
nano-hole filaments (Epulse = 4.5 µJ, Ʌ = 1072 nm) and their widely varying response to liquids with
varying RI values (color coded and as labelled in (a)) after undergoing 0 (a), 2 (b), 4 (c) and 6 (d)
minutes of chemical etching time. The unetched nano-holes having smallest diameter (a) provided the
strongest reflection resonances for a majority of the liquids evaluated, while an increasing hole
diameter (b to d) provided an increasing Bragg wavelength shift for improving refractive index
sensitivity, but with weakening overall reflection strength. The π-shift resonance for laser-formed and
chemically etched holes appears as narrow (50 to 370 pm) transmission windows that sharpened the
sensing resolution of Bragg resonance. However, a birefringent broadening of the π-defect resonance
plays out differently in the spectra according to the refractive index contrast and hole diameter. In the
sequence of increasing hole diameter (a-d) for air filled holes (negative refractive index contrast, n =
-0.45), the π-defect displays a moderately rising birefringence of δλ = 210 to 370 pm for chemical
etching times rising from 0 to 6 min. Birefringence broadening of δλ  20 pm is also strongly evident
in select cases of positive refractive index contrast (i.e., n = 0.1 at 2 and 4 min etch for Oil6 and Oil7
in (b) or (c)). Otherwise, the birefringence is unresolved for a majority of the liquid solvents (1.31 <
nH< 1.67) and etching time (0 – 6 min) presented here [104].

69
optical resonances in the holes that interfere with the photonic stopbands. In this way, strong
stopbands enabled a signal to noise ratio (SNR) amenable for optical sensing over the moderately
large range of reported RI values, 1–1.67. Scaling to a larger hole diameter would otherwise
diminish the SNR and narrow the RI-sensing range, for example, falling sevenfold to the 1.32–
1.41 range as reported in ref. [84] for 1.665 µm hole width. The benefit of the sub-micron hole
dimensions is thus narrower and stronger stopbands that avoid ambiguity in sensing multiple Bragg
resonances in distributed FBG systems. Moreover, the nanohole FBG avoids a low limit cut off in
the RI (i.e., n ~ 1.48) for total internal reflection as in the case of evanescence-based FBG
refractometry [75-77] .

The nanohole array presents a responsive optical fiber sensor, retaining cladding strength while
readily wetting with numerous solvents that can reach into the open photonic structure formed
along with the fiber core. The nanoholes are amenable to hosting a wide range of materials, for
example, such as nematic liquid crystals (section 6.2). The small, 200 nm diameter capillaries were
found to impose a strong axial molecular alignment of NLC, manifesting in strong optical
birefringence in the fiber polarization modes that facilitated an all-fiber dynamically switchable
polarization filter [105]. The facile means of laser writing and patterning of high-aspect-ratio
nanoholes introduced here thus opens the realm for creating strong and compact photonic stopband
devices directly in traditional optical fibers while facilitating environmental sensing through a
thick robust cladding. The line-by-line writing is an intrinsically fast process, being single-step
and scalable to sub-second exposure times with only modest repetition rates of ~1 kHz in
comparison with current fabrication techniques [84,169].

70
6.2 In‐Fiber Switchable Polarization Filter
Preface: The filament FBG writing in a single-mode fiber (SMF-28) was first
demonstrated by Dr. Erden Ertorer. The method have further developed for
opening high aspect ratio nano-channels for in-fiber refractive index sensing and
polarization filtering.
This research direction entails collaborative research with prof. Tigran Galstian
group at University of Laval and his postdoc Dr Tigran Dadalyan. The results of
this project have been published in one journal paper [105] and presented in
multiple conferences [170,171], with myself the first author.

Optical fiber polarizers are available with different forms, materials, and sizes to offer a high
extinction ratio together with low insertion loss, and wide working bandwidth for advantages in
numerous applications. In-fiber polarizers are found in fiber lasers, polarization division
multiplexers [172,173], and fiber sensors [174] for multi-faceted applications in
telecommunication, sensing, and medical fields. Unlike traditional free-space optics, in-fiber
polarizers impose additional challenges on meeting high polarization extinction with low-loss and
alignment constraints limited by a narrow and ultra-compact fiber geometry.

In-fiber polarizer designs are available in three main categories. In the first group, a large part of
the fiber cladding is mechanically or chemically removed and covered by a thin layer of graphene
[175,176], metal [177] or liquid crystals (LC) [178] that serve to evanescently attenuate one
polarization state of the guided fiber mode. In the second category, the fiber core is micro-formed
into a birefringent chiral guide [179,180] that induces a high bending loss in one polarization mode.
In the last category, structured laser irradiation of the core provides a minimally invasive means
of embedding tilted Bragg gratings [181] or long period gratings [182] to diffract away one
polarization mode in a birefringent fiber. Such gratings further enable the spectral tailoring of the
polarization states. In all these approaches, polarization responses are predominantly static.

For this reason, the active polarization discrimination capability of nematic liquid crystals (NLCs)
has been explored for a possible dynamic polarization control in optical fibers. The anisotropic
molecules of NLCs are favored for their ease of natural self-alignment (directed along with a unit
vector n, named director [95]) that results in high optical birefringence, typically on the order of
Δn = n∥ – n⊥ ≅ 0.15 at 25 °C, where n⊥ and n∥ are, respectively, the ordinary and extraordinary
refractive indices. The dynamic control may be obtained by an external stimulus such as electric
or magnetic fields that reorient the birefringence director. The birefringence can also be tuned with

71
temperature, for example, with n∥ decreasing and n⊥ increasing when approaching the phase
transition temperature, e.g., ≈34.5 °C for the 5CB [95].

The embedding of such NLC materials inside the core waveguide of optical fiber has been
promising to create dynamically reconfigurable components and devices [183,184]. However,
there are still some challenges. The evanescent field-based devices have offered a tunable
attenuation, but without polarization control. The filling of the micro-holes in photonic crystal
fiber with NLCs [185,186] enabled a high modal overlap of the guided light with the NLC, but at
the cost of dramatically altering or entirely removing the stopband structure for waveguiding of
the light. Similarly, the presence of appreciable NLC volume in the core of standard silica optical
fiber disturbs the modal properties as the typical values of n∥ ≈1.7 and n⊥ ≈1.5 for bulk NLC
materials are notably higher than the refractive indices of the cladding (ncl) and the core (nc) of
widely deployed single-mode telecommunication fibers. For this reason, the laser structuring of
nano-scale ripples [98] in combination with laser machining of buried microchannels [99] will not
meet the requirement to align NLCs and provide polarization control without disturbing the optical
properties of the fiber waveguide.

Harnessing the NLC birefringence properties directly in the core thus requires a new means for
nano-structuring of the internal volume of the fiber core in a way that does not impose dramatic
changes in the fundamental modal properties or lead to high radiative loss. In addition to inducing
a favorable “geometric” alignment of the NLCs, the underlying nano-structure should also enable
dynamic polarization responses together with the possibility for spectral tailoring of the
polarization control.

One promising direction to reach internally into the core waveguide of an optical fiber is by laser
micro-explosion of nanocapillaries from “filament-shaped” beams. Rather than relying on unstable
Kerr lensing effects, nano-channels with a high aspect ratio have been opened in bulk glass with
femtosecond lasers pre-shaped into an optical Bessel beam [85,88]. Zhang et al. [89] have used
this approach to open an array of nano-holes through a laser written waveguide, providing a Bragg
grating waveguide inside of a bulk glass plate. In the case of focusing on cylindrical-shaped fiber,
our group has harnessed surface aberration of glass plates to create long and uniform laser
filaments that pierce the core waveguide which is described in full detail in chapter 4. This method

72
enabled low-contrast refractive index modification (Chapter 5) or micro-explosion of nano-
capillaries (Chapter 6), which formed into fiber Bragg gratings (FBGs) when assembled in arrays.

To meet objectives for in-fiber polarization control, this section has harnessed the nanocapillary
FBG for drawing NLC into the waveguiding core of the fiber to induce strong polarization
dependent loss (PDL) on the stopbands in the telecommunication band. The nanometric sized
hollow-filament devices provided the present opportunity for infiltration and alignment of NLC
(Figure 6.5(a)) into the waveguiding core without overly influencing the modal or propagation
properties (i.e., effective refractive of the NLC zone, neff) due to the small fractional size of the
capillaries. As a result, low loss in-fiber polarizers having dynamic controllable properties have
been studied here with commercially available NLC materials and standard telecommunication
optical fiber. The anisotropic properties of such NLCs were evaluated by recording the FBG
spectral responses of hollow filament arrays when infiltrated with NLCs. Birefringent splitting of

Figure 6.5: Schematic of an in-fiber Bragg grating for polarization switching based on NLC-filled
nano-filament cavity arrays. At room temperature, NLCs favored homogeneous alignment as
depicted in the cross-sectional view (a) of the nano-holes opened to ≈200 nm diameter by
femtosecond laser nano-explosion. The nano-hole FBGs (b) were loaded with NLC by the
capillarity effect and the stopbands were spectrally tested for thermo-optic and electro-rotation
responses using a thermal heater and capacitive cell. When tuned to the FBG stopband for vertical
polarization (0o, z axis, n∥ = 1.661), only horizontal polarized light (90°, y axis, n⊥ = 1.511)
transmits through the grating [105].

73
the stopbands confirmed a preferential alignment of the NLCs inside of the nano-capillary
(Figure 6.5(a)).

The laser structured fiber was fixed onto the surface of a glass substrate, with the orientation of
the nanoholes aligned either perpendicular or parallel with the substrate surface. A second plate
was glued to the fiber-plate assembly to complete an NLC cell as shown schematically in
Figure 6.5(b) for the parallel alignment configuration. The inner surfaces of substrates were coated
with indium tin oxide (ITO) and were spaced to a 250 µm gap to contain the fiber. This permitted
a variable electric field to be applied transversely to the fiber either parallel or perpendicular with
the nano-capillaries. The above-mentioned NLC cell was positioned on a heat plate. This
configuration (Figure 6.5) thus provides both thermo-optic and electro-rotation probing
capabilities.

The nano-holes were readily wetted and filled with an NLC (for example, 4′-n-pentyl-4-
cyanobiphenyl, or 5CB) when the material was introduced between the cell substrates, being
driven by the capillarity effect. Polarizing microscope studies confirmed that the NLC director was
self-aligned along the filament axis without treatment of the inner surfaces of the nano- holes. This
alignment minimizes the elastic deformation energy. At 25 °C, the refractive indices of bulk NLC's
are expected to be n∥ = 1.661 and n⊥ = 1.511 at 1550 nm wavelength when fully aligned [187].
Lower values of birefringence may arise in the “confined” capillary geometry due to the small
effective volume as well as to alignment imperfections of the NLC.

Figure 6.6(a) shows the experimental arrangement applied for optical characterization of the
polarization dependent spectral responses generated by the nano-hole FBGs before and after
infiltration with NLC. A tunable external cavity laser with 1 pm bandwidth and up to 4 mW output
power provided linearly polarized light to a polarization-maintaining fiber circulator. After the
circulator, the polarization state of the light entering the FBG was controlled with a motorized
fiber polarization controller. The transmission and reflection spectra of the FBG were recorded
with an OSA positioned either following the FBG (as shown in Figure. 6.6(a)) for transmission

74
 Figure 6.6: (a) Schematic arrangement for optical characterization of the reflection and
transmission polarization responses in the LC filled FBG. (b) Transmission spectra
recorded through a second-order FBG of 1000 nano-holes (Λ = 1.072 µm) without LC
filling, comparing the two states of linear polarization aligning parallel (0°) and
perpendicular (90°) with respect to filament orientation. The air-filled capillaries induce a
small birefringent Bragg shift of Δλ = 0.25 nm [105].

or on the fiber circulator (Figure. 6.6(a)) for reflection while sweeping the laser output wavelength
across the telecommunication band (1545–1560 nm).

Figure. 6.6(b) shows the transmission spectra recorded for two orthogonal linear polarization states
from a uniform second-order FBG (Λ = 1072 nm) having 1000 nano-holes with open (air-filled)
nano-holes. Relatively strong (≈15 dB) and narrow (≈1 nm at 3 dB) stopbands have aligned except
for a small birefringence of the capillary shape that shifted the Bragg peaks by Δλ = 250 pm
between the perpendicular (90°, red line) and parallel (0°, blue line) polarization states as defined
with respect to the capillary axis (Figure. 6.5). The filament “antenna” geometry enabled a
relatively strong radiation and cladding mode coupling [31,100], the latter contributing to an ≈8 dB
loss extending to a short wavelength from the ≈1 nm Bragg resonance (Figure. 6.6(b)). Otherwise,
a smaller loss of ≈3 dB seen on the longer wavelength side of the Bragg resonance (1551–1560
nm) can be attributed to optical scattering from irregular patterning and nano-scale surface
roughness in the capillary arrays. These effects are compounded by a high refractive index contrast
of Δn ≅ 0.46 at the air to core-waveguide interface.

The validation of strong and responsive polarization-dependent photonic stopbands is presented in

the transmission and reflection spectra of Figure 6.7(a,b) for the nano-hole array filled with NLC.
Under a more modest contrast (Δn = 0.05) between the refractive index of the core (nc = 1.46) and

75
ordinary refractive index of NLC (n⊥ = 1.51), the nano-hole array provided narrow (≈1 nm) Bragg
resonances for perpendicular polarized probing light (Figure 6.7(a), green band). Relatively
symmetric stopbands and side lobes are noted in both the recorded reflection and transmission
spectra that attest to the highly ordered and uniform filament pitch over the ≈1 mm grating length.

Under higher refractive index contrast (Δn = 0.2) with the core (nc), the extraordinary refractive
index of NLC (n∥ = 1.66) induces stronger (up to ≈20 dB) and broader (≈5 nm) reflection and
transmission stopbands (Figure 6.7(b), amber band) that have shifted sharply to longer wavelength.
This index contrast further contributes to higher coupling and scattering losses that result in a
pronounced asymmetry in the band structure.

When probing with a pure polarization state (parallel or perpendicular), the evolution of the
spectral profile, reflection peak, and bandwidth (Figures 6.6(b) and 6.7(a,b)) have followed closely
with our previous study [104] of isotropic liquids, being highly sensitive to both the magnitude
and the sign of the refractive index contrast. The drawing of the optical field into capillaries with
positive values of refractive index (i.e., NLC in Figure 6.7) generated much stronger and broader
stopbands than for the negative index changes that pushed light out of the capillary (i.e., air-filled
FBG, Figures 6.6(b)) [104]. However, the highly birefringent grating now offers a new means for
dropping a preferred polarization state in two narrow spectral reflection windows (≈5 nm amber
color band for 0° and ≈1.5 nm green band for 90° in Figure 6.7(a,b)), tailored by the combination
of Bragg reflection and polarization dependent NLC.

EME modelling of the transmission and reflection spectra expected from a nano-hole array of the
same experimental design (Λ = 1072 nm, 1000 holes) provided the simulation spectra in
Figure 6.7(d,e) (dotted lines), showing good correspondence with the experimental spectral
recordings (solid lines) reproduced from Figure 6.7(d,e). A nano-hole diameter of 200 nm (±10
nm) provided the best matching of simulated stopband and side-lobe positions to the observed
spectra. The optimized refractive index values of n⊥ = 1.510, and n∥ = 1.651 align closely with the
fully aligned NLC case of n⊥ = 1.511, and n∥ = 1.661 expected for room temperature [187].
Relatively symmetric stopbands and side lobes are noted in both simulation and experimental data
for ordinary refractive index probed with perpendicular (90°) polarized light. When probing with
parallel polarized (0°) light, the extraordinary refractive index of the NLC generated less
symmetric stopbands in both the simulated and recorded spectra.

76
Although there are shortfalls in the overall stopband efficiencies in Figure 6.7(d,e) (i.e., ≈3 dB in
reflection, and ≈20 dB versus 40 dB extinction in transmission), the Bragg shifts and stopband
positions (Figure 6.7(d, e)) collectively point to the strong preferential alignment of NLC by the
nano-sized capillary for room temperature operation. The present simulations also points to a
moderately strong band of cladding mode coupling loss (≈17.5 dB) arising for the highest index
contrast case, n||, (Figure 6.7(d), 1550.2 nm). However, this PDL is much broader in the
experimental spectrum (Figure 6.7(d), solid blue), possibly owing to misalignment errors and
surface roughness of the nano-holes in the ideal case. The strong birefringence of the NLC-filled

Figure 6.7: Reflection and transmission spectra recorded from a uniform FBG (Λ = 1072 nm, 1000 holes)
filled with NLC for (a) perpendicular (90°) and (b) parallel (0°) polarization states. The respective
stopbands have shifted apart according to the perpendicular (a; green) and parallel (b; amber) polarization
alignment with the ordinary (n⊥ = 1.511) and extraordinary (n∥ = 1.661) refractive indices of the NLC. The
spectra (c) of polarization extinction ratios (parallel (a) divided by perpendicular (b)) for transmission
(PERT, green line) and reflection (PERR, orange line) present distinct bands of high extinction of the parallel
(PERT, amber) or perpendicular (green) polarized light, which extends to short wavelengths (PERT, pink)
due to the high cladding mode loss (b) seen for the parallel polarization state. A large component of the
rejected parallel polarized light is reflected, appearing symmetrically, further provides in the
PERT spectrum with a negative value (amber and pink bands). The recorded transmission (d) and reflection
(e) spectra (solid lines) are seen to follow approximately with the stopband and sidelobe positions predicted
in the EME simulated spectra (dashed lines) [105].

77
capillaries thus provides an overly strong enhancement of cladding modes, radiation modes, and
insertion losses for parallel (0°) polarization (Figure 6.7(b), pink band) in comparison with a higher
transparency at shorter wavelengths in the perpendicular (90°) polarization case (Figure 6.7(a), λ
< 1550 nm). This PDL extends too far outside of the Bragg stopband, beyond the 1545 to 1560 nm
window in Figure 6.7(b).

The PDL presents an opportunity for designing polarization attenuators with tailored wavelength
and spectral bandwidth. For example, the presence of NLC (Figure 6.7(a,b)) provides a relatively
high PER of up to 20 dB for transmission spectra (PERT green) and up to 25 dB for reflection
spectra (PERR orange) as shown in Figure 6.7(c). The PERT and PERR were calculated from the
ratio of transmission (Figure 6.7(a)) and reflection (Figure 6.7(b)) spectra, respectively, observed
for parallel to perpendicular polarization states. The sign difference between the PERT and PERR
ratios show the transmitted and reflected light can be highly polarized, while carrying the opposite
polarization states. In transmission, the parallel polarization extinction extends over tens of
nanometers, providing a wide passband (Figure 6.7(c), amber and pink bands) for the
perpendicular polarization except for the 1551 to 1552 nm Bragg stopband (Figure 6.7(c), green
band). In reflection (Figure 6.7(c), orange), the capillary grating preferentially drops the parallel
polarization state in a broad ≈5 nm stopband (amber colored band).

For non-orthogonal polarization states of probing light, the NLC-embedded FBG provided
polarization control as shown in Figure 6.8(a), with transmission components geometrically
projected onto the parallel (0°) and perpendicular (90°) axes with varying angles of 0° to 90° as
defined with respect to the filament axis (Figure 6.5). Figure 6.8(a) shows a systematic broadening
and strengthening of the stop-band as the polarization angles of probing light are tuned from 90°
to 0°. The transmission values for the 1555 nm wavelength, taken at the center of the strong Bragg
stopband (Figure 6.8(a), vertical black line), are seen to follow a near ideal linear polarizer
response (Figure 6.8(b), black squares) derived from Malus’ law. In this FBG stop-band (1545 to
1560 nm wavelength), the PDL was limited to a maximum ≈20 dB extinction ratio. However, the
linear polarization response also extended to outside of the transmission stopband (i.e., λ > 1560
nm, and λ < 1545 nm), offering a control of the polarization state over the Poincare sphere reaching
up to the full birefringence of bulk NLC (Δn ≅ 0.15, at 1550 nm wavelength). The expected phase
retardance, Δφ, arising between the 0° and 90° polarization states yielded an estimated maximum
value of

78
 2𝜋
Δφ = × Δn × ηL × ηMFD × L ≅ 0.76 π
𝜆

for a longitudinal filament duty cycle of ηL ≅ 0.2, a transverse filament-modal overlap

ηMFD ≅ 0.019, a λ = 1600 nm wavelength, and a grating length of L ≅ 1.072 mm. Equation (1)
demonstrates the feasibility to design customized polarization retarders at a desired region of the
spectrum with a low PDL.

Thermo-optic tuning of the stopbands opens a further direction for polarization and spectral tuning
of the NLC-filled FBG as shown in Figure 6.9(a). This response was most pronounced for the
parallel polarization state, presented in the progression of transmission spectra with heating of the
FBG cell from 20 to 35 °C. The strong and broad stopband has weakened by up to 21 dB, shifted
to a shorter ≈1552.5 nm wavelength and narrowed to ≈0.8 nm bandwidth. In contrast, the
transmission spectrum for the perpendicular polarization state had progressed from the room
temperature case in Figure 6.9(a), to align nearly identically with the parallel polarization case
shown at 35 °C (Figure 6.9(a)). Only a weak birefringent Bragg shift of < 50 pm differentiated the
parallel and perpendicular polarization stopbands, attesting the transition of the NLC to the
expected isotropic liquid state with refractive index n|| = n⊥ ≅ 1.56 at the critical temperature (TC)
of ≈35 °C [187]. Hence, a full thermo-optic response has been observed in the NLC-filled FBG.

Figure 6.8: A sequence of (a) transmission spectra recorded through a second-order FBG (1000
nanoholes) filled with NLC for incident linearly polarized light aligned at varying angles of 0° to 90°
with respect to filament orientation. Polar plot of the (b) transmittance measured at 1555 nm wavelength
from (a) (black squares) and comparison with the ideal linear polarizer response (red circles) from
Malus’ law [105].

79
The transmission stopband for the perpendicular polarization state underwent a 0.7 nm wavelength
tuning of the Bragg resonance (not presented) over the tested temperature range (20 to 35 °C),
leading to slight broadening (from 0.8 to 0.92 nm) and strengthening (from 10 to 18 dB) of the
stopband. In the parallel polarization state, the transmission stopband narrowed from ≈5 to ≈1 nm
and weakened by up to 21 dB, presenting the opportunity for narrow-band, thermo-optic switching
with a strong variable attenuation in a specific polarization state. This thermo-optic polarization
response is exemplified by plotting the ratio of transmission spectra recorded at 26, 30 and 35 °C
with respect to the room temperature (20 °C) spectrum, yielding the dynamically tunable
polarization extinction response shown in Figure 6.9(b) for the three different temperature
changes, ΔT = 6, 10, and 15 °C, respectively. One notes the parallel polarization attenuation has
scaled by up to ≈22 dB in the case of a 15 °C temperature change (Figure 6.9(b), orange).

Given the positive dielectric anisotropy, Δɛ = ɛ∥ – ɛ⊥, known for the present NLC (5CB) when
under an AC electrical field modulation (1 kHz, square wave AC signal) [95], the axis of NLC
filaments was aligned perpendicular to the external electric field of the NLC cell (Figure 6.5). An
electro-rotation detuning of the NLC alignment was not observable in the FBG polarization spectra
when tested to the maximum field limit of 10 kV cm–1 for an AC frequency f = 1 kHz (square

Figure 6.9: The thermo-optic response of 2nd order FBG (1000 nanoholes) filled with NLC showing
a sequence of transmission spectra (a) recorded for parallel polarization over temperature range of 20
to 35 °C. The ratio of spectra (b) recorded at 26, 30 and 35 °C with respect to the room temperature
(20 °C) spectrum demonstrates a dynamically tunable polarization extinction response. The induced
thermo-optic attenuation for different temperature changes is shown in (b) for three different values
(ΔT = 6, 10, 15 °C) [105].

80
wave). The second type of NLC (MLC-2048 [188]) with a “dual frequency” response (DF-NLC)
was applied to the electro-optic cell (Figure 6.5) and tested with an electric field applied parallel
to capillaries. Both positive and negative values of the dielectric anisotropy were probed with AC
frequencies (square wave) applied at 6 and 65 kHz that lie respectively above and below the cross-
over frequency fc = 10 kHz, at room temperature. Electro-rotation responses in the stopbands were
again not discernible up to the maximum available field of 10 kV cm–1.

The interaction strength function of LC’s director is known anchoring energy [189] .The absent
electro-rotation modulation in both cases of parallel and perpendicular alignment of the electric
field, E, to the filament axis suggests a strong surface anchoring [190] of the NLC by the inner
walls of a narrow diameter nano-hole. A strong surface anchoring energy is thus naturally provided
by the narrow capillary geometry without any pre-treatment steps as required in traditional NLC
cells. However, the ultra-small cross-sectional diameter, d = 200 nm, of the capillary inhibits the
dielectric torque-induced orientational deformation of the NLC [191,192] raising the level of the
dielectric torque, G ∝ ΔɛE2, required to overcome the elastic energy density, Wela, of orientational
deformation. Indeed, the oriental deformation is limited by the diameter of the nano-holes,
underpinning an inelastic response of the NLC as the elastic energy density scales as Wela ∝ 1/d2.
To counter the rising inelasticity, equally strong scaling of the external field over the NLC layer is
required inside of the filament. Twin-holed optical fibers [193] provide one direction for scaling
up the fields by closer placement of the electrodes internally within the fiber cladding. With a 10-
fold reduction in the electrode spacing, D (from 250 to 25 µm), one can anticipate a 100× increase
in the dielectric torque, G ∝ E2 ∝ 1/d2. Laser structuring offers one direction micro- or nano-
structuring of open ports within the fiber core wherein fields perpendicular to the axis of holes
may be applied to the limits of dielectric break down, ≈5 MV cm–1, in fused silica [194], if we
consider the same dielectric properties for the fiber and for the NLC, see ref. [192]. Future work
on micro-structured fiber electrodes is thus a promising direction for enabling electro-rotation
responses to be induced in the present nano-capillary NLC gratings. Otherwise, thermo-optic
modulation has demonstrated the significant potential for dynamic polarization state control of
transmission and reflection of FBGs by using NLC materials.

Chapter 5 and 6 exploited the flexibility of laser filament writing method for fabrication of a
visible all-fiber spectrometer, cantilever based displacement sensor, optofluidic sensor, and

81
switchable polarization filter. The discussion on the significance of results is presented in next
Section 6.3. The conclusion and future work are in Chapter 7.

82
6.3 Discussion and Significance
The underlying objective of this thesis (Section 1.2) is centered on demonstrating the utility of
ultrafast laser filaments in creating unique designs of Bragg grating devices in optical fiber. The
study (Chapter 4) demonstrated that arrays of the laser-formed modification tracks could be
aligned to high resolution. Moreover, the laser parameters facilitated the formation of grating
elements having both low- and high-contrast in refractive index (i.e., n  10-3 and 0.46,
respectively) inside of single-mode optical fibers crossing over a large spectrum from the visible
to the IR bands.

For the case of low-contrast gratings, one objective sought to harness the narrowed radiative
patterns expected by optical scattering from grating elements having long and thin ‘line’ shapes.
In the development of the all-fiber spectrometer, the filament antenna array yielded radiative
modes of 1st order Bragg resonance that were narrowed to an unprecedented small angle of ~2o,
(Section 5.1) improving the brightness of the spectrometer. The filament writing methods have
been pushed to state-of-the-art limits in resolution of fiber Bragg gratings, presenting sufficiently
small grating pitches of  = 364 nm such that the fiber spectrometer emitted across the visible
spectrum to reach the blue edge of all comparable FBG devices at 450 nm wavelength. However,
the chirped periodicity of the grating (Section 5.1 ) was unique amongst filament-based FBGs in
permitting both spectral and azimuthal focusing of the external radiation patterns. In combination,
these elements yielded a compact fiber spectrometer that radiated a bright and highly resolved
spectrum ( = 335 pm) into a flat field positioned externally to the fiber cladding, facilitating
spectral recording with a CCD sensor, setting a new milestone by reaching the blue end of the
visible spectrum. Therefore, the low-contrast filament grating laser writing was inherently facile
and flexible for enabling novel high resolution grating designs that can efficiently tap out the fiber
waveguiding light, with high-finesse engineering of the diffractive focusing and dispersion in ways
not previously demonstrated.

In a second application objective for the low-contrast gratings, strain-sensing of the filament
arrays were adapted for displacement sensing. The fiber sensor was tested in a cantilever
arrangement presented in Section 5.2 where optomechanical responses were enhanced with stress
concentration. The off-axis positioning of the grating from the neutral axis was key for enabling
displacement optical sensing in single-mode fiber. The narrow and thin geometry of the filament

83
array afforded in confining sensing response into a zone of uniform strain field induced by the
lateral displacement, or alternatively, to null the response when filaments were orthogonally
oriented. In this way, a combination of parallel and rotated filament gratings provided a means for
sensor calibration, improved resolution, and azimuthally resolved displacement sensing. The
ability to measure transverse displacements paves the way toward developing more efficient
optical accelerometers or vibration sensors.

A major objective for this thesis was to scale up the energy of laser pulses and trigger a more
aggressive nano-explosion along the filament zone that would open nano-channel cavities inside
of standard telecommunication optical fiber as otherwise previously demonstrated in bulk glasses
[85]. The methods in this work (Section 4.1) overcomes astigmatic focusing by the cylindrical
fiber shape and distortions from nonlinear processes such a Kerr lensing and plasma defocusing to
force open capillaries with diameters of ~200 nm, to the capillaries formed on straight lines with
high aspect ratio of 500:1, and aligned with high density (~1 µm = 1000 lines/mm). The benefits
of the present method over prior art [79-84] are the non-contact and minimally invasive processing
for generating capillaries with high uniformity and order, reaching through the full fiber cross-
section without disturbing the integrity of the fiber cladding. The result was a facile capability in
writing photonic bandgap structures with high contrast in refractive index along the fiber core. In
this way, fibre Bragg gratings with strong second order stop bands were fabricated by the
methodology introduced in Chapter 4.

Having developed a new capability to generate ordered nano-capillaries in fiber, the final thesis
objective centered on the demonstrating capillary action by the nano-channels in bringing the
outside cladding environment directly into the core waveguide. Two directions of application were
explored. In the first, refractive index sensing was studied over a variety of FBG parameters to
optimize capillary diameter, pitch, grating length, and design, yielding strong grating responses
and effective capillary action over many fluids. Chemical etching was successfully employed to
tune the laser-formed holes and permit photonic-bandgap engineering of the stop bands. An
extraordinary range of RI sensing was demonstrated from n = 1 in air to 1.67 in oil, generating
large Bragg resonance shifts due to the physical presence of liquid media in the fiber core
(Section 6.1). One shortcoming of the high RI contrast in the gratings was a lowered sensing
resolution arising from the extraordinarily wide stopbands. However, the facile laser writing

84
enabled formation of π-shifted gratings, generating sharp resonances of ~150 pm, thus recovering
a high sensing precision.

In the second application, infiltration of anisotropic liquids such as nematic liquid crystals (NLC)
to the laser formed nano-capillaries and their polarization response were studied (in Section 6.2)
to induce birefringence and electro-rotation effects. The small, 200 nm diameter capillary was
shown to impose a strong axial molecular alignment of NLC, manifesting in strong optical
birefringence in the fiber polarization modes as presented by the shifted fiber Bragg grating
stopbands. The NLC polarization offered a diverse interplay with the FBG stopbands, cladding,
and radiation modes that were shown tunable by the thermo-optic effect. The key contribution
centered on the in-fiber dynamic control of polarization state by capillary and thermal effects,
while also exploring the opportunity for electro-rotation polarization modulation. The results
(Section 6.2) demonstrated broad implications for creating new types of polarization optics inside
of optical fiber. The flexible laser writing offered a facile means of tailoring the polarization
dependent Bragg grating spectral responses that are expected to be extend outside of the present
C-band study into infrared or visible bands, as well as into other types of fibers. Moreover, the
thermo- and electro-rotation responses of the NLC in the nano-capillary grating were promising
for intra-fiber dynamically switchable polarization control.

The combination of results here mark new milestones in laser processing of fiber optic devices,
harnessing shaped beams and ultrafast optical interactions to control the micro-structuring of
delicate fiber-optic components.

85
 Conclusion and Future Works

7.1 Conclusion
In summary, this work harnessed the flexibility and precision of femtosecond laser micro- and
nano-machining for fabrication of the state of art FBG based sensors. The filament processing
technique introduced previously in ref. [31] was improved in this work to enable fabrication of
low- and high-contrast filament structures in single-mode fibers. A 1D array of filament tracks
were assembled for generating FBGs, with new designs and properties tested in new directions of
application.

In this work, the low-contrast laser filaments have been harnessed to exploit their radiation modes
for fabrication of an all-fiber spectrometer while their thin profile was exploited in a displacement
sensing application (Section 5.1 and 5.2). Further studies and advances in writing of high-contrast
(hollow) filament tracks inside of single-mode optical fibers enabled the fabrication of a robust
optofluidic sensor. The hollow filament tracks offered the unique opportunity in fiber optics to
boost the RI contrast of FBGs from n = ~0.001 to 0.46, presenting an extra-ordinary opportunity
for engineering of unusually strong photonic bandgap devices in single-mode optical fibers.
Moreover, the open grating structure facilitated fluidic flow of the outside environment into the
core waveguiding region without significantly compromising the strength of the grating. In this
way, the filament-based FBGs opened new avenues for refractive index sensing and polarization
controlling as demonstrated in Section 6.1 and 6.2, respectively.

In Section 5.1, a compact all-fiber grating spectrometer is presented in the visible spectrum for the
first time, enabled by the single-pulse laser writing of long and uniform filament tracks. The visible
response marks the state of the art in point-by-point FBG writing, with a grating period of 364 nm.
The unique chirped filament geometry removed the need for focusing or collimating optics, while
generating favorably narrowed radiation patterns of bright and highly resolved lines of 335 pm
bandwidth. In this way, the fiber spectrometer did not require imaging or focusing lenses, thus
presenting an exceptional small footprint (less than 3 mm3) in comparison with free-space or nano-
fabricated on-chip spectrometers.

86
Section 5.2, builds on the laser writing of filament arrays inside of single-mode optical fiber to
generate a first order Bragg grating. The novelty centers on the first application of this grating to
displacement sensing, tested in a cantilever arrangement. Unlike other fiber Bragg gratings, the
filament array defines an ultra-thin sensing volume for characterizing the photoelastic response
from a single plane of bending moment. Off-axis placement of the filaments within the narrow
confines of a core waveguide is ideally suited for enhancing the displacement sensing in single
mode optical fibers. The filament geometry provided a sinusoidal response on the azimuth,
conforming with optomechanical modeling. Multiple gratings have been overlaid in a single fiber
to harvest more precise and complete displacement data based on vector sensing capabilities and
cross-calibration that improved the azimuthal sensitivity and corrected for non-concentric fiber
core. Section 5.2 provided the groundwork for integrating multifunctional sensor elements in all
types of optical fibers, with extension to distributed sensing that would potentially enhance
portable and biocompatible micro-devices. The present sensor is suited to meet specific
requirements in a wide range of application directions for millimeter-scale displacements such as
structural health monitoring, material flowmeter, and accelerometer in high temperature and harsh
environments.

The technical breakthrough in Chapter 6 centered on harnessing the laser filament ‘nano-
explosion’ to open nano-hole arrays fully through the cladding and core waveguide of single-mode
optical fibers. The process is non-contact and less invasive than any other prior art, which left the
fiber cladding fully intact. Alternatively, this thesis opened the door to a new realm of creating
strong and compact photonic-stop band devices directly in traditional optical fibers while
facilitating environmental sensing through a thick robust cladding that retains the strength of the
fiber. While there are already existing demonstrations of nano- fabricated Bragg gratings, the nano-
holes could only be opened after destruction of large parts of the fiber cladding and using time-
consuming and difficult multi-step processes. The present nano-capillary is fully punched out with
a single laser pulse that can potentially scale to fabrication speeds of 1 million holes per second.

The hollow-filament tracks provided high aspect ratio nano-holes with tunable hole diameters from
200 to 700 nm that extended fully through the core and cladding zones. The holes served as nano-
fluidic channels for drawing solvents or oils from the surrounding cladding environment directly
into the fiber core without inducing high losses or de-activating waveguiding. The nano-holes were

87
assembled and remained isolated without breakthrough at high spatial densities of 1000 filaments
per mm.

Section 6.1 centered on the application of hollow-filament FBGs as a new means for optical
sensing of the fiber cladding environment, without directing any of the light away from the guiding
core of the fiber. Such nano-capillary arrays presented strong, second-order Bragg grating
stopbands (i.e., λ = 1550 nm for Λ = 1.072 µm periodicity) in the C-telecommunication band,
inducing large ≈40 nm Bragg shifts for in-fiber refractive index sensing when tested with isotropic
fluids (n = 1.0 – 1.67). The narrow π-shifted stopbands provided strong optical responses to sense
the local environment, as demonstrated in the wide 1548–1590 nm shift in Bragg wavelengths.
The RIU sensitivity reached as high as 600 nm/RIU for the case of highest RI (n= 1.66) and largest
hole diameter (700 nm). This RIU sensitivity is comparable to the best FBG-based demonstrations
to date (i.e., 945 nm/RIU in ref. [169]).

Section 6.2 has shown that the hollow filament grating presented a strong capillarity effect to draw
nematic liquid crystal (NLC) into homogeneous alignment with the cylindrical walls and presented
a high birefringent response in the second-order Bragg stopband. The NLC provided a strong
extinction ratio of up to 20 dB over a ≈5 nm band with only a moderate insertion loss of <1 dB
arising between the two polarization states of the shifted stopbands. The Bragg grating followed
the thermo-optic response of the NLC to further offered dynamic tuning and switching of the
polarization extinction response without an increase in the insertion loss.

The unique hollow filament grating defined a novel platform for driving LC's polarization
responses that is potentially extensible to other fiber types, for example, in the broad infrared and
visible spectral coverage (section 6.2). The reported LC-filled nano-hole array underpinned a new
type of FBG. This is promising for in-fiber dynamic control of polarization state by capillary,
thermal, and electro-rotation effects. Although electro-rotation responses would be highly
desirable for enabling high-speed polarization responses, the present methods and device designs
are suited to meet specific requirements in a wide range of application directions where the
orientational or thermal response time of the LC is acceptable for “adaptation” or “reconfiguration”
functions.

88
7.2 Future Work
The fabrication method in this thesis is extensible to other types of fibers and fiber materials, or to
the structuring of two-dimensional arrays of filaments in the fiber. The 2D photonic bandgap
structure of such grating patterns provides a novel opportunity for harnessing 2D photonics crystal
structures along the light propagation direction of optical fiber for the first time and thereby
engineer the bandgap and radiation modes. As an example, femtosecond laser point-by-point
writing of 2D low- and high-contrast filament arrays could be extended in the core waveguide of
single-mode or multi-mode optical fibers. The filament writing would enable the 2D photonic
bandgap engineering of the confined, cladding and radiation modes in the optical fiber, permitting
Bragg coupling to a single external radiation mode as demonstrated for an all-fiber spectrometer
application. The azimuthal radiative scattering properties of the filament together with the new
potential for 2D photonic bandgap design opens a powerful means to engineer and shape the
radiation patterns both inside of and outside of the fiber to broaden the range of FBG applications.

For the case of the all-fiber spectrometer, the main limitation in the present device is the low
efficiency which was around 1.5 - 2% on each side of the fiber. A new direction for optimizing the
efficiency performance of the spectrometer could be explored by forming 2D array of nanohole
filament structures in the guiding core and interact with a larger portion of the mode profile. A
further improvement may be obtained from the 2D ‘blazed’ design of the hollow filament array
structure to redistribute the efficiency of different diffractive orders. Photonic bandgap design
may be used to redirect the -1st order external radiation mode into a single and bright +1st order
radiation mode, increasing the overall spectrometer efficiency. Several other options for improving
the efficiency may be considered, which begin with strengthening the grating. Longer FBGs are
straightforward to fabricate, which will require a redesign of the chirping rate and the imaging
distance as well as the consideration of Fresnel or total internal reflection loss and aberration at
the air–cladding interface. Using a higher laser exposure to strengthen the refractive index contrast
is limited here by the optical focusing resolution limit of aberrated beams at 364 nm grating pitch.

In another frontier, beam shaping systems based on Spatial Light Modulators (SLMs) have proven
to be a new inscribing technique for rapid and flexible writing of VBGs and FBGs in bulk and
fiber glasses with low photosensitivity response. The SLM beam shaping methodology further
allows the re-shaping of ultrashort laser pulses with Gaussian beam profile to other intensity

89
profiles such as light sheets or multi-foci patterns. The re-shaping of intensity profiles allows the
fabrication of various structures in the core. Fabrication of FBGs with SLM, further provides a
unique opportunity of controlling the coupling of light to the cladding and radiation modes which
is not possible with other interference-based methods [195,196].

90
 References

[1] G. P. Agrawal, “Fiber-Optic Communication Systems: Fourth Edition,” John Wiley & Sons.
(2011).
[2] R. -J. Essiambre and R. W. Tkach, "Capacity Trends and Limits of Optical Communication
Networks," in Proceedings of the IEEE, 100(5), 1035-1055(2012).
[3] K. Bergman, M. Glick, M. Bahadori, Q. Cheng, and S. Rumley, “Recent advances in optical
technologies for data centers: a review,” Optica, 5(11) 1354-1370 (2018).
[4] K. O. Hill and G. Meltz, “1997. Fiber Bragg grating technology fundamentals and overview,” J.
Lightwave Technol., 15(8) 1263–1276 (1997).
[5] A. D. Kersey et al., “Fiber grating sensors,” J. Lightwave Technol., 15(8), 1442–1463 (1997).
[6] N. Sabri, S. A. Aljunid, M. S. Salim, and S. Fouad, “Fiber Optic Sensors: Short Review and
Applications,” Springer Series in Materials Science 204, 99–311 (2015).
[7] D. Chen et al., “An in-line fiber optic fabry-perot sensor for high-temperature vibration
measurement,” Micromachines, 11(3), 252 (2020).
[8] X. Li, Y. Shao, Y. Yu, Y. Zhang, and S. Wei, “A highly sensitive fiber-optic Fabry–Perot
interferometer based on internal reflection mirrors for refractive index measurement,” Sensors,
16(6) ,794 (2016).
[9] C. Campanella, A. Cuccovillo, C. Campanella, A. Yurt, and V. Passaro, “Fibre Bragg Grating
Based Strain Sensors: Review of Technology and Applications,” Sensors, 18(9), 3115 (2018).
[10] B. Lu et al., “Distributed optical fiber hydrophone based on -OTDR and its field test,” Optics
Express, 29(3), 3147–3162 (2021).
[11] S. J. Mihailov, “Fiber Bragg Grating Sensors for Harsh Environments,” Sensors ,12(2), 1898–1918
(2012).
[12] I. Floris, J. M. Adam, P. A. Calderón, and S. Sales, “Fiber Optic Shape Sensors: A comprehensive
review,” Opt Lasers Eng, 139, 106508 (2021).
[13] S. Pissadakis, “Lab-in-a-fiber sensors: A review,” Microelectron Eng, 217, 111105 (2019).
[14] N. Roveri, G. Pepe, F. Mezzani, A. Carcaterra, A. Culla, and S. Milana, “OPTYRE—Real Time
Estimation of Rolling Resistance for Intelligent Tyres,” Sensors 19(23),5119 (2019).

91
[15] P. Vaiano, B. Carotenuto, M. Pisco, A. Ricciardi, G. Quero, M. Consales, A. Crescitelli, E.
Esposito, and A. Cusano, “Lab on Fiber Technology for biological sensing applications,” Laser
Photon Rev, 10(6), 922–961, (2016).
[16] W. Liang, Y. Huang, Y. Xu, R. K. Lee, and A. Yariv, “2005. Highly sensitive fiber Bragg grating
refractive index sensors,” Appl. Phys. Lett., 86(15), 151122, (2005).
[17] B. Li, L. Jiang, S. Wang, L. Zhou, H. Xiao, and T. Hai-Lung, “Ultra-abrupt tapered fiber Mach-
Zehnder interferometer sensors,” Sensors 11(6), 5729–5739 (2011).
[18] X. Fang, C. R. Liao, and D. N. Wang, “Femtosecond laser fabricated fiber Bragg grating in
microfiber for refractive index sensing,” Opt. Lett. 35(7), 1007–1009 (2010).
[19] D. Pallarés-Aldeiturriaga, P. Roldán-Varona, L. Rodríguez-Cobo, and J. M. López-Higuera,
“Optical Fiber Sensors by Direct Laser Processing: A Review,” Sensors 20(23), 6971 (2020).
[20] M. Ochoa, J. F. Algorri, P. Roldan-Varona, L. Rodriguez-Cobo, and J. M. Lopez-Higuera, “Recent
Advances in Biomedical Photonic Sensors: A Focus on Optical-Fibre-Based Sensing,” Sensors
21(19), 6469 (2021)
[21] B. Gibson, C. Martelli, J. Canning, N. Groothoff, P. Olivero, and S. Huntington, “Micromachining
structured optical fibers using focused ion beam milling,” Opt. Lett. 32(11), 1575-1577 (2007).
[22] M. Bernier, F. Trépanier, J. Carrier, and R. Vallée, “High mechanical strength fiber Bragg gratings
made with infrared femtosecond pulses and a phase mask,” Opt. Lett. 39(12), 3646–3649 (2014).
[23] F. Bilodeau, D. C. Johnson, S. Thériault, B. Malo, J. Albert, and K. O. Hill, “An All-Fiber Dense-
Wavelength-Division Multiplexer/Demultiplexer Using Photoimprinted Bragg Gratings,” IEEE
Photonics Technology Letters, 7(4), 388–390 (1995).
[24] K. Ennser, M. N. Zervas, and R. I. Laming, “Optimization of apodized linearly chirped fiber
gratings for optical communications,” IEEE J Quantum Electron 34(5), 770–778, (1998).
[25] Y. J. Rao, D. J. Webb, D. A. Jackson, L. Zhang, and I. Bennion, “In-fiber bragg-grating
temperature sensor system for medical applications,” Journal of Lightwave Technology 15(5),
779–784 (1997).
[26] R. Kashyap, Fiber Bragg Gratings. Elsevier Inc., (2010).
[27] B. Malo, K. O. Hill, F. Bilodeau, D. C. Johnson, and J. Albert, “Point-by-point fabrication of
micro-Bragg gratings in photosensitive fibre using single excimer pulse refractive index
modification techniques,” Electron Lett, 29(18), 1668–1669 (1993).
[28] A. Martinez, M. Dubov, I. Khrushchev, and I. Bennion, “Direct writing of fibre Bragg gratings by
femtosecond laser,” Electron Lett, 40(19), 1170–1172 (2004).

92
[29] K. Zhou, M. Dubov, C. Mou, L. Zhang, V. K. Mezentsev, and I. Bennion, “Line-by-line fiber
bragg grating made by femtosecond laser,” IEEE Photonics Technology Letters 22(16), 1190–
1192 (2010).
[30] C. Voigtländer, R. Becker, J. Thomas, D. Richter, A. Singh, A. Tünnermann, and S. Nolte,
“Ultrashort pulse inscription of tailored fiber Bragg gratings with a phase mask and a deformed
wavefront,” Optical Materials Express, 1(4), 633-642 (2011).
[31] E. Ertorer, M. Haque, J. Li, and P. R. Herman, “Femtosecond laser filaments for rapid and flexible
writing of fiber Bragg grating,” Opt. Express, 26(7), 9323 (2018).
[32] J. Thomas, N. Jovanovic, R. Becker, G. Marshall, M. Withford, A. Tünnermann, S. Nolte, and M.
Steel, “Cladding mode coupling in highly localized fiber Bragg gratings: modal properties and
transmission spectra,” Opt. Express, 19(1), 325-341 (2011).
[33] D. Grobnic, C. W. Smelser, S. J. Mihailov, R. B. Walker and P. Lu, "Fiber Bragg gratings with
suppressed cladding modes made in SMF-28 with a femtosecond IR laser and a phase mask," in
IEEE Photonics Technology Letters, 16(8), 864-1866 (2004).
[34] J. Albert and Y. Y. Shevchenko, “Plasmon resonances in gold-coated tilted fiber Bragg gratings,”
Optics Letters, 32(3), 211–213 (2007).
[35] J. L. Wagener, T. A. Strasser, J. R. Pedrazzani, J. DeMarco, and D. J. DiGiovanni, “Fiber grating
optical spectrum analyzer tap,” IEE Conference Publication, 5, 65–68 (1997).
[36] H. Qin, Q. He, Y. Moreno, Z. Xing, X. Guo, Z. Yan, Q. Sun, K. Zhou, D. Liu, and L. Zhang,
“Compact linear polarization spectrometer based on radiation mode shaped in-fiber diffraction
grating,” Optics Letters, 44(21), 5129-5132 (2019).
[37] C. Waltermann, P. Guehlke, J. Koch, and W. Schippers, “Shaping Spectra within Optical Fibers,”
Photonics Views, 16(2), 42–45 (2019).
[38] D. Kinet, P. Mégret, K. W. Goossen, L. Qiu, D. Heider, and C. Caucheteur, “Fiber Bragg Grating
Sensors toward Structural Health Monitoring in Composite Materials: Challenges and Solutions,”
Sensors (14)4, 7394–7419 (2014).
[39] D. lo Presti et al., “Fiber bragg gratings for medical applications and future challenges: A review,”
IEEE Access 8,156863–156888 (2020).
[40] J. Villatoro, E. Antonio-Lopez, A. Schülzgen, and R. Amezcua-Correa, “Miniature multicore
optical fiber vibration sensor,” Opt. Lett. 42(10), 2022-2025 (2017).

93
[41] W. Bao, N. Sahoo, Z. Sun, C. Wang, S. Liu, Y. Wang, and L. Zhang, "Selective fiber Bragg grating
inscription in four-core fiber for two-dimension vector bending sensing," Opt. Express 28(18),
26461-26469 (2020).
[42] M. Hou et al., “Two-dimensional vector bending sensor based on seven-core fiber Bragg gratings,”
Opt. Express, 26(18), 23770 (2018).
[43] J. Cui, Z. Liu, D. Gunawardena, Z. Zhao, and H. Tam, "Two-dimensional vector accelerometer
based on Bragg gratings inscribed in a multi-core fiber," Opt. Express 27(15), 20848-20856
(2019).
[44] T. Boilard, G. Bilodeau, S. Morency, Y. Messaddeq, R. Fortier, F. Trépanier, and M. Bernier,
"Curvature sensing using a hybrid polycarbonate-silica multicore fiber," Opt. Express 28(26),
39387-39399 (2020).
[45] J. Kong, X. Ouyang, A. Zhou, H. Yu, and L. Yuan, “Pure directional bending measurement with
a fiber Bragg grating at the connection joint of eccentric-core and single-mode fibers,” Journal of
Lightwave Technology, 34(14), 3288–3292 (2016).
[46] Y. Ouyang, J. Liu, X. Xu, Y. Zhao, and A. Zhou, “Phase-shifted eccentric core fiber bragg grating
fabricated by electric arc discharge for directional bending measurement,” Sensors 18(4), 1168
(2018).
[47] D. Zheng, Z. Cai, I. Floris, J. Madrigal, W. Pan, X. Zou, and S. Sales, "Temperature-insensitive
optical tilt sensor based on a single eccentric-core fiber Bragg grating," Opt. Lett. 44(22), 5570-
5573 (2019).
[48] J. H. Osório et al., “Bragg gratings in surface-core fibers: Refractive index and directional
curvature sensing,” Optical Fiber Technology, 34, 86–90 (2017).
[49] F. Chen, X. Qiao, R. Wang, D. Su, and Q. Rong, “Orientation-dependent fiber-optic displacement
sensor using a fiber Bragg grating inscribed in a side-hole fiber,” Appl Opt, 57(13), 3581 (2018).
[50] S. Kibben, M. Kropp, G. Dumstorff, M. Koerdt, W. Lang, and F. Vollertsen, “Bend sensor based
on fibreoptics and concept for a compact evaluation unit,” Production Engineering 7(1), 15–22
(2013).
[51] X. Sun, Z. Chang, L. Zeng, L. Zhang, Y. Hu, and J. Duan, “Simultaneous vector bending and
temperature sensing based on eccentric multi-mode fiber Bragg gratings,” Sens Actuators A Phys.
331, 112903 (2021).

94
[52] W. Bao, X. Qiao, and Q. Rong, “Fiber-optic vector accelerometer using orthogonal bragg grating
inscription over innermost cladding of a multi-clad fiber,” Journal of Lightwave Technology,
37(11), 2706–2712 (2019).
[53] B. Jiang et al., “Co-located angularly offset fiber Bragg grating pair for temperature-compensated
unambiguous 3D shape sensing,” Applied Optics, 60(14), 4185–4189 (2021).
[54] D. Feng, J. Albert, and X. Qiao, “Off-axis ultraviolet-written fiber Bragg gratings for directional
bending measurements,” Opt. Lett. 41(6), 1201–1204 (2016).
[55] S. Zhang, Y. Liu, H. Guo, A. Zhou, and L. Yuan, “Highly Sensitive Vector Curvature Sensor
Based on Two Juxtaposed Fiber Michelson Interferometers with Vernier-Like Effect,” IEEE Sens
J, 19(6), 2148–2154 (2019).
[56] J. Wo, Q. Sun, X. Li, D. Liu, and P. P. Shum, “Biconical-taper-assisted fiber interferometer with
modes coupling enhancement for high-sensitive curvature measurement,” Appl Phys B 115(1), 1–
8 (2014).
[57] J. Li et al., “Micro-Machined Optical Fiber Side-Cantilevers for Acceleration Measurement,”
IEEE Photonics Technology Letters 29(21), 1836–1839 (2017).
[58] Q. Zhang, T. Zhu, J. Zhang, and K. S. Chiang, “Micro-fiber-based FBG sensor for simultaneous
measurement of vibration and temperature,” IEEE Photonics Technology Letters 25(18), 1751–
1753 (2013).
[59] L. Zhang, Y. Jiang, J. Jia, P. Wang, S. Wang, and L. Jiang, “Fiber-optic micro vibration sensors
fabricated by a femtosecond laser,” Opt Lasers Eng 110, 207–210 (2018).
[60] L. Y. Shao and J. Albert, “Lateral force sensor based on a core-offset tilted fiber Bragg grating,”
Opt Commun, 284(7), 1855–1858 (2011).
[61] S. Cai et al., “Fiber-Optic Accelerometer Using Tilted Grating Inscribed in Depressed Cladding
Fibers,” IEEE Photonics Technology Letters 29(24), 2171–2174, (2017).
[62] T. Guo, F. Liu, B. O. Guan, and J. Albert, “Tilted fiber grating mechanical and biochemical
sensors,” Opt Laser Technol 78,19–33 (2016).
[63] D. Su, X. Qiao, F. Chen, and Q. Rong, “Higher order coupling mode for orientation-dependent
bend measurement using an off-axis FBG inscription over few-mode fiber,” IEEE Sens J, 19(4),
1368–1372 (2019).
[64] A. Doering et al., “Multiple off-axis fiber Bragg gratings for 3D shape sensing,” Applied Optics
57(28), 8125–8133 (2018).

95
[65] A. Martinez, Y. Lai, M. Dubov, I. Y. Khmshchev, and I. Bennion, “Vector bending sensors based
on fibre Bragg gratings inscribed by infrared femtosecond laser,” Electron Lett, 41(8), 472–474
(2005).
[66] D. Viveiros et al., “Femtosecond laser direct written off-axis fiber Bragg gratings for sensing
applications,” Opt Laser Technol, 128, 106227 (2020).
[67] F. Chen, X. Li, R. Wang, X. Qiao, and X. Qiao, “Two-dimensional vector accelerometer based on
orthogonal Bragg gratings inscribed in a standard single-mode fiber cladding,” Opt. Lett. 46(12),
2992-2995 (2021).
[68] C. Monat, P. Domachuk, and B. J. Eggleton, “Integrated optofluidics: a new river of light,” Nat.
Photonics 1(2), 106–114 (2007).
[69] G. Pitruzzello and T. F. Krauss, “Photonic crystal resonances for sensing and imaging,” J. Opt.
20(7), 073004 (2018).
[70] D. Q. Yang, B. Duan, X. Liu, A. Q. Wang, X. G. Li, and Y. F. Ji, “Photonic crystal nanobeam
cavities for nanoscale optical sensing: a review,” Micromachines 11(1), 72 (2020).
[71] A. F. Gavela, D. G. García, J. C. Ramirez, and L. M. Lechuga, “Last advances in silicon-based
optical biosensors,” Sensors 16(3), 285 (2016).
[72] M. G. Scullion, Y. Arita, T. F. Krauss, and K. Dholakia, “Enhancement of optical forces using
slow light in a photonic crystal waveguide,” Optica 2(9), 816–821 (2015).
[73] E. Luan, H. Shoman, D. M. Ratner, K. C. Cheung, and L. Chrostowski, “Silicon photonic
biosensors using label-free detection,” Sensors, 18(10), 3519 (2018).
[74] Y. Zhi, X. C. Yu, Q. Gong, L. Yang, and Y. F. Xiao, “Single nanoparticle detection using optical
microcavities,” Adv. Mater. 29(12), 1604920 (2017).
[75] A. Cusano, D. Paladino, and A. Iadicicco, “Microstructured fiber Bragg gratings,” J. Lightwave
Technol. 27(11), 1663–1697 (2009).
[76] Y. Xu, P. Lu, L. Chen, and X. Bao, “Recent developments in micro-structured fiber optic sensors,”
Fibers 5(1), 3 (2017).
[77] A. Urrutia, I. del Villar, P. Zubiate, and C. R. Zamarreño, “A comprehensive review of optical
fiber refractometers: Toward a standard comparative criterion,” Laser Photonics Rev. 13(11),
1900094 (2019).
[78] Z. Liu, Z. F. Zhang, H. Y. Tam, and X. Tao, “Multifunctional smart optical fibers: materials,
fabrication, and sensing applications,” Photonics 6(2), 48 (2019).

96
[79] Y. Liu, C. Meng, A. P. Zhang, Y. Xiao, H. Yu, and L. Tong, “Compact microfiber Bragg gratings
with high-index contrast,” Opt. Lett. 36(16), 3115–3117 (2011).
[80] K. P. Nayak et al., “Cavity formation on an optical nanofiber using focused ion beam milling
technique,” Opt. Express, 19(15), 14040–14050 (2011).
[81] P. Zhao, Y. Li, J. Zhang, L. Shi, and X. Zhang, “Nanohole induced microfiber Bragg gratings,”
Opt. Express 20(27), 28625–28630 (2012).
[82] W. Li, J. Du, V. G. Truong, and S. Nic Chormaic, “Optical nanofiber-based cavity induced by
periodic air-nanohole arrays,” Appl. Phys. Lett. 110(25), 253102 (2017).
[83] K. P. Nayak and K. Hakuta, “Photonic crystal formation on optical nanofibers using femtosecond
laser ablation technique,” Opt. Express 21(2),2480–2490 (2013).
[84] C. Chen, H.-B. Sun, Q.-D. Chen, R. Yang, and Y.-S. Yu, “Rapid fabrication of microhole array
structured optical fibers,” Opt. Lett. 36(19), 3879-3881 (2011).
[85] M. K. Bhuyan et al., “High aspect ratio nanochannel machining using single shot femtosecond
Bessel beams,” Appl. Phys. Lett. 97(8), 081102 (2010).
[86] M. K. Bhuyan et al., “High aspect ratio taper-free microchannel fabrication using femtosecond
Bessel beams,” Opt. Express 18(2), 566–574 (2010).
[87] F. Courvoisier, J. M. Dudley, L. Furfaro, M. Jacquot, M. K. Bhuyan, and P.-A. Lacourt, “Surface
nanoprocessing with nondiffracting femtosecond Bessel beams,” Opt. Lett. 34(20), 3163-3165,
(2009).
[88] T. Chen, G. Zhang, Y. Wang, X. Li, R. Stoian, and G. Cheng, “Reconstructing of Embedded High-
Aspect-Ratio Nano-Voids Generated by Ultrafast Laser Bessel Beams,” Micromachines, 11(7),
671 (2020).
[89] C. D’Amico, G. Cheng, G. Zhang, M. Bhuyan, and R. Stoian, “Efficient point-by-point Bragg
gratings fabricated in embedded laser-written silica waveguides using ultrafast Bessel beams,”
Opt. Lett. 43(9), 2161–2164 (2018).
[90] C. D’Amico, G. Cheng, G. Zhang, M. K. Bhuyan, R. Stoian, and Y. Wang, “Ultrashort Bessel
beam photoinscription of Bragg grating waveguides and their application as temperature sensors,”
Photonics Research 7(7), 806–814 (2019).
[91] S. Ertman, K. Rutkowska, and T. R. Wolinski, “Recent progress in liquid-crystal optical fibers and
their applications in photonics,” Journal of Lightwave Technology 37(11), 2516–2526 (2019).

97
[92] T. J. Chen and S. H. Chen, “Propagation of Lower-Order Modes in a Radially Anisotropic
Cylindrical Waveguide with Liquid Crystal Cladding,” Journal of Lightwave Technology 13(8),
1698–1705 (1995).
[93] S. H. Chen and T. J. Chen, “Observation of mode selection in a radially anisotropic cylindrical
waveguide with liquid‐crystal cladding,” Appl Phys Lett. 64(15), 1893 (1998).
[94] A. K. Pitilakis, D. C. Zografopoulos, and E. E. Kriezis, “In-line polarization controller based on
liquid-crystal photonic crystal fibers,” Journal of Lightwave Technology 29(17), 2560–2569
(2011).
[95] P.-G. de Gennes, The physics of liquid crystals. Oxford, UK: Clarendon Press, 1993.
[96] E. S. Goldburt and P. S. J. Russell, “Electro‐optical response of a liquid‐crystalline fiber coupler,”
Appl Phys Lett, vol. 48(1), 10 (1998), doi: 10.1063/1.96772.
[97] T. Tanggaard Larsen et al., “Optical devices based on liquid crystal photonic bandgap fibres,” Opt.
Express 11(20), 2589–2596 (2003).
[98] Y. Liao et al., “Alignment of liquid crystal molecules in a micro-cell fabricated by femtosecond
laser,” Chem Phys Lett. 498(1–3), 188–191 (2010).
[99] F. Serra et al., “Nematic Liquid Crystals Embedded in Cubic Microlattices: Memory Effects and
Bistable Pixels,” Adv Funct Mater, 23(32), 3990–3994 (2013).
[100] A. Rahnama, K. M. Aghdami, Y. H. Kim, and P. R. Herman, “Ultracompact Lens-Less
‘Spectrometer in Fiber’ Based on Chirped Filament-Array Gratings,” Adv Photonics Res. 1(2),
2000026 (2020).
[101] A. Rahnama, H. Mahlooji, G. Djogo, F. Azhari, and P. R. Herman, “Filament-arrayed Bragg
gratings for azimuthally resolved displacement sensing in single-mode fibers,” Opt. Express 30(3),
4189-4201(2022).
[102] K. K. Lee et al., “Temperature-compensated fiber-optic 3D shape sensor based on femtosecond
laser direct-written Bragg grating waveguides,” Opt. Express 21(20), 24076-24086 (2013).
[103] H. Zhang, S. Ho, S. M. Eaton, J. Li, and P. R. Herman, "Three-dimensional optical sensing network
written in fused silica glass with femtosecond laser," Opt. Express 16(18), 14015-14023 (2008).
[104] K. Mahmoud Aghdami, A. Rahnama, E. Ertorer, and P. R. Herman, “Laser nano-filament
explosion for enabling open-grating sensing in optical fibre,” Nat. Commun. 12(1), 6344 (2021).
[105] A. Rahnama, T. Dadalyan, K. Mahmoud Aghdami, T. Galstian, and P. R. Herman, “In-Fiber
Switchable Polarization Filter Based on Liquid Crystal Filled Hollow-Filament Bragg Gratings,”
Adv. Opt. Mater. 9(19), 2100054 (2021).

98
[106] X. Guo, Z. Xue, and Y. Zhang, “Manufacturing of 3D multifunctional microelectronic devices:
challenges and opportunities,” NPG Asia Materials 11(1),1–7 (2019).
[107] D. Tan, Z. Wang, B. Xu, and J. Qiu, “Photonic circuits written by femtosecond laser in glass:
improved fabrication and recent progress in photonic devices,” Adv. Photon. 3(2), 024002 (2021).
[108] L. Jonušauskas, R. Martínez Vázquez, F. Bragheri, and P. Paiè, “3D Manufacturing of Glass
Microstructures Using Femtosecond Laser,” Micromachines 12(5), 499 (2021).
[109] F. He et al., “Femtosecond Laser Fabrication of Monolithically Integrated Microfluidic Sensors in
Glass,” Sensors 14(10), 19402-19440 (2014).
[110] L. Shah, A. Arai, S. Eaton, and P. Herman., “Waveguide writing in fused silica with a femtosecond
fiber laser at 522 nm and 1 MHz repetition rate,” Opt. Express, 13(6), 1999-2006 (2005).
[111] J. He, B. Xu, X. Xu, C. Liao, and Y. Wang, “Review of Femtosecond-Laser-Inscribed Fiber Bragg
Gratings: Fabrication Technologies and Sensing Applications,” Photonic Sensors 11(2), 203–226
(2021).
[112] L. Kelemen, E. Lepera, B. Horváth, P. Ormos, R. Osellame, and R. Martínez Vázquez, “Direct
writing of optical microresonators in a lab-on-a-chip for label-free biosensing,” Lab Chip 19(11),
1985–1990 (2019).
[113] J. M. Maia, V. A. Amorim, D. Viveiros, and P. V. S. Marques, “Femtosecond laser
micromachining of an optofluidics-based monolithic whispering-gallery mode resonator coupled
to a suspended waveguide,” Sci. Rep. 11, 9128 (2021).
[114] M. Spagnolo, R. Motta, R. Memeo, F. Pellegatta, A. Crespi, and R. Osellame, "Resonant opto-
mechanical modulators and switches by femtosecond laser micromachining," Opt. Express
28(16), 23133-23142 (2020).
[115] S. He, Q. Yang, B. Zhang, Y. Ren, H. Liu, P. Wu, Y. Yao, F. Chen, “A waveguide mode modulator
based on femtosecond laser direct writing in KTN crystals,” Results Phys. 18, 103307 (2020).
[116] J. Han, Y. Zhang, C. Liao, Y. Jiang, Y. Wang, C. Lin, S. Liu, J. Wang, Z. Zhang, J. Zhou, and Y.
Wang, "Fiber-interface directional coupler inscribed by femtosecond laser for refractive index
measurements," Opt. Express 28(10), 14263-14270 (2020).
[117] J. Grenier, L. Fernandes, and P. Herman, "Femtosecond laser inscription of asymmetric directional
couplers for in-fiber optical taps and fiber cladding photonics," Opt. Express 23(13), 16760-16771
(2015).
[118] R. Stoian, “Volume photoinscription of glasses: three-dimensional micro- and nanostructuring
with ultrashort laser pulses,” Appl Phys A Mater Sci Process 126(6), 1–30 (2020).

99
[119] M. Ams et al., “Investigation of ultrafast laser-photonic material interactions: Challenges for
directly written glass photonics,” IEEE Journal on Selected Topics in Quantum Electronics, 14(5),
1370–1388 (2008).
[120] S. S. Mao et al., “Dynamics of femtosecond laser interactions with dielectrics,” Applied Physics
A 79(7), 1695–1709 (2004).
[121] M. v. Fedorov, “L. V. Keldysh’s ‘Ionization in the Field of a Strong Electromagnetic Wave’ and
modern physics of atomic interaction with a strong laser field,” Journal of Experimental and
Theoretical Physics 122(3), 449–455, (2016).
[122] K. Itoh, W. Watanabe, S. Nolte, and C. B. Schaffer, “Ultrafast Processes for Bulk Modification of
Transparent Materials,” MRS Bulletin 31(8), 620–625, (2011).
[123] R. Osellame, H. J. W. M. Hoekstra, G. Cerullo, and M. Pollnau, “Femtosecond laser
microstructuring: an enabling tool for optofluidic lab-on-chips,” Laser Photon. Rev. 5(3), 442–463
(2011).
[124] J. W. Chan, T. R. Huser, S. H. Risbud, J. S. Hayden, and D. M. Krol, “Waveguide fabrication in
phosphate glasses using femtosecond laser pulses,” Appl Phys Lett 82(15), 2371, (2003).
[125] L. Sudrie, M. Franco, B. Prade, and A. Mysyrowicz, “Study of damage in fused silica induced by
ultra-short IR laser pulses,” Opt Commun. 191(3), 333–339 (2001).
[126] C. Hnatovsky et al., “Pulse duration dependence of femtosecond-laser-fabricated nanogratings in
fused silica,” Appl. Phys. Lett. 87(1), 014104 (2005).
[127] E. N. Glezer and E. Mazur, “Ultrafast-laser driven micro-explosions in transparent materials,”
Appl. Phys. Lett, 71(7), 882 (1998).
[128] H. D. Nguyen et al., “Non-Diffractive Bessel Beams for Ultrafast Laser Scanning Platform and
Proof-Of-Concept Side-Wall Polishing of Additively Manufactured Parts,” Micromachines
11(11), 974 (2020).
[129] R. Stoian, M. K. Bhuyan, G. Zhang, G. Cheng, R. Meyer, and F. Courvoisier, “Ultrafast Bessel
beams: Advanced tools for laser materials processing,” Advanced Optical Technologies 7(3), 165–
174 (2018).
[130] M. Duocastella and C. B. Arnold, “Bessel and annular beams for materials processing,” Laser
Photon. Rev. 6(5), 607–621 (2012).
[131] P. Salter, M. Baum, I. Alexeev, M. Schmidt, and M. Booth, "Exploring the depth range for three-
dimensional laser machining with aberration correction," Opt. Express 22(15), 17644-17656
(2014).

100
[132] F. Ahmed, M. S. Lee, H. Sekita, T. Sumiyoshi, and M. Kamata, “Display glass cutting by
femtosecond laser induced single shot periodic void array,” Applied Physics A 93(1), 189–192
(2008).
[133] J. Li, E. Ertorer, and P. R. Herman, “Ultrafast laser burst-train filamentation for non-contact
scribing of optical glasses,” Opt. Express 27(18), 25078-25090 (2019).
[134] M. Mikutis, T. Kudrius, G. Šlekys, D. Paipulas, and S. Juodkazis, "High 90% efficiency Bragg
gratings formed in fused silica by femtosecond Gauss-Bessel laser beams," Opt. Mater. Express
3(11), 1862-1871 (2013).
[135] F. Ahmed and M. B. G. Jun, “Filament-based fabrication and performance analysis of fiber Bragg
grating sensors using ultrashort pulse laser,” J. Micro. Nanomanuf. 2(2), 021007 (2014).
[136] A. Lamberti, G. Luyckx, W. van Paepegem, A. Rezayat, and S. Vanlanduit, “Detection,
Localization and Quantification of Impact Events on a Stiffened Composite Panel with Embedded
Fiber Bragg Grating Sensor Networks,” Sensors 17(4), 743(2017).
[137] G. Meltz, W. Morey, and W. Glenn, "Formation of Bragg gratings in optical fibers by a transverse
holographic method," Opt. Lett. 14(15), 823-825 (1989).
[138] S. Loranger, V. Lambin-Iezzi, and R. Kashyap, "Reproducible ultra-long FBGs in phase corrected
non-uniform fibers," Optica 4(9), 1143-1146 (2017).
[139] G. Marshall, R. Williams, N. Jovanovic, M. Steel, and M. Withford, "Point-by-point written fiber-
Bragg gratings and their application in complex grating designs," Opt. Express 18(19), 19844-
19859 (2010).
[140] P. Lu, S. Mihailov, H. Ding, D. Grobnic, R. Walker, D. Coulas, C. Hnatovsky, and A. Naumov,
"Plane-by-Plane Inscription of Grating Structures in Optical Fibers," J. Lightwave Technol. 36(4),
926-931 (2018).
[141] P. Roldán-Varona, D. Pallarés-Aldeiturriaga, L. Rodríguez-Cobo, and J. López-Higuera, "Slit
Beam Shaping Technique for Femtosecond Laser Inscription of Enhanced Plane-by-Plane FBGs,"
J. Lightwave Technol. 38(16), 4526-4532 (2020).
[142] A. Rahnama, Y. Kim, K. M. Aghdami, and P. R. Herman, "2D filament grating array: enabling an
efficient, high-resolution lens-less all-fiber spectrometer," Proc. SPIE 11676, Frontiers in Ultrafast
Optics: Biomedical, Scientific, and Industrial Applications XXI, 116760Y (2021).
[143] L. Rapp et al., “High aspect ratio micro-explosions in the bulk of sapphire generated by
femtosecond Bessel beams,” Sci. Rep. 6, 34286 (2016)

101
[144] T. Chen, G. Zhang, Y. Wang, X. Li, R. Stoian, and G. Cheng, “Reconstructing of Embedded High-
Aspect-Ratio Nano-Voids Generated by Ultrafast Laser Bessel Beams,” Micromachines 11(7),
671 (2020).
[145] J. del Hoyo, R. Meyer, L. Furfaro, and F. Courvoisier, “Nanoscale confinement of energy
deposition in glass by double ultrafast Bessel pulses,” Nanophotonics 10(3), 1089–1097 (2021).
[146] M. Haque, K. K. C. Lee, S. Ho, L. A. Fernandes, and P. R. Herman, “Chemical-assisted
femtosecond laser writing of lab-in-fibers,” Lab Chip 14(19), 3817–3829 (2014).
[147] M. Haque and P. R. Herman, “Chemical‐assisted femtosecond laser writing of optical resonator
arrays,” Laser Photonics Rev. 9(6), 656–665 (2015).
[148] G. Djogo, J. Li, S. Ho, M. Haque, E. Ertorer, J. Liu, X. Song, J. Suo, and P. R Herman,
“Femtosecond laser additive and subtractive micro-processing: enabling a high-channel-density
silica interposer for multicore fibre to silicon-photonic packaging,” International Journal of
Extreme Manufacturing 1(4), 045002 (2019).
[149] G. Djogo, A. Rahimnouri, S. Ho, J. Li, and P. R. Herman, “Femtosecond laser micro-structuring
of air-disk mirrors in glass for out-of-plane waveguide coupling in silicon photonic packaging,”
Proc. SPIE 11676, Frontiers in Ultrafast Optics: Biomedical, Scientific, and Industrial
Applications XXI, 116760S (2021).
[150] A. Rahimnouri, G. Djogo, and P. Herman, "3D Laser Structured Mirror-Waveguide Circuits: a
New Optical PCB Platform for Silicon Photonics," in 2021 Conference on Lasers and Electro-
Optics Europe and European Quantum Electronics Conference, OSA Technical Digest (Optica
Publishing Group, 2021), paper cm_2_5.
[151] G. Djogo, A. Rahimnouri, J. Li, S. Ho, and P. R. Herman, “Fiber-to-silicon photonics interposer
for high-density vertical coupling through laser structured mirror-waveguide circuits,” Proc. SPIE
PC12007, Optical Interconnects XXII, PC120070H (2022).
[152] A. Rahnama, Y. H. Kim, K. M. Aghdami, and P. R. Herman, “Visible-light, All-fiber Spectrometer
Based on Radiative Emission From a Chirped Filament Grating Array,” OSA Advanced Photonics
Congress (AP) 2020 (IPR, NP, NOMA, Networks, PVLED, PSC, SPPCom, SOF) (2020), paper
SoTh1H.4, vol. Part F190-SOF 2020, p. SoTh1H.4, Jul. 2020.
[153] W. Neumann, “Fundamentals of Dispersive Optical Spectroscopy Systems,” Fundamentals of
Dispersive Optical Spectroscopy Systems, Jul. 2014.
[154] X. Ma, M. Li, and J. J. He, “CMOS-compatible integrated spectrometer based on echelle
diffraction grating and MSM photodetector array,” IEEE Photonics J. 5(2), 6600807 (2013).

102
[155] P. Cheben, J. Schmid, A. Delâge, A. Densmore, S. Janz, B. Lamontagne, J. Lapointe, E. Post, P.
Waldron, and D. Xu, "A high-resolution silicon-on-insulator arrayed waveguide grating
microspectrometer with sub-micrometer aperture waveguides," Opt. Express 15(5), 2299-2306
(2007).
[156] G. Calafiore et al., “Holographic planar lightwave circuit for on-chip spectroscopy,” Light:
Science & Applications 3(9), e203 (2014).
[157] J. Field, S. Berry, R. Bannerman, D. Smith, C. Gawith, P. Smith, and J. Gates, "Highly-chirped
Bragg gratings for integrated silica spectrometers," Opt. Express 28(14), 21247-21259 (2020).
[158] A. Rahnama, H. Mahlooji, G. Djogo, F. Azhari, and P. R. Herman, “Harnessing filament arrays
for azimuthally resolved displacement sensing in fiber Bragg gratings,” Proc. SPIE PC11991,
Frontiers in Ultrafast Optics: Biomedical, Scientific, and Industrial Applications XXII,
PC119910L (2022).
[159] H. Mahlooji, A. Rahnama, G. Djogo, F. Azhari, and P. R. Herman, “Off-Axis Filament Based
Fiber Bragg Gratings for Azimuthally Resolved Displacement Sensing,” 2021 Conference on
Lasers and Electro-Optics Europe and European Quantum Electronics Conference, CLEO/Europe-
EQEC 2021, Jun. 2021.
[160] A. Martinez, Y. Lai, M. Dubov, I. Y. Khmshchev, and I. Bennion, “Vector bending sensors based
on fibre Bragg gratings inscribed by infrared femtosecond laser,” Electron. Lett. 41(8), 472–474
(2005).
[161] L. Xiong, G. Jiang, Y. Guo, Y. Kuang, and H. Liu, "Investigation of the Temperature
Compensation of FBGs Encapsulated With Different Methods and Subjected to Different
Temperature Change Rates," J. Lightwave Technol. 37(3), 917-926 (2019).
[162] M. M. Khan, N. Panwar, and R. Dhawan, “Modified cantilever beam shaped FBG based
accelerometer with self temperature compensation,” Sens Actuators A Phys. 205, 79–85, (2014).
[163] T. Qiu, S. Yang, and A. Wang, "Experimental investigation of point-by-point off-axis Bragg
gratings inscribed by a femtosecond laser in few-mode fibers," Opt. Express 28(25), 37553-37565
(2020).
[164] F. Chen, R. Wang, X. Li, and X. Qiao, "Orientation-dependent fiber-optic accelerometer based on
eccentric fiber Bragg grating," Opt. Express 29(18), 28574-28581 (2021).
[165] J. He, Z. Chen, X. Xu, J. He, B. Xu, B. Du, K. Guo, R. Chen, and Y. Wang, "Femtosecond laser
line-by-line inscription of apodized fiber Bragg gratings," Opt. Lett. 46(22), 5663-5666 (2021).

103
[166] T. Goebel, G. Bharathan, M. Ams, M. Heck, R. Krämer, C. Matzdorf, D. Richter, M. Siems, A.
Fuerbach, and S. Nolte, "Realization of aperiodic fiber Bragg gratings with ultrashort laser pulses
and the line-by-line technique," Opt. Lett. 43(15), 3794-3797 (2018).
[167] K. M. Aghdami, E. Ertorer, A. Rahnama, and P. R. Herman, “Femtosecond laser opening of
hollow-filament arrays: The Fiber Bragg Grating Opto-fluidic Sensor,” 2019 Photonics North, PN
2019, May 2019.
[168] K. O. Hill and G. Meltz, “Fiber Bragg grating technology fundamentals and overview,” Journal of
Lightwave Technology, 15(8), 1263–1276 (1997).
[169] K. Zhou, Y. Lai, X. Chen, K. Sugden, L. Zhang, and I. Bennion, "A refractometer based on a
micro-slot in a fiber Bragg grating formed by chemically assisted femtosecond laser processing,"
Opt. Express 15(24), 15848-15853 (2007).
[170] A. Rahnama, T. Dadalyan, K. M. Aghdami, T. Galstian, and P. R. Herman, “All-fiber, dynamically
tunable, polarization filter based on liquid crystal filled nano-channels,” Proc. SPIE PC12012,
Advanced Fabrication Technologies for Micro/Nano Optics and Photonics XV, PC120120K (5
March 2022).
[171] A. Rahnama, T. Dadalyan, K. M. Aghdami, T. Galstian and P. R. Herman, "In-fiber Polarization
Control Using Nematic Liquid Crystal in Nano-Capillary Bragg Grating Array," 2021 Conference
on Lasers and Electro-Optics (CLEO), 2021, pp. 1-2.
[172] L. Feng, Y. Li, H. Zhou, S. Wu, C. Yang, W. Li, J. Qiu, H. Guo, X. Hong, and J. Wu, "All-fiber
polarization-division multiplexing to mode-division multiplexing conversion for hybrid optical
networks," Appl. Opt. 57(36), 10528-10533 (2018).
[173] E. Rochat, S. Walker, and M. Parker, "Polarisation and wavelength division multiplexing at
1.55μm for bandwidth enhancement of multimode fibre based access networks," Opt. Express
12(10), 2280-2292 (2004).
[174] C. Caucheteur, T. Guo, and J. Albert, "Polarization-Assisted Fiber Bragg Grating Sensors: Tutorial
and Review," J. Lightwave Technol. 35(16), 3311-3322 (2017)..
[175] Q. Bao et al., “Broadband graphene polarizer,” Nature Photon. 5, 411–415 (2011).
[176] H. Zhang, N. Healy, L. Shen, C. Huang, N. Aspiotis, D. Hewak, and A. Peacock, "Graphene-Based
Fiber Polarizer With PVB-Enhanced Light Interaction," J. Lightwave Technol. 34(15), 3563-3567
(2016).
[177] J. Feth and C. Chang, "Metal-clad fiber-optic cutoff polarizer," Opt. Lett. 11(6), 386-388 (1986).

104
[178] K. Liu, W. Sorin, and H. Shaw, "Single-mode-fiber evanescent polarizer/amplitude modulator
using liquid crystals," Opt. Lett. 11(3), 180-182 (1986).
[179] V. I. Kopp and A. Z. Genack, “Adding twist,” Nature Photonics 5(8), 470–472 (2011).
[180] L. Yang, L. Xue, C. Li, J. Su, and J. Qian, "Adiabatic circular polarizer based on chiral fiber
grating," Opt. Express 19(3), 2251-2256 (2011).
[181] Z. Yan, K. Zhou, and L. Zhang, "In-fiber linear polarizer based on UV-inscribed 45&#xB0; tilted
grating in polarization maintaining fiber," Opt. Lett. 37(18), 3819-3821 (2012).
[182] B. Ortega et al., “High-performance optical fiber polarizers based on long-period gratings in
birefringent optical fibers,” IEEE Photonics Technology Letters 9(10), 1370–1372 (1997).
[183] S. H. Chen and T. J. Chen, “Observation of mode selection in a radially anisotropic cylindrical
waveguide with liquid‐crystal cladding,” Appl. Phys. Lett. 64(15), 1893 (1998).
[184] S. Ertman, K. Rutkowska, and T. Woliński, "Recent Progress in Liquid-Crystal Optical Fibers and
Their Applications in Photonics," J. Lightwave Technol. 37(11), 2516-2526 (2019).
[185] E. S. Goldburt and P. S. J. Russell, “Electro‐optical response of a liquid‐crystalline fiber coupler,”
Appl. Phys. Lett. 48(1), 10, (1998).
[186] T. Larsen, A. Bjarklev, D. Hermann, and J. Broeng, "Optical devices based on liquid crystal
photonic bandgap fibres," Opt. Express 11(20), 2589-2596 (2003).
[187] V. Tkachenko et al., “Nematic Liquid Crystal Optical Dispersion in the Visible-Near Infrared
Range,” Molecular Crystals and Liquid Crystals 454, 263-271(2006).
[188] O. Pishnyak, S. Sato, and O. Lavrentovich, "Electrically tunable lens based on a dual-frequency
nematic liquid crystal," Appl. Opt. 45(19), 4576-4582 (2006).
[189] J. Kim, “Effective Anchoring Energy of Nematic Liquid Crystals on Stripe Anchoring Patterns”
Jpn. J. Appl. Phys. 52, 080201(2013).
[190] K. Takatoh, M. Sakamoto, R. Hasegawa, M. Koden, and N. Itoh, “Alignment Technology and
Applications of Liquid Crystal Devices,” Alignment Technology and Applications of Liquid
Crystal Devices (2005).
[191] O. Sova, V. Reshetnyak, and T. Galstian, "Modulation transfer function of liquid crystal
microlenses and microprisms using double dielectric layer," Appl. Opt. 57(1), 18-24 (2018)..
[192] K. Asatryan, V. Presnyakov, A. Tork, A. Zohrabyan, A. Bagramyan, and T. Galstian, "Optical lens
with electrically variable focus using an optically hidden dielectric structure," Opt. Express
18(13), 13981-13992 (2010)..

105
 [193] T. Fujiwara, D. Wong, and S. Fleming, “Large Electrooptic Modulation in a Thermally-Poled
Germanosilicate Fiber,” IEEE Photonics Technology Letters 7(10), 1177–1179 (1995).
[194] W.M. Haynes, D.R. Lide, and T.J. Bruno, 2016. CRC handbook of chemistry and physics. CRC
press.
[195] P. Zavyalova, A. Rahnama, E. Alimohammadian, J. Li, S. Sivanandam, and P. R. Herman "From
filaments to light-sheets: tailoring the spectrum of fiber Bragg gratings with femtosecond lasers",
Proc. SPIE 11676, Frontiers in Ultrafast Optics: Biomedical, Scientific, and Industrial
Applications XXI, 116760X (30 March 2021).
[196] P. Zavyalova, S. Sivanandam, P. R. Herman, E. Alimohammadian, and A. Rahnama "Tunable
fibre Bragg grating arrays for spectral cross-correlation", Proc. SPIE 12184, Ground-based and

106

View publication stats