Investigation of structured fibres
for nonlinear endomicroscopy
imaging
A thesis submitted for the degree of
Doctor of Philosophy
by
Navin Prakash Ghimire
Supervisors
Prof. Min Gu
Dr.Hongchun Bao
Dr. Xiangping Li
Centre for Micro-Photonics
Faculty of Science, Engineering and Technology
Swinburne University of Technology
Melbourne, Australia
2015
Declaration
ii
Declaration
I, Navin Prakash Ghimire, declare that this thesis entitled:
Investigation of structured fibres for nonlinear endomicroscopy imaging
is my own work and has not been submitted previously, in whole or in part, in
respect of any other academic award.
Navin Prakash Ghimire
Centre for Micro-Photonics
Faculty of Science, Engineering and Technology
Swinburne University of Technology
Melbourne, Australia
Dated this day, 22 April, 2015
Acknowledgments
iii
Acknowledgments
I am thankful to the Centre for Micro-Photonics at Swinburne University of
Technology for giving me an opportunity to pursue my Doctor of Philosophy. I
would like to thank my supervisors Prof. Min Gu , Dr. Hongchun Bao, Dr.
Xiangping Li for their effort in my research skills development. Prof. Min Gu’s
tireless effort and guidance have helped me conduct and convey scientific research
in a professional manner. I thank Dr. Hongchun Bao and Dr. Xiangping Li for their
suggestions and discussion in consolidating results.
I would like thank administrative staff and technicians for helping to conduct the
research smoothly. Thanks go to all my colleagues at the Centre for Micro-
Photonics for valuable discussion and suggestions during the period.
Finally, I would like to thank my parents, all my teachers, family and friends for
their encouragement and support.
Navin Prakash Ghimire
22 April, 2015
Abstract
iv
Abstract
Broadband excitation and collection in a fibre-optic nonlinear endomicroscope is
realized by using a single hollow-core double clad photonics crystal fibre (DCPCF)
and a gradient index (GRIN) lens. Femtosecond pulses with central wavelengths in
the range of 750 - 850 nm can be directly delivered through the core of the fibre for
nonlinear excitation without pre-chirping. A gradient index lens with numerical
aperture of 0.8 designed to operate over the near-infrared wavelength range is used
for simultaneously focusing the laser beam from the fibre for nonlinear excitation
and collecting the fluorescent signal from samples. It is possible to optimally excite
different fluorescent markers with different excitation wavelengths without
externally adjusting a dispersion compensation for each wavelength using this
Abstract
v
system. This compact system is suitable to perform nonlinear imaging of multiple
fluorophors in the wavelength range of 750-850 nm.
In the contest of implementing super-resolution techniques such as stimulated
emission depletion (STED) in fibre-optic endoscopy imaging, beam propagation
for different input polarization states ( linear and cylindircal with/ without
superimposed the vortex phase ) through the core of the hollow-core photonics
crystal fibre (HC-PCF) fibre is characterized. Our experiment suggests that the
doughnut mode with the centre intensity null can be propagated through the core of
the hollow-core photonic crystal fibre. A doughnut mode and a fundamental mode
coupling can be achieved simultaneously in the core of the fibre by varying the
beam width and hence the effective numerical aperture (NA) through the single
coupling objective. This is a curcial step towards implementing the single
wavelength fibre-optic super-resolution endomicroscopic imaging.
Finally, realizing the need of a robust solid silica fibre to withstand the fast
scanning requirement of a miniaturized video rate imaging system and also with
the view of implementation of such fibre in a fibre based STED endomicroscope, a
double-clad solid silica fibre is newly designed for applications that need the high
efficiency operation of light at two colors. Ultra-short pulses at a central
wavelength of 800 nm are delivered by the core of the double-clad fibre which can
realize the transmission of the optical pulses with a net chromatic dispersion of
Abstract
vi
zero. This is achieved by integrating the double-clad fibre with a pair of long
period gratings, which allows optical pulses to propagate in a higher order mode
(LP02) in the middle of the fibre as well as in a fundamental mode (LP01) at the
beginning and the end of the fibre. The index profile of the double-clad fibre is
engineered so that the higher order mode has high anomalous dispersion that can be
used to compensate for normal dispersion of the fundamental mode. By controlling
the lengths of the fibre where pulses are in a fundamental and in a higher order
mode, the fibre with total zero dispersion can be realized. The new double-clad
fibre can collect 100% of visible light within the NA of 0.21 with a loss of the
optical pulses less than 1%. The design of this fibre is essential for applications
including fibre-optic nonlinear imaging for compactness, robustness, and low
optical power loss in dispersion compensation.
Table of contents
vii
Table of contents
Declaration ii
Acknowledgments iii
Abstract iv
Table of contents vii
List of figures and tables xi
Chapter 1: Introduction
1.1 Introduction 1
1.2 Thesis preview 7
Chapter 2: Background and literature review
2.1 Nonlinear microscopy 10
2.2 Optical property of tissue 13
Table of contents
viii
2.3 Fluorescence collection and imaging depth 16
2.4 Optical resolution 20
2.4.1 Super-resolution techniques 22
2.4.2 Single wavelength stimulated emission
and depletion microscopy imaging 27
2.5 Fibre-optic imaging system 28
2.5.1 Optical fibres used in endoscopy 30
2.5.2 Endoscope probe components 35
2.6 Ultra-fast pulse propagation 40
2.7 Dispersion in Optical fibre 41
2.7.1 Long period fibre gratings 45
2.7.2 Dispersion compensation using LPG 47
2.7.3 Long period fibre grating design 48
Chapter 3: Hollow-core photonics crystal fibre for
broadband nonlinear endoscopy
3.1 Introduction 52
3.2Advantages of HC-PCF in nonlinear endoscopy 53
3.3 Pulse width measurement through HC-PCF fibre 55
3.4 Broadband excitation and collection in
nonlinear endomicroscopy 58
3.5 Experimental results and discussion 63
3.6 Conclusion 64
Table of contents
ix
Chapter 4: Propagation of a doughnut beam through
hollow-core photonics crystal fibre
4.1Introduction 65
4.2 Experimental setups 68
4.3 Experimental results 69
4.3.1 Linear polarization states 68
4.3.2 Radial polarization states 70
4.3.3 Azimuthal polarization states 72
4.3.4 Circular polarization states 74
4.4 Conclusion 84
Chapter 5: Design of zero dispersive double-clad fibre
for operation of two color light
5.1 Introduction 85
5.2 Zero chromatic dispersion 88
5.3 Mode conversions 96
5.4 High efficient operation of two color light 100
5.5 Conclusion 102
Table of contents
x
Chapter 6: Conclusion
6.1 Thesis conclusion 103
6.2 Future work 106
Bibliography 107
Appendix
Dispersion relation -cladding modes 134
Author’s Publications 139
List of figures and table
xi
List of figures and tables
Fig1.1 Diagram illustrating nonlinear interaction of light with matter …………...11
Fig. 2.2: Refractive index and image profile of GRIN lens………………..…......37
Fig 3.1: Experimental setup for pulse width measurement with frequency resolved
optical gating (FROG); Laser: Ti: Sapphire laser (Spectra-Physics, Mai Tai HP,
~100 fs,80 MHz, 690-1040 nm), ND neutral density, OBJ 1 & OBJ2: 20x 0.25 NA;
XS1&XS2:3D fibre coupling stage; M1&M2: Ultrafast mirrors; FROG setup,
Swamp Optics, GRENOUILLE-008-50-USB)……………………………………55
Fig 3.2: (a) Spectral intensity and phase, (b) autocorrelation trace of (c) measured
and (d) retrieved pulse after the propagation through the 1.5 m length…………...56
List of figures and table
xii
Fig. 3.3: Pulse width measured at the output of the 1.5 m HC-PCF for different
wavelengths at 100 mw………………………………………………………….57
Fig. 3.4: (a) Schematic diagram of the experimental set-up for a broadband
excitation and collection system for single fibre nonlinear endomicroscopy. PMT:
photo-multiplier tube. (b) Enlarged part of the probe consisting of a HC-PCF and a
GRIN lens. (c)-(e) Mode profiles of the HC-PCF for different wavelengths. (f)
Output power at different wavelengths for the input power of 20 mW through a 1.5
m HC-PCF……......................................................................................................58
Fig. 3.5: Log-Log plot of the two-photon-excited fluorescence intensity (If) versus
the excitation laser power (Ip ) of fluorescent beads for wavelengths of 760 nm,
810 nm and 850 nm. …………………………………………………………...…60
Fig. 3.6: Log-Log plot of the two-photon-excited fluorescence intensity versus the
excitation laser power for different lengths of the fibre. ……………………...…61
Fig. 3.7: (a)-(f) Two-photon fluorescence images of 1 μm fluorescent beads (scale
bar: 5 µm) ; (g) Lateral resolution (full width at half maximum) of 1 μm fluorescent
beads for the excitation laser power of 4.5 mW at the samples. (h)-(i) Two-photon
fluorescence images of 2 μm fluorescent beads and the Rhodamine B dye (scale
List of figures and table
xiii
bar: 10 µm) at 800 nm (h) with emission filter Semrock – FF01-647/51 and ( i)
Schott – BG18. (j) Log-Log plot of two-photon-excited fluorescence intensity of
Rhodamine B dye versus excitation laser power at 800 nm………………..……..62
Fig. 4.1: (a) Schematic diagram of the experimental set-up for the characterization
of doughnut beams through a hollow-core photonics crystal fibre. ND: neutral
Density, OBJ: objectives, BM: beam manipulation, VA: variable aperture, HC-
PCF: zero dispersion hollow-core photonic crystal around the wavelength of ~807
nm, CCD: charge couple device, A: Analyser ……………………..……….…….68
Fig. 4.2: (a) Intensity ratio of dip and peak of the fibre mode for different
numerical aperture. (b)-(f). Mode profiles for the input linear polarization states
and at different angles of the analyser with respect to the vertical
axis……………………………………………………………….………………..70
Fig. 4.3: (a) Intensity ratio of dip and peak of the fibre mode for different
numerical aperture. (b)-(f). Mode profiles for the input linear polarization states
superimposed with the the vortex phase and at different angles of the analyser with
respect to the vertical axis……………………………………………….…..……72
List of figures and table
xiv
Fig. 4.4: (a) Intensity ratio of dip and peak of the fibre mode for different
numerical aperture. (b)-(f). Mode profiles for the input radial polarization states
and at different angles of the analyser with respect to the vertical axis…………..73
Fig. 4.5: (a) Intensity ratio of dip and peak of the fibre mode for different
numerical aperture. (b)-(f). Mode profiles for the input radial polarization states
superimposed with the vortex phase and at different angles of the analyser with
respect to the vertical axis ………………………………………………………..75
Fig. 4.6: (a) Intensity ratio of dip and peak of the fibre mode for different
numerical aperture value. (b)-(f). Mode profiles for the input azimuthal
polarization states and at different angles of the analyser with respect to the vertical
axis ………………………………………………………………………..………76
Fig. 4.7: (a) Numerical aperture versus the normalized intensity ( Imin / Imax) for
an azimuthal polarization state overlapped with a the vortex phase. (b)-(f). Mode
profiles for the input azimuthal polarization state superimposed with the vortex
phase and at different angle of the analyser with respect to the vertical axis kept
before the CCD. ……………………………………………………….………….78
List of figures and table
xv
Fig. 4.8: (a) Intensity ratio of dip and peak of the fibre mode for different
numerical aperture. (b)-(f). Mode profiles for the input circular polarization states
and at different angles of the analyser with respect to the vertical axis
………………………………………………………………………..……..……..79
Fig. 4.9: (a) Intensity ratio of dip and peak of the fibre mode for different
numerical aperture. (b)-(f). Mode profiles for the input circular polarization states
superimposed with the vortex phase and at different angles of the analyser with
respect to the vertical axis . (g) Doughnut mode profile at the fibre output, (h) cross
section intensity profile……………………………………………………………82
Fig. 5.1: (a) Schematic structure of a DCF intergrated with a pair of gratings for
achieving zero net chromatic dispersion. (b) Refractive index profile of the DCF.89
Fig.5.2: Mode evolution for mode LP01 and mode LP02 at wavelengths 725 nm
(blue), 775 nm (green) and 825 nm (red) for the w-DCF. The refractive index
profile is shown in background (black)……………………………………….…..90
List of figures and table
xvi
Fig. 5.3: Waveguide dispersion Dw of a higher order mode in the DCF (a) for
different core diameters d1, (b) for different thicknesses of the index dip region d2,
and (c) for different cladding thicknesses d3. …………………………….…….92
Fig. 5.4: (a) Dispersion values for the higher order mode. (b) Dispersion values for
the fundamental mode…………………………………..………………..…...…93
Fig. 5.5: Total dispersion of the fibre under the condition (L1+L3) :L2 = 1:3.77…94
Fig. 5.6: Mode conversion of the first (a) and the second (b) long period grating
are the intensity of input in mode LP01, input in mode LP02,
output in mode LP01 and output in mode LP02……………………………………98
Fig.5.7: (a) Schematic diagram of the DCF for operating two color light. (b) The
fibre collection efficiency η of the visible light versus the NA of the beam coupled
by an objective………………………………..………………………………….101
Table 2.1: Comparison of DCF and DCPCF features for fibre endoscopes…….32
outoutininHFHF
IIII ,,,
Chapter 1
1
Chapter 1
Introduction
1.1 Introduction
Nonlinear imaging is one of the best optical imaging tools available today for
investigating cellular functions, the cell - cell interaction and the cell migration in
dense thick deep tissues. Deep tissue imaging is possible because of nonlinear
excitation with a long wavelength for achieving reduced scattering/absorption and
diffuse fluorescence collection from a spatially localized focus spot that limits
photo toxicity/bleaching in the out of focus region [1, 2]. Nonlinear endoscopy
with a small probe and a long thin optical fibre has more advantages in flexibility,
handiness, and compactness than bench-top imaging systems for biomedical and
Chapter 1
2
clinical imaging applications [3-5]. Miniaturized hand-held fibre-optic devices are
ideal for their flexibility and potential use in live suspects long term imaging
studies to be able to trace the development stages of different complication in
different body parts that are difficult to access with minimal invasion. Nonlinear
fibre-optic imaging has added the advantage of improved resolution and deeper
tissue penetration over other linear systems [5, 7]. Fibre-optic nonlinear imaging
devices are the only available tools for use in biomedical research for the three-
dimensional (3D) visualization of cellular layers, long term imaging studies and
minimally invasive clinical diagnosis and surgical procedures of difficult to access
areas of living and moving suspects.
Different fibre types have been employed in nonlinear microscopic imaging
systems over a decade to improve the image quality by efficient excitation and
collection of the fluorescence signal [8-11]. The double-clad fibre structure has
made it possible both to excite samples through the core and simultaneously collect
the fluorescence signal through the cladding of the fibre. The use of a fibre coupler
in the system makes it possible to separate the excitation and fluorescence signal
within the fibre [12-15]. Combination of the zero dispersive fibre, the double-clad
fibre, the fibre coupler and miniaturized probes with 3D imaging capabilities would
make it feasible now to have all fibre endoscopy. Till now single mode fibre
(SMF), solid silica double-clad fibre (DCF) and double-clad photonics crystal fibre
(DCPCF) have been used in nonlinear endoscopy that utilizes a single fibre both
Chapter 1
3
for excitation and fluorescence signal collection. These studies show room for
improvements in various aspects. Though there are fibre designs that impose zero
dispersion to the ultra-fast pulses in the near-infrared region there are no studies
available that use single fibre for both excitation and signal collection. In this work
we make an investigation of structured fibres for single fibre nonlinear
endomicroscopy.
Although one type of fibre is preferred over the other types of fibre, the major
hurdle with fibre-optic nonlinear imaging is still imposed by the group velocity
dispersion in fibres. When femtosecond excitation pulses propagate through a fibre,
they suffer chromatic dispersion from the fibre which broadens the pulses and
reduces the nonlinear excitation efficiency. Femtosecond pulses with a central
wavelength of 800 nm are widely used in nonlinear imaging because the peak of
two-photon excitation wavelengths of fluorescein and acriflavine solutions is near
800 nm. They are the only fluorophores that has been approved for use in human.
All solid silica fibres have a high normal chromatic dispersion at a wavelength of
800 nm. To solve this optical dispersion problem, a pre-chirp unit external to a
fibre is normally used in fibre-optic nonlinear imaging systems for the dispersion
compensation [12, 16-20]. A pre-chirp unit generally employs grating pairs,
prisms, chirp mirrors and acoustic-optics modulators (AOM) [21, 22]. By adjusting
the distance between mirrors/prisms/gratings, the angular orientation of these
components and the beam size, the different frequency components of a pulsed
Chapter 1
4
beam obtain different optical path lengths and thus compensate for the chromatic
dispersion in the fibre [23]. However, endoscopic systems with a pre-chirp unit are
not only cumbersome to operate but also limited for the dispersion compensation of
pulsed beams with different central wavelengths. In addition, the dispersion
compensation by grating pairs or prisms requires a laser beam to have multiple
reflections or transmissions through the units, which induce high loss to the laser
beam power [24, 25]. For this reason the design and operation of fibre-optic
nonlinear imaging systems is limited to a single near-infrared (NIR) wavelength
and broadband excitation and collection feature in a single fibre based endoscopy
device is not realized until recently.
Resolution at the diffraction limited value in a fibre-optic nonlinear imaging system
is another major challenge, which researchers around the globe are trying to tackle.
Resolution is directly proportional to the wavelength used and inversely
proportional to the numerical aperture (NA) of a lens used. Thus shorter
wavelengths and higher NA oil immersion objective lenses are used to resolve a
feature size approximately half of the wavelength. Biological imaging of a thick
tissue sample imposes restriction to use of a short wavelength and use of oil
immersion objective in a contact mode. Thus smallest feature size that could be
imaged under the in vivo condition of thin sample is limited until recently before
the availability of bench-top super-resolution imaging technology.
Chapter 1
5
To directly visualize the sub-cellular structure (smaller than the wavelength
dimension), fluorescence microscopy combines the specific anti-body/dye labeling
technique with light microscope system for achieving diffraction limited/unlimited
bio-imaging [26, 27]. Recently the optical nonlinearity of the fluorophores has been
used in combination to achieve diffraction unlimited imaging of fluorescent
labelled sub-cellular structures. Resolution beyond the diffraction barrier was
achieved by the combination of specific dyes/antibodies labelling of samples and
use of two lasers under the stringent alignment and synchronized condition [27-30].
Fibre-optic based endomicroscopic devices can offer the inherent alignment of two
beams for incorporating this advanced super-resolution imaging features like
stimulated emission depletion (STED) imaging. Detailed study of the propagation
of multiple wavelengths, multiple polarization states of optical-beams for
broadband excitation/emission is necessary to make these concepts feasible. Also,
using one wavelength for both two-photon excitation and quenching of the
fluorescence makes the optimization of optics of the STED system simple and
possible for 3D-imaging of thick scattering tissue samples.
In this thesis, we investigate the use of hollow-core photonics crystal fibre (HC-
PCF) in fibre-optic nonlinear endoscopy as well as theoretically design a DCF with
two values of the NA for the efficient transmission of two colors of light, near-
infrared optical pulses as well as the visible continuous wave (CW) bream. We
demonstrate the feasibility of broadband excitation and collection in a single fibre
Chapter 1
6
based nonlinear endomicorsocopic system using a piece of HC-PCF integrated with
a gradient index lens (GRIN). The DCF fibre designed to have two inbuilt long
period gratings (LPGs) enables the transmission of the near-infrared optical pulses
with zero net chromatic dispersion in its core as well as the high efficiency
collection of visible light coupled by an objective lens. This DCF is promising to
be used in a compact nonlinear fibre-optic imaging system.
Chapter 1
7
1.2 Thesis preview
Chapter 1 introduces the fibre-optic nonlinear imaging system, it’s applications and
explains the motivation behind the research works conducted along with methods
implemented. It also briefly describes the contents of each chapter.
Chapter 2 contains the theoretical background and review of the topics dealt in this
thesis. Topics like nonlinear microscopy and endoscopy, fibre-optic nonlinear
imaging, dispersion and its compensation in optical fibres, super-resolution
techniques and single wavelength super-resolution imaging are explained and
reviewed in detail in this chapter. The chapter starts with fundamentals of light
mater interaction and nonlinear microscopy. Optical properties of tissues like
absorption, scattering and their wavelength dependence and the propagation of
light through tissue are discussed in relation to imaging. Ways to maximize the
extent of excitation and collection from the depth for two-photon imaging are
explained in detail. Optical resolution, diffraction limits and super-resolution
techniques are explained in brief. Different types of optical fibres used in nonlinear
microscopy and different technical challenges like linear and nonlinear properties
of the optical fibre are described in relation to achieving the zero dispersion and
thus understanding the propagation of ultra-fast pulse at near-infrared wavelengths
Chapter 1
8
for the successful implementation and design of fibre-optic endomicroscopy
system. Various relations for the calculation of linear dispersion and nonlinear
parameters like the third harmonic distortion, dispersion length, and nonlinear
length are given. The imaging property of a typical GRIN lens is described in
detail as this component is used in our experiments.
Chapter 3 presents the advantages of HC-PCF in nonlinear microscopy,
experimental results on pulse width measurement through HC-PCF using
frequency resolved optical gating (FROG), linear threshold power level
measurement of different length of the fibre and demonstrate the feasibility of
broadband excitation and collection in a single fibre based nonlinear
endomicorsocopic system using a piece of HC-PCF integrated with a GRIN lens.
Chapter 4 explains the advantages and importance of fibre-optic based super-
resolution endoscopy and advantages of the single wavelength implementation of
such a system. This chapter contains the experimental study on the generation of
fundamental and doughnut beams at different values of numerical aperture (NA)
and the characterization of the propagation of fundamental and doughnut beams for
different input polarization states through HC-PCF which is a crucial step towards
implementation of such a system.
Chapter 1
9
Chapter 5 starts with the review of fibre types and addition of different features
over time for use in nonlinear endoscopy since beginning. The problem of
chromatic dispersion and drawbacks of using pre-chirp units in nonlinear
microscopic system are explained. Chromatic dispersion compensation using a
higher order mode fibre is discussed and the necessity of new fibre design that
meets the requirement for both nonlinear imaging at near-infrared wavelengths and
its possible implementation in fibre-optic based STED technology requiring the
operation of two color wavelengths is presented. The mechanism to achieve zero
chromatic dispersion within the fibre is explained. The complete design and the
nonlinear parameter calculation are given along with simulation results for the
optimization of fibre design. Features of the newly designed fibre and the potential
use in fibre-optic super-resolution imaging system are discussed.
Chapter 6 concludes the thesis with the future work.
Chapter 2
10
Chapter 2
Background and literature review
2.1 Nonlinear microscopy
The fundamental physics of light matter interaction is utilized in optical
microscopy for imaging. The induced dipole moment by the vectorial electric field
during the light-matter interaction is given by [28],
.......
3)3(
2)2()1( EEEP
, (2.1)
where )(i is the i
th order nonlinear susceptibility tensor. The first order
)1( is
responsible for absorption and reflection of light in materials. Linear microscopic
systems like wide field microscopy and confocal microscopy utilizes the first order
interaction of light with imaging samples. The second order )2( is responsible for
Chapter 2
11
nonlinear optical phenomena like the second harmonic generation while the two-
photon absorption, the third harmonic generation and coherent anti stroke Raman
scattering are due to the third order susceptibility )3( . Nonlinear optical
phenomena use multi-photon processes so the corresponding optical microscopy is
called multi-photon microscopy. The imaging techniques that use the nonlinear
interaction between light and matter are two-photon excited fluorescence (TPEF)
microscopy, second harmonic generation (SHG) microscopy, third harmonic
generation (THG) microscopy, coherent anti-stokes Raman scattering (CARS).
Apart from imaging application, other non-imaging applications of nonlinear
excitation that are used in biological research are multi-photon fluorescence
correlation spectroscopy (MP-FCS), multi-photon cross correlation spectroscopy
(MP-FCCS) [29-32].
Fig1.1 Diagram illustrating nonlinear interaction of light with matter [33]
Chapter 2
12
Development of fluorescent probes over the period of time plays a major role in
bringing the microscopy to the current state of development though both
endogenous and exogenous fluorochromes can be used in bio-imaging [26-27].
Current state-of-art fluorescence optical microscopy utilizes optical nonlinearity as
well as nonlinear response (saturation of fluorescence emission rate with intensity
of light) of fluorescent probes. Recently developed microscopes are capable of
providing information about subcellular structures, biophysical and biochemical
phenomena.
Among other available techniques, TPEF microscopy is the mostly used nonlinear
bio-imaging technique for a high quality thick tissue imaging in vitro/ in vivo. The
possibility of the two-photon absorption was shown theoretically in 1931 by Maria
Goppert. The first two-photon fluorescence microscopy is demonstrated in 1990 by
Denk et.al [34]. For two-photon absorption to occur, two-photons have to interact
with other particles within ~10-16
seconds. This gives two-photon absorption event
a quadratic relation to light’s intensity. This intensity square dependence gives the
three dimensional localization of the excitation as the excitation probability outside
the focal region falls off by the fourth power with distance along optical axis. Two-
photon imaging is superior to single photon imaging owing to the limitation of
absorption to small focal volume [35]. This confined absorption has the advantage
of inherent 3D optical sectioning effect, reduced out of focus photo-bleaching and
Chapter 2
13
improvement in signal to noise ratio in thick tissue imaging [36-39]. The use of
long wavelengths also gives deeper tissue penetration and less absorption being
closer to the optical transparency window of tissues [7, 40].
2.2 Optical property of tissue
It is very important to understand the detailed optical property of specific tissue
types under investigation for imaging and other light assisted treatments [40]. In
general, biological tissues are relatively more transparent over the optical window
of 1000 nm - 1400 nm. In animal tissue Haemoglobin, lipids, water is the main
absorber at the near-infrared (NIR) wavelengths range. Deoxyribonucleic acid
(DNA), haemoglobin, lipids, amino acids, structural proteins are the main absorber
at Ultraviolet (UV) and visible part of the spectrum. DNA and proteins are
generally auto-fluorescence under UV and visible excitation. There is about 30%
protein in soft tissues. Within 10 micron depth all the portion of excitation
wavelength at 6-7 micron will be absorbed in protein due to the excitation of amid
bond in this wavelength range. Scattering is another important phenomenon which
is mainly due to in-homogeneities of refractive index in tissue because of the
presence of heterogeneous mixture of molecular structures. The strength of
scattering of light in tissue depends on the size and mixture proportion of different
molecules. Scattering is termed elastic if the photon energy remains unchanged
Chapter 2
14
after deflection. Elastic scattering phenomena like Rayleigh and Mie are the major
scattering process in tissues. Rayleigh scattering is predominant phenomena for
particle size much smaller than wavelength. It is nearly isotropic scattering and
wavelength dependent.
Raleigh scattering from dipole scatter much small than the wavelength of light is
given by [41],
),cos1(8 2
24
24
0
R
NII (2.2)
where I is the Intensity of light scattered, R is the distance from the scatter, N is
the number of scatters, α is the polarizability, θ is scattering angle, is wavelength
of light .
Mie scattering is a major phenomenon for particle size comparable to wavelength
of light. It is not strongly dependent on wavelength and is forward directional
scattering. Cell components like mitochondria, lysosomes, vesicles, collagen fibrils
are responsible for Mie scattering in biological tissue while cell membrane are
responsible for Rayleigh scattering in a cell.
Different scattering parameters for Mie scattering in tissue are
Scattering cross section ( s) is given by
s= QsAs , (2.3)
where Qs is scattering efficiency , As is the area of scattering particles.
Chapter 2
15
Scattering coefficient ( s ) is given by
sss NQ , (2.4)
where sQ is the scattering efficiency , sN is the number density of scattering
particles.
Strength of scattering is a measure in terms of mean free path ( ls ) which is average
distance between scattering events. For cell size comparable to the wavelength
scattering is mostly directed in the forward direction. Scattering mean free path ( ls)
is given by inverse of the scattering coefficient. Anisotropy (g) is a measure of
forward direction retained photon after a single scattering event with θ as the
scattering angles of photon. Parameter Qs and g are dependent on the size of the
cell structures, the index of the cell and it’s medium.
Reduced scattering coefficient is given by,
ss g )1(' , (2.5)
This coefficient is the measure of cumulative effect of forward scattering events,
incorporating the scattering coefficient and anisotropy together. Other parameter of
importance for imaging in tissue is the physical path length and optical path length.
The ratio of the optical path length to the physical path length is 4 or greater than 4
according to the type of tissues. The propagation of the photon through tissues can
be simulated with the knowledge of these parameters using Monte Carlo, transport
theory or diffusion theory [42, 43]. Modeling of optical properties of soft tissues
Chapter 2
16
like in visceral organs are given in ref. [41], shows that refractive indices of all the
constituent cellular components should be considered to get accurate optical
properties of the tissue under consideration.
Various diagnostic and treatment process have been developed utilizing the change
in absorption, scattering by tissues for example the amount of scattering from tissue
may vary due to diseases conditions for example, reduction of the Nicotanimide di-
hydrogen phosphate (NADH) concentration in tissues is one of the indications of
the tumour tissue as there is consumption of a large amount of oxygen. The reduce
in auto fluorescence level from NADH in this situation has thus been utilized for
early detection of cancerous tissue. Laser assisted in situ keratomileusis (LASIK)
utilize the absorption of light source in the tissue as the primary mechanism for
treatment.
2.3 Fluorescence collection and imaging depth
The selection of wavelength for excitation of given fluorophore/s determines
penetration depth in nonlinear imaging. Fluorophores are chosen according to their
two-photon cross-section. It is shown that scattered infra-red photons are not
effective to excite fluorophores in TPEF systems [44]. This is an advantage in
Chapter 2
17
using long wavelengths as they are relatively less scattered, which can increase
excitation and may compensate the need of more photons for two-photon excitation
(TPE) at long wavelengths. For nonlinear optical microscopy like two-photon
imaging only non-scattered light from the focal volume is considered as signals.
The signal power follows Lambert-beer like exponential decline with imaging
depth z. Two-photon signal from depth declines as [33]
,)(/22
2ss lzl
z
PE eeF
(2.6)
where ls is the scattering length, F2PE is the fluorescence signal at depth z.
Available power at desired wavelengths and the pulse width are key parameters to
be considered in choosing laser system for nonlinear imaging. Almost all biological
tissues scatter light strongly but with localized nonlinear action in two-photon
absorption processes, multiply scattered signal photons can be assigned to their
origin providing image from several hundred microns deep of living tissue.
In TPE fluorescence imaging optimization of signal collection and detection
efficiency is the most important factor. For scattering tissues, ballistic fluorescence
photon becomes negligible beyond a few tens of microns deep from the surface so
the efficient fluorescence detection requires large field of view to collect scattered
light from the focal volume [32, 45]. Large field of view objective has low
magnification thus high numerical aperture (NA) with low magnification will have
particularly high collection efficiencies [44]. For TPF imaging from the deep
tissue, detection systems should be designed to collect the diffuse light.
Chapter 2
18
Fluorescence F generated in the focal plane decreases proportionally to the square
of the fraction of the ballistic excitation photons,
,)/exp(.
2
0
ex
slzPF (2.7)
Maximum imaging depth in TPEF microscopy is given by
),)(ln( max0max zP
TPlZ
ex
s
(2.8)
where is the collection efficiency of the system, Zmax is the maximum imaging
depth, ls is the scattering length for excitation in tissues, P0 is the average power
incident on the surface of the samples, α is the setup parameter that depends on
fluorophore properties as well as detector and shot noise, is the pulse duration , T
is the pulse period.
The probability of the number of absorption of the two-photon pair ( na) is given
by [34].
,2
)()(
222
2
2
excPp
avg
excpahc
NA
f
Pgn
(2.9)
where gp is a unit less factor that depends on the temporal laser pulse shape ( 1 for
rectangular, 0.66 for Gaussian , 0.59 for hyperbolic-secant-squared ),
)(2 exc is
the molecular cross-section, Pavg is the average power, NA is the numerical
aperture, h is the plank’s constant, c is the velocity of light in space, p is the
pulse duration, fp is the repetition rate. The extent of two-photon fluorescent
excitation (TPFE) can be manipulated by altering parameters like pulse width,
Chapter 2
19
pulse repetition rate, pulse shape and selection of suitable fluorophores and their
excitation with given NA of optics.
Two-photon excitation (TPE) cross-section for picoseconds and continuous wave
(CW) excitation beam are larger than TPE femto-second cross-section but local
heating effect of samples is the problem for considering pico-seconds and CW laser
to be used for TPE fluorescence imaging [39,46]. Fluorescence intensity If in two-
photon excitation of fluorophores depends on the pulse width and the pulse
repetition rate and hence the average power of the laser beam used. Linear region
in log – log relationship between the input power and the signal power is the
verification of occurrence of the two-photon excitation in fluorescence samples
during measurements.
Long wavelengths have the advantage of deeper tissue penetration and less
scattering in tissues but the diffraction limited spot size is comparatively bigger
than that for short wavelengths. Combination of super-resolution techniques and
deep tissue imaging using longer wavelengths has the possibility of deep tissue
super-resolution for 3D imaging from thick tissue biological samples with suitable
far-red fluorophore labelling.
Chapter 2
20
2.4 Optical resolution
Optical resolution of light microscope is its ability to distinguish the two point
objects in close proximity [47]. This is fundamentally limited by the wave nature of
light. A point object will be projected as blur spot of finite size by optical imaging
system called point spread function (PSF). Two points closer than full width half
maximum (FWHM) of the PSF will be difficult to resolve because of the
substantial overlap in their image. Resolution is given by ∆(x,y)= /(βn sin(α) in the
lateral direction and ∆z= β /n sin2(α) in the axial direction, where is the
wavelength , n is the refractive index of medium and α is the half angle of the
imaging system. With best optics for bio-imaging, this value comes between 200
nm - 300 nm laterally and 500 nm - 700 nm axially. It is not possible to directly
visualize the sub-cellular structure smaller than this dimension. Fluorescence
microscopy combines the molecule specific labelling technique with light
microscope for the direct visualization [48-53]. Nonlinear interaction of light with
matter is utilized in nonlinear fluorescence microscopy which gives the capability
of inherent optical sectioning for 3D imaging, deep tissue penetration with
improved signal to noise ratio and slightly improved resolution [54-56]. Further,
nonlinear response in time and space of the different fluorescence labels are
utilized in recently developed microscopy techniques enabling one to achieve
Chapter 2
21
resolution above by one order of magnitude over conventional microscopy
techniques. Factor of √β in spatial resolution and improvement in signal to noise
ratio by rejecting out of focus fluorescent background was achieved using the
pinhole detection scheme in confocal microscopy. Reduction in point spread
function (PSF) is achieved by nonlinear excitation in nonlinear microscopy. Near
symmetrical PSF was achieved by increasing effective system NA by using two
opposing objectives for excitation and detection in 4Pi and I5M microscopy.
Structured illumination microscopy uses higher harmonics in illumination scheme
to increase image resolution. These different approaches helped to extend the far-
field optical resolution but they are fundamentally limited by the wave nature of
light. Recently the nonlinear saturation of fluorophore is utilized to reduce the PSF
beyond the diffraction limit. It is demonstrated that with the saturation of emission
rates of fluorophores one can create emission distribution with sharp edges and
strongly confined non-emissive region that was not possible in linear systems. The
saturation of the upper transition state of fluorophores faster than its recovery is
achieved by high intensity laser beam matching the difference in energy of this
electronic transition. Exact shape of the emission distribution is manipulated by the
distribution of intensity. The most popular use of this technique is stimulated
emission and depletion (STED) microscopy which uses high intensity doughnut
shaped beam to limit fluorophore emission around the periphery of the diffraction
limited emission spot at the centre formed by another laser. Similarly the
nonlinearity in time is utilized for high resolution by particle localization methods.
Chapter 2
22
It is now possible to calculate the exact position of single emitters with their
centroid of emission pattern in samples with high labelling density. To get the
detailed structural information, the precise control of emission of individual
fluorescent labels within the diffraction-limited volume and separating their
emission in time are used to construct the high-resolution image. Using these
recently developed high resolution techniques it is now possible to uncover the
biological process at subcellular and molecular level but these techniques are only
available in bench-top systems in a research laboratory environment.
2.4.1 Super-resolution techniques
Far-field optical technique like fluorescence microscopy can now achieve higher
resolution which is comparable to near field microscopy like near field scanning
optical microscopy (NSOM) that can provide 20 nm-50 nm resolution. Three-
dimensional imaging with an optical resolution, as high as ~ 20 nm in the lateral
direction and 40-50 nm in axial dimension has been achieved by far field
microscopy [57, 58]. There are a number of super-resolution techniques currently
available in far field microscopy which can in general be categorized into methods
that sharpen the PSF of fluorophores using nonlinear responses of fluorophores
such as in stimulated emission depletion microscopy (STED), reversible saturable
optically linear fluorescence transitions (RESOLFTs), structured illumination
microscopy (SIM), saturated structured illumination microscopy (SSIM), nonlinear
Chapter 2
23
structured illumination (NSI), selective plane illumination microscopy (SPIM),
digital laser scanning microscopy (DLSM), or that localize the individual
fluorescent molecules such as in photo activated localization microscopy (PALM),
fluorescence PALM(FPALM), stochastic optical reconstruction microscopy
(STORM), fluorescence imaging with one nanometre accuracy (FIONA) [47-53,
59-74] . Some of these techniques are slightly different from each other and some
of them also use one or more techniques in combination to achieve high resolution
[75-77].
In brief, their working principles are described as follows. In SIM, samples are
illuminated with light pattern of different spatial frequencies (produced by using
gratings), that captures high frequency information from features sizes smaller than
wavelength. Fourier signal post-processing techniques are applied to shift this high
frequency information to lower frequency within diffraction limit. SIM can only
achieve a factor of two improvements in resolution below the diffraction limit. In
SSIM, which uses SIM with saturation of fluorescence emission, the resolution is
not fundamentally limited by the diffraction but by the level of fluorescence
saturation practically achievable. SSIM demonstrated 50 nm resolution in two
dimensions [78]. Nonlinear structured illumination (NSI) uses harmonic generation
from the saturated excitation state of the fluorescent in samples and mathematical
processing of these harmonics. NSI may not be suitable for live imaging as
described by the author [78].
Chapter 2
24
The imaging depth as well as resolution can also be optimized by simultaneous
temporal and spatial focussing of pulse. Good spatial and temporal focussing of
pulse used as described in ref. [79] can be achieved by using 4Pi technique that
uses two divided beams from a point source which are focussed at a point using
identical and opposing high NA objectives [80]. At common focal volume,
resulting illumination PSF is given by two coherent beam interference. The
fluorescence signal is collected and focused to a point like detector. 4Pi technique
achieve 4 times narrower central PSF maximum than that of conventional PSF in
confocal microscopy, but this narrowing is accompanied by side lobes which limit
resolution in such systems. Post processing techniques to delete these side lobes are
required during reconstruction of the 3D images. With simple de-convolution
algorithm as described in [81] axial resolution as 0.5 micron can be achieved using
two-photon excitation with 4Pi confocal microscopy setup. Spatial resolution also
depends on labelling intensity of fluorophores (ls = Vs/V) and the probe size and the
preservation of ultra-structures during the sample preparation. This technique
cannot be used in live cell imaging and is possible only for bench top SHG setup.
Fluorescence image is formed by spatial co-ordinates of fluorophores thus
determination if position of each fluorescent probe molecules in samples with high
precision form the basis of super-resolution by single molecule imaging. A single
fluorophore is visualized by its diffraction limited blur in its image but the position
of the fluorophore can be precisely known by the centroid calculation of intensity
Chapter 2
25
distribution within PSF. In a well distributed fluorescent samples with the
assumption of a photon emission from each individual fluorophore, the
localization precision is given by [57]
Δ localization ≈ Δ / √ σ, (2.9)
whrere, Δ localization is the localization precision, Δ is the PSF size and N is the
number of measurement of fluorophores. For multiple molecules within the
diffraction limited proximity the separation of signal in overlaped regions in
spectral domain or time domain is necessary. Use of photo switchable fluorescent
dyes or proteins and activating them in different points of time each fluorophores
can be precisely localized to reconstruct the images during processing. In practice
resolution of ~ 20 nm has been achieved experimentally.
STED microscopy overcomes the diffraction barrier for resolution by reducing the
effective size of focus by switching off the fluorescence ability of the fluorophores
in the outer part of excitation focus. This switching of fluorophore is achived by
forced transition between electronically excited state (ON) and the electronic
ground state of the fluorophore(OFF) by incidence of a photon that matches the
energy gap between lower excitation state and ground state other than the
excitation laser. The spectral range of detector is chosen in such a way that it could
not detect the wavelength of the emitted photon after the quenching process.
Chapter 2
26
FWHM of the region from which the fluorescence can be emitted in STED with
first approximation is given by [82].
,1sin2
)(
sIIn
Ir
(2.10)
where is the STED wavelength, n is the refractive index of medium, α is the
semi-aperture angles of the objective lens and I is the intensity maximum of
engulfing zero, and Is is the characteristic intensity required for reducing the
fluorescence ability by a factor of two for dye being used. In STED, raising the
STED laser power the saturated depletion region expands. I/Is is much larger than
unity but RESOLFTs using photo switched fluorescent FP595 and dark state is
achieved at much less laser intensity. Focus engineering is necessary to a produce
doughnut intensity pattern for the quenching beam. Sample preparation and choice
of suitable fluorophores and their properties are one of the critical design
parameters for imaging in STED [83]. STED can also be used for live cell imaging
using auto-fluorescent proteins in tissues. Auto fluorophores like Yellow
fluorescent protein, blue fluorescent protein can be used to get signal along with
other organic dyes like Tetra-Methyl-Rhodamine (TMR), and process like enzyme
mediated labelling with Oregon Green, or self-labelling tag, for live STED
imaging. Atto 532 , Atto647N are the most used dyes for STED microscopy. Auto-
fluorescence intensity and the change in the spectrum measurement can distinguish
normal and various stages of cancer cells at different internal organs [84]. For
example , the diagnostic algorithm based on the combination of the fluorescence
Chapter 2
27
peak intensity ratios of I-350/I-470 at 280 nm excitation and I-390/I-470 at 330 nm
excitation yielded a sensitivity of 100% [95% confidence interval (CI) 0.95-1.0]
and specificity of 100% (95% CI 0.90-1.0) [85]. Similarly, suitable fluorophores
are needed to implement TPF imaging at 1200 nm wavelength and make such
comparative study of fluorescence intensity ratio at this excitation wavelength for
the diagnostic purpose. An auto-fluorescent protein called Neptune, which is far-
red fluorescent protein with excitation peak at 600 nm or above have been, reported
recently [86]; that may be used for intra-vital imaging in mammals with two-
photon excitation at 1200 nm.
2.4.2 Single wavelength stimulated emission and depletion imaging
Requirement to use two laser beams and associated complex optical design,
inflexibility in selection of dye and high cost and difficulty in incorporating them
into other commercial microscopy system are some of the technical challenges
remaining for wide use of bench-top single wavelength STED microscopy. Also
the use of two different wavelengths in this system makes the imaging of the thick
scattering tissue samples impossible due to different amount of chromatic
aberration in optics and tissues. The use of one wavelength both for two-photon
excitation and quenching of the fluorescence makes the optics optimization task
simple as well as the use of STED system possible for 3D imaging of thick
scattering tissue samples. Recent works [87-89] shows the feasibility of using
Chapter 2
28
conventional dyes and single wavelength bench-top STED microscopy with
reduced system complexity for super-resolution biological imaging.
2.5 Fibre-optic imaging systems
Different design trials to build ultimate compact systems for bio-imaging in vivo
environment are going on. Each of the systems built so far has some advantages
and limitations not found in others. The core concept is to develop miniaturized
optical endomicroscope/endoscope with high resolution, large working distance
and wide field of view with video rate 3D image acquisition capability to collect as
much information as possible from internal organs of living suspects [90-91].
Combining two or three imaging modalities like two-photon absorption (TPA),
SHG, Optical coherence tomography (OCT), CARS and wide range of
fluorophores for simultaneous multi-colour excitation) have also been tried to get
more information, at the cost of increased complexity of the device. Imaging
endoscope contains different components for beam and signal delivery and
scanning mechanism [30, 92-94]. Fibre and micro-optics techniques are used to
make the system compact and mult-functional at the same time [31]. A well
designed fibre-optic imaging probe can make it possible to access internal parts in
minimum invasive manner and are suitable for clinical diagnostic and surgical
Chapter 2
29
procedures, which otherwise would have been a major procedure . These fibre-
optic imaging systems with suitable probes have inherent flexibility to be used with
living suspects for the long-term imaging studies in various fields to understand the
detailed progressive development stages of different diseases and other natural
phenomena. In this direction research has been focus in last two decades and fibre-
optical systems are now considerably reduced in their system size and increase in
their functionality. Fibres in imaging devices are basically used for the remote
delivery of light and the collection of signals. Optical endoscopy systems with
endoscopes probe of very small diameters (few millimetres) are most suited to
image hollow tissue cavities within living body. Different fibre-optic fluorescence
imaging modalities have been used to date like single-fibre confocal, dual-axis
confocal, fibre-bundle confocal, two-photon fluorescence , single-fibre two-
photon, fibre-bundle two-photon, multi-focal two-photon, dual-clad two-photon
systems. They have different sets of advantages and limitations as described in ref.
[95].
Technical problems like group velocity dispersion (GVD), self-phase modulation
(SPM), self-steepening arise when fibres are used for high energy ultrafast pulse
propagation required for non-linear imaging. GVD is due to different spectral
components of a pulse travel at slightly different speeds within the fibre core which
lead to the temporal broadening of the pulse. SPM is a result of the intensity
dependent refractive index of the fibre material. Self-steepening is a non-linear
Chapter 2
30
effect arising due to the intensity dependent group velocity. The primary cause of
temporal pulse broadening in fibre based system is GVD but with the increase in
the pulse intensity, non-linear broadening of pulse will also occur.
2.5.1 Optical fibres used in endoscopy
The step index optical fibre with an inner core of radius ‘a’ has 1 or 2% higher
index given as ‘n1’ then the outer cladding of lower index given as ‘n2’ and light
is transmitted in them by total internal reflection. Numerical aperture (NA) of the
fibre is given by
,)( 2/12
2
2
1 nnNA (2.11)
Optical fibres are characterised by their normalized wave number V given by
V= (βπ a σA)/ , (2.12)
Fibre supports single mode (LP01) if it satisfies inequality condition V< 2.405 or
few mode (LP01 and LP02) if V< 3.8317 (for step index profile) and multimode
else. Multi-mode fibre (V> 2.405) guides more than one spatial mode. Transmitted
modes in the fibre increase approximately quadratically (as a function of V) with
the increase in core the diameter ‘a’. Single mode fibre (SMF) deliver light
efficiently and also acts as a pinhole detector rejecting out of focus fluorescence
emission in two-photon fluorescence imaging systems [8].
Chapter 2
31
Photonic crystal fibres are special optical fibres and have periodic arrays of air
holes in silica glass medium. A photonic crystal fibre is superior to others due to its
capability to endless single mode operation, dispersion engineering and high non-
linearity [96-97]. Hollow-core Photonic band gap fibre (PBF) guides wavelengths
within the transmission band in the central air core. There is no problem of SPM in
this fibre as there is no material in the core. PBF does not rely on total internal
reflection for guiding light. PBF uses diffractive effects with wavelength scale air-
silica arrays. Periodic arrays of air-silica lattice create photonic band gap which
prevents light at certain wavelength range to propagate in the cladding. These
wavelengths are localized in core and transmitted. However higher mode dispersive
pulse broadening can be more severe than in conventional fibres so for fibre
lengths greater than few meters needs additional dispersion compensation for
higher order modes.
Another class of fibre with a silica core and air-silica micro-structures as cladding
called large-mode-area photonic crystal fibre (LMA-PCF). It has large core
diameter up to ~35 m. SPM effect is reduced and light guidance is ‘endlessly
single mode’ in such fibres. Higher mode dispersion is less severe in these fibres
than in hollow-air core fibres.
A double-clad fibre (DCF) has two claddings apart from the central core. They are
especially important in single fibre endoscopes as they can be excited through the
Chapter 2
32
core and collect signals using cladding with high NA. This feature has improved
the detection efficiency in nonlinear imaging modalities by two order of magnitude
compared to nonlinear imaging using single modes. Double-clad photonic crystal
fibre (DCPCF) has large-mode-area silica core, periodic array of air-silica micro-
structure with low NA as the inner cladding and solid silica the outer region as
outer cladding with high NA. This kind of double-clad photonic crystal fibre is of
particular interest as low NA cladding can be used for guiding ultra-short excitation
pulses with reduced SPM and the outer cladding with high NA can be used for the
efficient collection of fluorescence signal. Outer multi-mode silica-air interface
can have NA around ~0.6 for the visible light. Apart from dual excitation and
collection a double-clad photonic crystal fibre can be formed into a coupler which
can be used to separate the visible fluorescence from the infrared excitation. Fused
tapering method can be used to fabricate double -clad PCF couplers [13].
Table 2.1: Comparison of DCF and DCPCF features suitable for fibre endoscopes
SN Parameters
DCF DCPCF
1 Core size 3.6 micron 16 micron
2 Inner clad 125 micron 135 micron
3 Outer clad 250 micron 350 micron
4 NA core 0.19 0.04
5 NA clad 0.21 0.62
Chapter 2
33
6 Coupling efficiency
laser to fibre
91% 82%
7 Power at core 70% 38%
8 Degree of
polarisation
0.95 @18mW
0.78 @90mW
0.98@50mW
9 Endoscopic system
with lens at fibre tip
Focal length :740 micron ;
spot size : 9um @1310nm;
CSF(lf)=450micron
curvature 90micron
Field of view 2.5mm X 2mm
Probe size: 0.4mm
Lens NA 0.12;
Spot size : 6micron
@800nm
10 Cost wise less costly costly
11 Signal collection comparatively low efficiency
due to low NA
high efficiency due
to high NA
12 Mechanical property robust not so robust ( major
drawback for fibre
scanning imaging
systems)
13 Dispersion
has high dispersion (GVD :
45,127 fs^2/m @805 +/-
4.5nm ) over 60nm
bandwidth )
has low dispersion(
GVD of Photonic
band gap fibre (PBF)
-14622 fs^2/m)
Chapter 2
34
14 Coupler Coupler with polymer inner
clad (contact length 16cm)
have achieved 30% clad to
clad coupling of the visible
signal.
Coupler permits
separation of single
mode excitation and
multimode visible
signals at the
detection arm.
15 Consideration for
excitation at 1200nm
Long excitation wavelength
needs more power for
excitation and DCFs
effectively excite
fluorescence due to more
power carried in core.
16 After dispersion
compensation
In DCF de-polarisation effect
can be improved
A single cladding micro structure fibre with a small silica core yielding highly
nonlinear behaviours with interaction to light is being used positively to harness
SPM for broadening the spectrum of ultra-short pulses for the super-continuum
generation.
Chapter 2
35
Fibre bundles are also used in imaging. The fibre bundle consistng of up to
~100,000 individual step-index fibres are closely packed maintaining relative
arrangement through the length. They are able to transmit intensity image in
pixelated form. Laser scanning can be achieved at the distal end of samples by the
sequential illumination of the individual fibre in the bundle which can be taken as
advantage but reduction of lateral optical resolution (calculated as core-to-core
distance divided by optical magnification) is the main disadvantage of such system.
2.5.2 Endoscope probe components
SMF gives severe temporal and spectral broadening for ultra-fast laser pulses.
Their NA is low (~0.1) for the signal collection, which is far from wavelength
range mostly used for excitation in a nonlinear imaging system. Multimode fibres
could be used but they also lack the capability to focus to a diffraction limited spot
on samples for the efficient excitation without use of special techniques or other
micro-optics like use of gradient index (GRIN) lens. The use of fibre could make
the system compact but can’t achieve beam scanning without using other beam
scanning components like micro-electro mechanical system (MEMS). There are
ways to manage dispersion and reduce nonlinearity by the use of LMA-PCF or
hollow-core PCF for excitation alone but equal design consideration is needed for
back collected visible fluorescence signal. DCPCF is a good candidate for
Chapter 2
36
optimization of both excitation and collected signal. DCPCF has LMA core for
single mode excitation with reduced nonlinear effects, microstructure inner
cladding for multi-mode guidance of visible signals. The inner cladding is
separated by the web of silica bridges that are small in dimension than the
wavelength of light guided in the inner core. GRIN lens are used to focus the beam
from the fibre to a focus spot on samples.
GRIN lenses are sub-millimetre in size. GRIN lens uses a variable concentration of
dopant in glass to achieve characteristic parabolic refractive index profile given by,
)2/1()( 2
0 Arnrn
(2.13)
where n0 is the refractive index on the centre axis, r is the distance from the central
axis. GRIσ lens has no curvature to achieve focus instead it’s nearly parabolic
refractive index profile gives focussing effect if cut in proper size. The period of
the sinusoidal path (pitch of the lens is given by
2/LAp , (2.14)
where L is the lens length, √A is the gradient constant.
Adjustment of the distance between the fibre and GRIN lens and choice of pitch of
the GRIN lens in system are variables for achieving variable working distance and
focus spot size [94, 95]. Effective NA of the system and hence the optimization of
the two-photon fluorescence excitation in a single fibre based imaging system is
determined by the distance between three port coupler and the GRIN lens with
different pitch values [ 13].
Chapter 2
37
Fig. 2.2: Refractive index and image profile of GRIN lens [13].
For GRIN lens immersed in a refractive index medium nm( m= 1,2 ) , a laser beam
emerging from the fibre at a distance d1 to the GRIN lens surface can be focused to
an image distance d2 given by,
)cos()sin()sin()/()cos( 101
2
0211202 ALnnALAdnALAnnALdnnd , (2.15)
where n0 is refractive index on central axis , n1 & n2 are refractive index of the
medium, L is the length of the GRIN lens, d1 & d2 are object and image distance
respectively from GRIN lens surface, √A is the gradient constant .
on the opposite side of the GRIN lens. GRIN lens can have NA in range 0.46-1.
Recently the formation of lens on the fibre tip by the collapse of air holes has been
used to acquire images using nonlinear microscopy [98]. Use of SMF couplers,
which act as a low pass filter for signal in the visible wavelength region and
provide inherent confocal pinhole, has made it possible for all fibre nonlinear
imaging system. Photonic crystal fibre (PCF) couplers with double-clad fibre have
also been implemented in nonlinear optical endoscopy [17, 99]. Recently super-
continuum light for excitation has been generated within fibres using highly
Chapter 2
38
nonlinear property of fibres enabling simultaneous CARS and TPEF images [100].
DCPCF preserves linear polarisation in core but there is depolarization effect in the
inner cladding region due to micro-structures in large core area. So the
development of polarization states maintaining fibres is needed to implement SHG
imaging using double-clad photonic crystal fibres [101].
To implement second harmonic generation phenomena for imaging using fibre –
optics, the fibre and fibre couplers used should have polarisation preserving
characteristics. Measure of polarisation preserving feature is given by
degree of polarisation ( ) defined as
)(
)(
minmax
minmax
II
II
, (2.16)
If the value of is close to 1 polarisation at every 90o shows birefringence effect of
the fibre. Linear polarisation in fused single mode fibre couplers is preserved at
certain incident angles for both CW and pulsed illumination [65].
DCF exhibits a low nonlinearity and self-modulation effect. DCF is also found to
have degraded two-photon signal levels and polarisation of signal if used in SHG.
Instead DCPCF has higher threshold of the nonlinearity and higher degree of
polarization states maintenance. DCPCF is found superior over DCF for SHG
imaging and two-photon imaging [24]. Compact two-photon fluorescence
microscope based on single mode three port fibre coupler has been demonstrated.
Chapter 2
39
In such a system coupler behaves as low-pass filter that can deliver an ultra-short
pulsed laser beam with 38% coupling efficiency in the infrared range (770 to 870
nm) as well as collect fluorescence signal in the visible range at low (1%) coupling
efficiency with single mode profiles [8].
DCPCF coupler can be used to separate the visible fluorescence from infrared
excitation. Two lengths of DCPCF are twisted and heated by hydrogen flame with
flame size of approximately 10 mm and then draw gradually in the fused region.
The elongation length determines the splitting ration of the coupler formed and
mode coupling in the core and clad region. If the elongation length is longer there
will be multi-mode in both arms of the coupler so optimization is needed in the
fabrication process. Axial resolution and signal level is dependent on the gap
length between the fibre coupler and the back surface of the GRIN lens used as in
two-photon imaging system using the single mode fibre coupler and GRIN lens.
Axial resolution of 10 m was obtained in a two-photon imaging system with the
DCPCF and the GRIN lens [99].
Piezo-electric driven 2D-tilt mirrors near the samples are used to scan beam from
single fibre in portable microscopy. Micro-electro mechanical system (MEMS)
mirrors are another rapidly emerged technology that could be actuated electro-
statically or electro-thermally. The size of MEMS size ranges from 0.5-2 mm and
gives up to 30 degree angular rotation with a low voltage. Their fabrication is
Chapter 2
40
complex process involving sequential material etching and deposition. Probe is the
major component in endoscope imaging system. There are many designs of probes
available in literature with various functionalities [93, 102-103]. Figures in ref. [93,
102-103] show the mostly used and the most robust recent design. Resonant
scanning type probe has both fibre and imaging lens resonating together attached to
a cantilever to reduce possible aberrations. Because of the mechanical robustness
drawback of DCPCF, it cannot be used in such a resonant scanning probe system.
In Dual-axis confocal imaging system one SMF delivers excitation light and a
second SMF mounted at angles collects fluorescence from the overlapping region
of the two fibre apertures. To achieve variable focus, liquid lens that can alter the
focus on electro wetting are also tested.
2.6 Ultra-fast pulse propagation
The understanding of the propagation of ultra-fast pulse near-infrared wavelengths
in optical fibres is critical to successful implementation of fibre-optic imaging
systems. For ultra-short pulse propagation in optical fibres we have to consider
both materials and waveguide dispersion. The material dispersion effect comes
from the frequency dependent refractive index of the fibre material, silica in our
case. It can be approximated by Sellmeier equation.
,13
122
2
2
i i
i
B
An
(2.17)
where, Ai and Bi are Sellmeier coefficients and is the wavelength in micron.
Chapter 2
41
The waveguide dispersion is determined from the propagation constant ( ) of the
optical pulse, which is given by
( )=n(ω) ω / c , (2.18)
where ω is the frequency of the pulse and c is the velocity of light. The dispersion
coefficients for the pulse propagation are obtained by its Taylor’s series expansion
around carrier frequency ω0 given by � = + � − � + �! � − � + �! � − � + ⋯, (2.19)
where, j is the mode number. Coefficient 1 is related to the group velocity of
pulse in the fibre by = /�� where ng is the group index. The second order
coefficient β represents the second derivative of propagation constant with respect
to the wavelength i.e dispersion of group velocity and is responsible for the pulse
broadening. β has positive sign in normal dispersion regime and has negative sign
in anomalous dispersion regime. The waveguide dispersion in optical fibres is
given by � = − � � , (2.20)
The coefficient γ in the equation is called the third order dispersion (TOD). The
effect of TOD becomes important near zero dispersion of the wavelength in the
design. Pulse propagation in optical fibres is described in terms of two
characteristics length name as dispersion length (Ld) and non-linear length (Lnl). � = ∣ ∣ , ��� = � , (2.21)
Chapter 2
42
Prominent effects during the pulse propagation are described in terms of the ration
of these lengths. When dispersion length is much shorter than the nonlinear length,
dispersion is prominent in the fibre. Similarly when the nonlinear length is much
shorter than the dispersion length the pulse propagation is nonlinear which doesn’t
affect temporal profile of the pulse but result in spectral changes in the pulse.
For pulses with a pulse width (To ) less than 1ps, which is the case for nonlinear
imaging using two-photon , high order nonlinear effects up to eleven have to be
consider in the Taylor series expansion and the simplified pulse propagation
equation with the normalized amplitude U is given as [104]
� + �� � � = ��6��′ � + −��� � |�| � + � � |�| � − ��� | |� , (2.22)
where, �′ = | | , � = 0 , �� = �0 is the absorption coefficient of the Silica.
The nonlinear parameters are calculated for our new fibre design in chapter 5,
which are required for the ultra-short pulse propagation in the fibre. Split step
Fourier method is used for the pulse propagation simulation. The split step method
calculates different values by splitting total length in small steps where dispersive
and nonlinear effects are assumed to act independently.
Chapter 2
43
2.7 Dispersion in optical fibres
When the short pulse passes through dispersive materials, the pulse width stretch
but the spectrum remains unchanged. In visible and NIR region, all materials have
positive dispersion (red frequency leads blue frequency). Dispersion is given in
unit “fsβ” (or for optical fibre in ps/km-nm). Positive dispersion can be offset by
adding negative dispersion using prism or grating pair in the path of the beam.
Dispersion in the optical path is the major problem in excitation and signal delivery
when using fibres as media. Pulse broadening resulting from the dispersion can be
estimated using
,))/(68.71( 2/122
inDinout (2.23)
where D is the total dispersion in femtoseconds squared. Typical bench-top multi-
photon microscopy system has 5000 fs2. In silica based optical fibre material
dispersion is around -70 ps/nm-km @ 800 nm and around -20 ps/nm-km @ 1200
nm. Increase in pulse width by dispersion decreases peak laser power on samples.
Total dispersion has different components. One of them is material dispersion,
which is due to the different values of refractive index imposed by the material for
different wavelength components of the pulse. The difference in time for the
different spectral components to reach the other end of the fibre i.e pulse
broadening is given by
Chapter 2
44
,0
0
2
0
22
0
d
nd
c
L
(2.24)
where n is the refractive index, c is the velocity of light, L is the length of the path ,
0 is the wavelength. Material dispersion (Dm) is given by,
,0
L
Dm (2.25)
Unit for material dispersion in optical fibre is given in ps/km.nm. Imperial relation
to calculate the material dispersion is given by Sellmeier’s equation (2.17).
Another dispersion mechanism that is present in single mode fibres defined by its
V number as given in equation (2.12) is the waveguide dispersion. Even in the
absence of material dispersion group velocity of the spectral component depends
on frequency (ω). The pulse broadening given by
,.)( 0
0
2
2
2
V
bVdV
dn
c
Lw
(2.26)
and the waveguide dispersion is given by
],)(
[2
2
21
dV
VbdV
nnD
c
w
(2.27)
The waveguide dispersion parameter inside the bracket of expression for Dw in
fibre can be obtained by the empirical expression given by following equations,
assuming weakly guiding condition.
Chapter 2
45
,)834.2(549.0080.0
)( 2
2
2
VdV
VbdV
(2.28)
both dispersion is given in combination by the expression
D = ,2
2
2
2
d
dc (2.29)
The fibre dispersion slope coefficient (S) is given by
S = D/∆ , (2.30)
The normal dispersion in the fibre and other optical components can be
compensated by the anomalous dispersion obtained by the grating pair for pre-
chirped pulse [105-107]. The problem in use of these gratings is the excitation
power loss reducing available power at the samples.
Fibre Bragg grating (FBG) can also be used for the dispersion compensation within
fibre but that will not serve purpose in imaging system due to the high reflection
[108]. Even the small amount of reflection could cause imaging artifices and not
permissible in biological imaging system. But for the case of transmission grating
like long period grating back reflection from grating is negligible.
2.7.1 Long period fibre gratings
Long period gratings (LPGs) are used as optical filters, gain flatteners, lens free
fibre to fibre coupler and for the dispersion compensation in optical communication
Chapter 2
46
systems apart from their use in different sensors [109-113]. They are used for light
coupling between two co-propagating modes in an optical fibre or wave guide.
LPG has pitch range in several micrometres to several millimetres range. Grating
created by acoustic wave along the fibre [114] or grating created by periodic stress
[115] have also been used for coupling light between two guided modes. In these
gratings, there exists a specific resonance wavelength at which the coupling
between the guided modes is the strongest.
Fibre gratings formed in the core of fibre by periodic modulation of refractive
index section that convert the fundamental mode present in the fibre into higher
order mode ( LP11 or LP02 ). LP01 to LP11 mode conversion is complicated because
of asymmetrical modes. However, LP01 to LP02 mode converter involve single
resonance being circularly symmetric mode for LP02. The spatial periodic (Λ)
modulation of refractive index in fibre core matches to inter-modal beat length
between the mode to be coupled given by the phase-matching condition.
,2
21
(2.31)
where 1 , 2 are the propagation constants of the two coupled modes.
At the wavelength where mode conversion occurs the optical power of the mode
LP01 is transferred to LP02 mode. With the decrease in power mode conversion
efficiency can be calculated in a transmission spectrum. The value of required
Chapter 2
47
refractive index difference ∆n for grating and fibre core radius rc for few modes in
the fibre are crucial design parameters.
LPG can be fabricated using femto-second laser beams in the core of fibres (which
are doped with Ge ) by inducing index modulation achieved by Ge and O2 bond
(defect) breakage [116-119].
2.7.2 In Fibre dispersion compensation using LPGs
The major problem of implementing all fibre based imaging system is the
dispersion compensation in fibres for femto-second laser pulses. In current
nonlinear imaging systems, prism and gratings pairs are used for the dispersion
compensation. It has been shown that normal dispersion in the fibre can be
compensated by anomalous dispersion achieved with long period grating in fibres
for higher order modes to make net zero dispersion in fibres for the wavelength
used. Nonlinear microscopic imaging system currently utilizes 800 nm as
excitation wavelength. Shifting the excitation wavelength to 1200 nm will increase
the penetration depth in tissues, but the system should be designed carefully within
available average power of laser sources to get two-photon fluorescence signals, so
dispersion issue will be more critical.
Chapter 2
48
There are methods for the dispersion compensation by converting the fundamental
mode to higher order modes and back to fundamental modes by using long period
grating at telecommunication wavelengths. There has been not much use of long
period gratings for the dispersion compensation at the wavelength of 800 nm for
endoscope imaging purpose. We propose the design of LPG for in-fibre dispersion
compensation at 800nm for two-photon endoscope imaging purpose.
2.7.3 Long Period fibre grating design
Resonance condition for fibre LPG is given by
,))()(( cladcoreres nn (2.32)
where ncore is the effective index of the fundamental core mode, nclad is the effective
index of the resonant cladding mode, Λ is the period of the core refractive index
modulation. The core and cladding effective indices can be expanded in Taylor
series about res. The impact of higher order derivatives in Taylor series for higher
order modes is less than 10% (error) to the value of core clad index difference at
the wavelength. For lower order modes (m<4) higher order values in the Taylor
expression contribute more errors and these values of dispersion should be taken
into account.The grating period that gives the coupling at the wavelength of p is
given by
Chapter 2
49
Λ= p / (neff-co –neff-cl ), (2.33)
where neff-co is determined from the dispersion relation to obtain the LP01 mode
normalized effective index b which is solved for the fibre from the expression
given by,
)(
)(
)1(
)1(1
0
1
0
1
bVK
bVKbV
bVJ
bVJbV
(2.34)
The effective indices of cladding modes can be determined by solving the
dispersion relation for three optical layers. Detailed dispersion relation derivation
for cladding modes is given in Appendix.
During LPGs fabrication, the refractive index difference achieved between UV
laser beams exposed portion and unexposed portion in the fibre core will determine
the grating characteristic like the coupling strength and the length of the grating.
The difference of effective index for the two co-propagating modes at chosen
resonant wavelength in the fibre will determine the period of the gratings required
for transfer of optical power between modes. Length of the grating is obtained from
the length required for the maximum power transfer condition. The amount of
anomalous dispersion achieved with this grating determines the distance between
the two LPG in series.
The length of gratings that gives the complete power transfer is
Lc= π/β , (2.35)
where the coupling coefficient ( ) between core mode LP01 and cladding modes
Chapter 2
50
with azimuthal value of 1 is given by [120] for step index fibres.
The imaging at the greater depth for available laser average power is limited by out
of focus fluorescence that is generated mainly near the surface of the sample [121-
124]. The ratio of focal (S) to out-of-focus fluorescence (B) is given by relation as
described in the manuscript of the fundamental imaging-depth limit in two-photon
microscopy is given by,
)/2exp()(2 2
2
s
s
lzznl
NA
D
S
(2.36)
where n is the tissue refractive index, ls is the scattering length.
For signal-to-noise ratio of unity (S/D= 1) at the excitation wavelength ‘ ’ of 1200
nm, tissue refractive index ‘n’ of 1.γγ and scattering length ‘ls’ of 390.65 m
(taking scattering coefficient of 2560 1m tissue at 1200 nm [125]) for NA of 0.5
and 0.21, imaging depths obtained are 1746.6 and 1290.2 respectively. These
values are two to three times greater than the 600 micron depth generally achieved
using two-photon microscopy at 800 nm wavelength.
This imaging depth can be improved further by simultaneous spatial focussing (by
focussing different wavelength component at same spot possible with aberration
free optics that uses two lenses with different materials) and temporal focussing of
pulse used as described in ref. [79]. Thus, if the dispersion is compensated then
using temporally focussed short pulses, penetration depth optimization is possible
in nonlinear microscopy [126-128].
Chapter 2
51
The STED system has been developed with common path for two beams from
different sources (excitation and depletion beam), which has made the STED
system as robust as previous; critical alignment of the two beams is not necessary.
The carefully designed phase plate (considering dispersion property of special glass
material for both beam at different wavelengths) produce required beam patterns
from the two beams which pass through it as described in ref. [129]. This setup was
able to observe dimension up to 60 nm.
Chapter 3
52
Chapter 3
Hollow-core photonics crystal fibre
for broadband nonlinear endoscopy
3.1 Introduction
A HC-PCF is unique for the use in nonlinear endoscopic imaging over solid-core
silica fibres because of its low chromatic dispersion, low nonlinearity, low loss and
a high damage threshold [130-132]. A HC-PCF is the most efficient fibre delivery
medium among fibre types, as the optical power in the HC-PCF propagates in the
air medium, which results in low transmission losses, low nonlinearity and low
Chapter 3
53
scattering [131,133-136]. In addition, this kind of fibre has no Fresnel loss in free
space coupling and gives a low background signal level [29,137]. The single mode
propagation at the 800 nm wavelength regime can be achieved in a HC-PCF
without sacrificing of the fibre core size. Further, a HC-PCF with zero group
velocity dispersion in the visible to near-infrared (NIR) wavelength range and
having large cladding numerical aperture (NA) with a large cladding diameter
allows us to have a high collection efficiency of the fluorescent signal in nonlinear
imaging.
In this Chapter, we demonstrate the feasibility of broadband excitation and
collection in a single fibre based nonlinear endomicorsocopy system using a piece
of hollow-core photonic crystal fibre (HC-PCF) integrated with a gradient index
(GRIN) lens.
3.2 Advantages of the HC-PCF in nonlinear endoscopy
Although the delivery of femtosecond pulses through a HC-PCF has been studied
for nonlinear excitation no endoscopy imaging has been achieved by collecting the
fluorescent signal through the cladding of the same piece of the fibre [102,138-
142]. Even for the investigation into a microscopy imaging system, the same piece
of the HC-PCF has not been used both for broadband nonlinear excitation and
Chapter 3
54
broadband fluorescent signal collection [143]. The collection of the fluorescent
signal in these systems is still conducted using a separate narrow band fibre other
than the HC-PCF [11,102 99, 138-142]. A single fibre endomicroscopy system
using a double-clad PCF (DC-PCF) has been developed both for excitation and
fluorescent signal collection but such a system requires a pre-chirp unit for
nonlinear excitation [16]. The ability to deliver the femtosecond light pulse in a
hollow-core over a broad range of wavelengths without the need for dispersion
compensation and the simultaneous collection of signal through the solid silica
cladding region gives the HC-PCF the advantage over other fibres used in single
fibre nonlinear endomicroscopy.
In this work, an endomicroscopy system formed by the integration of a single HC-
PCF and a GRIN lens is used. The system is tuneable over a broad wavelength
range of 750 – 850 nm without the need for the adjustment of the dispersion
compensation for each wavelength. The compactness of the system is improved by
avoiding the use of the pre-chirp units, which also reduces the optical power loss of
the system.
Chapter 3
55
3.3 Pulse width measurement through the HC-PCF fibre
Fig 3.1 Experimental setup for pulse width measurement with frequency resolved
optical gating (FROG); Laser: Ti: Sapphire laser (Spectra-Physics, Mai Tai HP,
~100 fs,80 MHz, 690-1040 nm), ND neutral density, OBJ 1 & OBJ2: 20x 0.25 NA;
XS1&XS2:3D fibre coupling stage; M1&M2: Ultrafast mirrors; FROG setup,
Swamp Optics, GRENOUILLE-008-50-USB).
Figure 3.1 shows the experimental setup for the pulse width measurement after the
propagation through 1.5 m HC-PCF. Figure 3.2 (a) shows the spectral intensity and
phase near zero wavelength of the fibre. The CCD image of the input pulse (Fig 3.2
(b)) within the frequency resolved optical gating (FROG) instrument and
Laser ND OBJ1
HC-PCF
M1 & M2
FROG
XS1
OBJ2
XS2
Chapter 3
56
algorithm retrieve pulse (Fig 3.2 (c)) give smooth the autocorrelation trace with the
FROG error of less than 1% as shown in Fig 3.2 (d).
Fig 3.2: (a) Spectral intensity and phase, (b) autocorrelation trace of (c) measured
and (d) retrieved pulse after the propagation through the 1.5 m length of HC-PCF.
(a) (b)
(c) (d)
Chapter 3
57
Fig. 3.3: Pule width measured at the output of the 1.5m HC-PCF for different
wavelengths at 100 mW
The output pulse width measured at the output end of the 1.5 m HC-PCF at the 100
mW input power by the FROG measurement is less than 100 fs for the wavelength
range from 750 nm to 850 nm as shown in Fig. 3.3. Therefore the compact
nonlinear endoscope as shown in Fig. 3.4(a) can be used for different wavelengths
without the requirement of the dispersion compensation. Thus, different bio-
markers can be efficiently used without the requirement for adjusting a pre-chirp
unit.
Chapter 3
58
3.4 Broadband excitation and collection in
nonlinear endomicroscopy
Fig. 3.4: (a) Schematic diagram of the experimental set-up for a broadband excitation and
collection system for single fibre nonlinear endomicroscopy. PMT: photo-multiplier tube.
(b) Enlarged part of the probe consisting of a HC-PCF and a GRIN lens. (c)-(e) Mode
profiles of the HC-PCF for different wavelengths. (f) Output power at different wavelengths
for the input power of 20 mW through a 1.5 m HC-PCF.
The schematic diagram of the experimental setup is shown in Fig. 3.4(a). It consists
of a Ti: Sapphire laser (Spectra-Physics, Mai Tai HP) which generates ultrafast
optical pulses with a pulse width of 100 fs, a repetition rate of 80 MHz and tuneable
Chapter 3
59
wavelengths around 690-1060 nm. Femtosecond pulses from the laser are coupled
through a 40x 0.85 NA objective to a HC-PCF (HC-800-01, NKT photonics, core
diameter 9.5 µm, core NA 0.2, cladding diameter 130 µm) with a zero dispersion
wavelength at 810 nm. For nonlinear imaging, a GRIN lens of numerical aperture 0.8
(GRINTECH, GT-MO-080-0415-810) is used in front of the output end of the fibre
for focussing the laser beam to the samples on a scanning stage and simultaneously
collecting the fluorescent signal back to fibre tip (Fig. 3.4(b)). The core and photonic
crystal regions do not support the visible signal as it is out of the band gap of the
fibre. Thus this part of the signal leaks out into the cladding region and propagates
with total internal reflection at the air and solid silica interface of the solid silica
cladding region after the outer acrylate coating of the fibre is removed. The
fluorescent signal is finally focused to a photo-multiplier tube (PMT). Band pass
filters (Schott - BG18 / Semrock - FF01-647/51) are used before the PMT to filter the
fluorescence signal from the reflected near-infrared light.
Figures 3.4(c)-(e) display the mode patterns in the HC-PCF at different wavelengths.
Since the wavelength 700 nm is out of the band gap of the photonic crystal in the HC-
PCF, the light could not be confined in the core, but distribute around the holey and
the cladding region of the fibre (Fig. 3.4(c)). On the other hand, the wavelength 800
nm is at the center of the photonic crystal band gap, the majority of light is confined
inside the core of the HC-PCF (Fig. 3.4(d)) and the remaining light leaks to the
cladding area. However, the wavelength 900 nm is close to the long wavelength edge
Chapter 3
60
of the photonic crystal band gap. Thus light confined in the core is decreased and
light leaked in the cladding increaseds (Fig. 3.4(e)) compared with the case in the
wavelength of 800 nm (Fig. 3.4(d)). Figure 3.4(f) shows the fibre output power
experimentally measured over the wavelength range for the given input power of 20
mW. Femtosecond pulses can be effectively delivered over a wavelength range of
750 - 850 nm through the core of the fibre. The transmission rates of the laser in the
fibre at the wavelength range is greater than 40% with highest 60% around
wavelength of 800 nm.
Fig. 3.5: Log-Log plot of the two-photon-excited fluorescence intensity (If) versus
the excitation laser power (Ip ) of fluorescent beads for wavelengths of 760 nm, 810
nm and 850 nm.
Dry fluorescent microspheres of diameters 1 μm and 2 μm (Fluoresbrite® Yellow
Green Microspheres, Polysciences Inc.) and Rhodamine B (Sigma Aldrich) were used
as test samples. The dependence of fluorescence intensity detected by the PMT (Fig.
(a)
Chapter 3
61
3.4(a)) on the excitation laser power at the samples for different wavelengths through
the HC-PCF fibre was measured. A plot of the fluorescence intensity against the
incident power on a logarithm scale for the central wavelengths of 760 nm, 810 nm
and 850 nm fits a straight line with a slope close to 2 (Fig. 3.5), indicating the
fluorescence signal detected by the PMT is due to the two-photon excitation for the
whole wavelength range.
Fig. 3.6: Log-Log plot of the two-photon-excited fluorescence intensity versus the
excitation laser power for different lengths of the fibre.
Figure 3.6 shows the plot of the two-photon-excited fluorescence intensity versus the
excitation laser power using dry fluorescent microspheres of diameter 1 μm as a
sample. For this experiment, the two-photon-excited fluorescence intensity for the
fibre with different lengths was measured by keeping the input coupling conditions
same. The threshold power where the fibre nonlinearity starts to play a role is
Chapter 3
62
different for different lengths of the fibre, as shown in Fig. 3.6. The threshold power
for fibre length of 200 cm, 140 cm and 80 cm are 75 mW, 92 mW, and 125 mW
respectively. These threshold power levels are higher than the corresponding power
levels that reported using other fibres [8,24] which is beneficial to the high efficiency
nonlinear endoscopy.
Fig. 3.7: (a)-(f) Two-photon fluorescence images of 1 μm fluorescent beads (scale bar: 5 µm) ; (g)
Lateral resolution (full width at half maximum) of 1 μm fluorescent beads for the excitation laser
power of 4.5 mW at the samples. (h)-(i) Two-photon fluorescence images of 2 μm fluorescent
beads and the Rhodamine B dye (scale bar: 10 µm) at 800 nm (h) with emission filter Semrock –
FF01-647/51 and ( i) Schott – BG18. (j) Log-Log plot of two-photon-excited fluorescence
intensity of Rhodamine B dye versus excitation laser power at 800 nm.
Chapter 3
63
3.5 Experimental results and discussion
Figures 3.7(a)-(f) show the two-photon-excited fluorescence images of the samples
containing the 1 µm diameter dry fluorescent microspheres over the wavelength
range, while Figs.3.7(h)-(i) show the two-photon-excited fluorescence images at 800
nm of the samples formed by mixing the 2 µm diameter fluorescent beads and
Rhodamine B fluorescent dye solution in water and dried on the glass slide. Emission
filters FF01-647/51, Schott BG18 are used to block the fluorescence from beads in
fig. 3.7(h) and from dye in fig. 3.7(i) respectively for imaging only the other.
Fig. 3.7(g) shows the lateral resolution (full-width at half maximum) for differrent
wavelengths. The slight degrading in resolution at short wavelengths is due to the
weak defocusing effect of the GRIN lens for the wavelength range. Fig. (j) Shows
the Log-Log plot of two-photon-excited fluorescence intensity of Rhodamine B
dye versus excitation laser power at the wavelength of 800 nm. It can be clearly
seen that Figs. 3.7(a)-(f) confirms the simultaneous boradband excitation and
collection (figs. (h)-(i) ), in the system shown Fig. 3.7(a), while Figs. 3.7(h)-(j)
demonstrate the multi-fluorophore two-photon imaging ability of the system.
Chapter 3
64
3.6 Conclusion
To summarize, broadband excitation and collection for a fibre-optic nonlinear
endo-microscopy system has been realized by using a single piece of the HC-PCF
integrated with a GRIN lens to operate in a NIR wavelength range from 750- 850
nm. The dispersion compensation adjustment for different wavelengths is not
required for nonlinear excitation over the range and thus multiple fluorescent
markers with different excitation peaks can be used simultaneously. The broadband
two-photon fluorescent signal can be collected through the cladding of the same
piece of the HC-PCF. The optical power level for system operation is reduced by
eliminating the requirement for lossy pre-chirp units for the chromatic dispersion
compensation. The HC-PCF has a higher nonlinear power threshold than other
types of fibres previously used in fibre-optic nonlinear endoscopy. The new endo-
microscopy imaging system is easy to operate, versatile, compact, and possible to
use low cost femtosecond pulsed fibre lasers.
Chapter 4
65
Chapter 4
Propagation of doughnut beams
through the hollow-core photonics
crystal fibre
4.1 Introduction
Different techniques have been available since a decade to achieve resolution
beyond long held diffraction limit barrier. Extensive reviews about different super-
resolution techniques and their applications can be found in a huge number of
literatures recently [144-150]. These systems are either hard to align, use complex
image restoration algorithm, or need specific laser/dye. All of them are limited to
bench top applications. The biological imaging community would tremendously
Chapter 4
66
benefit if these high resolution technologies could be implemented for routine in-
vivo imaging studies. With the use of optical fibre in biological imaging devices,
many bench top systems have been converted and are being used for clinical
diagnostic imaging and surgery; however the resolution of these devices are still
diffraction limited [3-5, 8-10,12,18-21,24-25,28, 99]. In resolution enhancement by
the spatial confinement of the excitation beam, for a scheme like in stimulated
emission depletion (STED) microscopy, phase plates are used on beam paths of
excitation and quenching laser beam to produce a doughnut shaped intensity profile
that overlaps with the central excitation spot. Resolution that could be achieved is
related to the dye property like the saturation intensity (Isat) and it’s rate(kqn)) and
the intensity of the quenching beam that could be achieved with a tight focus of a
high numerical aperture (NA) objective is given by the relation [144-150].
(4.1)
where λex is the excitation laser wavelength, n sinθ is the NA of objective used for
focussing two beams, the Isat value depends on the property of particular dye used
and Iqn is the intensity of the quenching laser beam achieved at the focus spot.
Previously fibre has been used in super-resolution imaging setup but its use is
limited mainly as the dispersion compensation element. Particularly the double-
clad fibre (DCF) structure is unique for use in nonlinear endoscopy as its different
layers can be used to excite dyes and collect the fluorescence signal simultaneously
,
sin2sat
qn
ex
I
In
res
Chapter 4
67
through the single piece of fibre [10, 16-18]. Recently, fibre-optical based super-
resolution imaging is demonstrated by using an azimuthal polarization state beam
instead of the use of the conventional doughnut beam for quenching the excitation
to achieved resolution beyond the diffraction limit in a fibre based endoscopic
imaging device [151]. With the first time demonstration of the concept, the system
can be further improved by using zero dispersive hollow-core photonics crystal
fibre (HC-PCF) and a single infra-red wavelength of for both excitation and
depletion. A single wavelength STED imaging system has also been successfully
demonstrated for bio-imaging at a wavelength of 770nm by using ATTO647N dye
[87-89, 152-154] in a bench-top system. Use of a single wavelength both for
excitation and quenching requires the implementation of pulse synchronization in
the time domain but simplifies the spatial overlap issues of two beams deep inside
the tissue owing the same optical response from components and tissues.
In this chapter we experimentally characterize the propagation of doughnut beams
through the core of a HC-PCF which is crucial for implementing compact single
wavelength single optical fibre based super-resolution imaging device for thick
biological samples. We use low NA coupling objectives to couple the light to HC-
PCF. The beam propagation condition for different input beam polarization states
and analysis after the fibre coupling objective are carried out.
Chapter 4
68
4.2 Experimental setup
Fig. 4.1: Schematic diagram of the experimental set-up for the
characterization of doughnut beams through a hollow-core photonics
crystal fibre. ND: neutral density, L1&L2: lens, HWP: half wave plate,
VPP: vortex phase plate, QWP: quarter wave plate, VA: variable
aperture, BM: beam manipulation, HC-PCF: zero dispersion hollow-
core photonic crystal around the wavelength of ~807 nm, CCD: charge
couple device, A: analyser
The schematic diagram of the experimental setup is shown in Fig. 4.1. It consists of
a Ti: Sapphire laser (Spectra-Physics, Mai Tai HP) which generates ultrafast
optical pulses with a pulse width of 100 fs, a repetition rate of 80 MHz and
tuneable wavelengths around 690-1060 nm. Femtosecond pulses from the laser are
coupled through a 10x 0.4 NA objective to a HC-PCF (HC-800-02, NKT
photonics, core diameter 7.5 µm, core NA 0.2, cladding diameter 130 µm) with a
Chapter 4
69
zero dispersion wavelength at 807 nm. Different input beam conditions are
achieved with different combinations of wave plates, a polarization state converter,
the vortex phase plates and a variable aperture in front of the fibre coupling
objective. The output of the fibre is imaged to the charged couple device (CCD) by
using the 40 x 0.65NA objectives. An analyser is inserted between the fibre output
and the CCD camera.
4.3 Experimental results
The HC-PCF can provide advantages over the fibre for implementing superresoltuion
imaging scheme. The input beam condition is not maintained at the output of the fibre
due to bifringence in the case of solid silica fibres, which is the major hurdle for
implementing STED scheme as circularly polarized beam superimposed with a phase
vortex are used to achieve a doughnut beam profile. The propagation medium at the
centre of the HC-PCF fibre is air and the propagation mechanism is fundamentally
different. HC-PCF can be designed to operate over a broadband at the designed centre
wavelength, which can provide an extra flexibility in choosing dyes arround 800 nm
for implementing STED.
The beam propagation condition through the HC-PCF for different input beam
polarization states with or without the superposition of the vortex phase and analysis
of the beam after the fibre coupling objective are given in the following sections.
Chapter 4
70
4.3.1 Linear polarization states without/with vortex phase
Fig. 4.2 (a) Intensity ratio of dip and peak of the fibre mode for
different NA. (b)-(f). Mode profiles for the linear input polarization
state and at different angles of the analyser with respect to the vertical
axis. Scale bar 5 µm.
0
0.25
0.5
0.75
1
0.1 0.2 0.3 0.4 0.5
Numerical Aperture (NA)
Imin
/Im
ax
(a)
(b) (c) (d) (e) (f)
Chapter 4
71
Figure 4.2 (a) shows the intensity ratio of dip to peak for linear polarization states.
Output beam profile from the fibre is imaged to CCD which changes from a
fundamental mode to a doughnut-like intensity profile with intensity deep in the
centre as NA of the input beam to the fibre input is increased. The ratio of
minimum intensity to maximum intensity at the centre achieved is around 0.5 for
this case. The ring shaped intensity profile breaks for different analyser angle with
respect to the vertical axis positioned between CCD and the fibre output as shown
in figure 4.2 (b)-(f).
Figure 4.3 (a) shows the intensity ratio of dip to peak for the case of linear
polarization states overlapped with a linear phase ramp obtained from the vortex
phase plate before coupling to the fibre input. Output beam profile from the fibre is
imaged to CCD which changes from a fundamental mode to doughnut-like
intensity profile with intensity deep in the centre as NA of the input beam to the
fibre input is increased. The intensity dip achieved is fast with change in NA for
this input polarization states. The ring shaped intensity profile breaks for different
analyser orientation to the vertical axis which is kept in front of CCD after the fibre
output as shown in figure 4.3 (b)-(f).
Chapter 4
72
Fig. 4.3 (a) Intensity ratio of dip and peak of the fibre mode for
different NA. (b)-(f). Mode profiles for the linear input polarization
states superimposed with the vortex phase and at different angles of
the analyser with respect to the vertical axis. Scale bar 5 µm.
0
0.25
0.5
0.75
1
0.1 0.2 0.3 0.4 0.5Numerical Aperture (NA)
Imin
/Im
ax
(b) (c) (d) (e) (f)
(a)
Chapter 4
73
4.3.2 Radial polarization states without/with vortex phase
Fig. 4.4 (a) Intensity ratio of dip and peak of the fibre mode for
different NA. (b)-(f). Mode profiles for the input radial polarization
states and at different angles of the analyser with respect to the vertical
axis . Scale bar 5 µm.
0
0.25
0.5
0.75
1
0.1 0.2 0.3 0.4 0.5
Numerical Aperture (NA)
Imin
/Im
ax
Numerical Aperture (NA)
Imin
/Im
ax
(b)
(a)
(c) (d) (e) (f)
Chapter 4
74
Figure 4.4(a) shows the intensity ratio of dip to peak for the case of radial input
polarization states. Output beam profile from the fibre is imaged to the CCD which
changes from a fundamental mode to doughnut-like intensity profile with intensity
dip in the centre as NA of the input beam to the fibre input is increased. In this
case, the fundamental mode intensity profile changes to the ring intensity profile
after redistribution of intensity for increase in NA. The doughnut-like intensity
profile after the analyser is achieved which breaks only for 45 º angle of analyser
to the vertical axis as shown in figure 4.4 (b)-(f).
Figure 4.5 (a) shows the intensity ratio of dip to peak for the case of radial
polarization states overlapped with the linear phase obtained from the phase plate
coupled to the fibre. Output beam profile imaged to the CCD camera change from a
fundamental mode to the doughnut-like intensity profile with intensity deep in the
centre as NA of the input beam to the fibre input is increased by opening the
variable aperture in front of the coupling objective. The intensity dip develops
slowly with change in NA for this polarization state. Orientation of side lobes are
changed for different analyser angle to the vertical axis between CCD and fibre
output suggesting radial polarization state at the output is change by the fibre as
shown in figure 4.5 (b)-(f).
Chapter 4
75
Fig. 4.5 (a) Intensity ratio of dip and peak of the fibre mode for
different NA. (b)-(f). Mode profiles for the input radial polarization
states superimposed with the vortex phase and at different angles of
the analyser with respect to the vertical axis . Scale bar 5 µm.
0
0.25
0.5
0.75
1
0.1 0.2 0.3 0.4 0.5
Numerical Aperture (NA)
Imin
/Im
ax
(b) (c) (d) (e) (f)
(a)
Chapter 4
76
4.3.3 Azimuthal Polarization states without/with vortex phase
Fig. 4.6 (a) Intensity ratio of dip and peak of the fibre mode for
different numerical aperture. (b)-(f). Mode profiles for the input
azimuthal polarization states and at different angles of the analyser with
respect to the vertical axis. Scale bar 5 µm.
0
0.25
0.5
0.75
1
0.1 0.2 0.3 0.4 0.5
Numerical Aperture (NA)
Imin
/Im
ax
(a)
(b) (c) (d) (e) (f)
Chapter 4
77
Figure 4.6 (a) shows the intensity ratio of dip to peak for the case of an azimuthal
input polarization state. Output beam profile from the fibre output imaged to the
CCD camera changes from a fundamental mode to a doughnut-like intensity profile
with intensity deep in the centre as NA of the input beam to the fibre input is
increased by opening the variable aperture in front of the coupling objective. The
fundamental mode intensity profile breaks into two side lobes for NA of 0.4. The
doughnut-like intensity profile is obtained after the analyser angle of 135 º to the
vertical axis on the CCD camera as shown in figure 4.6 (b)-(f).
Figure 4.7 (a) shows the numerical aperture verses the normalized intensity ( Imin /
Imax) for the case of an azimuthal polarization state overlapped with the vortex
phase obtained from the phase plate at the fibre input. Output beam profile from the
fibre imaged to the CCD change from a fundamental mode to azimuthal
polarisation states as NA is increased. Orientation of side lobe is changed for
different analyser angles to the vertical axis suggesting polarization states change
as propagated through the fibre as shown in 4.7 (b)-(f).
Chapter 4
78
Fig. 4.7 (a) numerical aperture verses the normalized intensity ( Imin /
Imax) for an azimuthal polarization state overlapped with the vortex
phase. (b)-(f). Mode profiles for the input azimuthal polarization state
superimposed with the vortex phase and at different orientation of the
analyser with respect to the vertical axis. Scale bar 5 µm.
0
0.25
0.5
0.75
1
0.1 0.2 0.3 0.4 0.5
Numerical Aperture (NA)
Imin
/Im
ax
(b) (c) (d) (e) (f)
(a)
Chapter 4
79
4.4.4 Circular Polarization states without/with vortex phase
Fig. 4.8 (a) Intensity ratio of dip and peak of the fibre mode for
different numerical aperture. (b)-(f). Mode profiles for the input
circular polarization states and at different angles of the analyser with
respect to the vertical axis . Scale bar 5 µm.
0
0.25
0.5
0.75
1
0.1 0.2 0.3 0.4 0.5
Numerical Aperture (NA)
Imin
/Im
ax
(b) (c) (d) (e) (f)
(a)
Chapter 4
80
Figures 4.8- 4.9 show the beam propagation condition through HC-PCF for
different input polarization states without/with the overlapping of linear phase
ramp obtained from the vortex phase plate. Figure 4.8 (a) shows the intensity ratio
of dip to peak for circular polarization states. Output beam profile from the fibre is
imaged to the CCD camera which changes from a fundamental mode to doughnut-
like intensity profile with intensity deep in the centre as NA of the input beam to
the fibre input is increased by opening the variable aperture in front of the coupling
objective. The ratio of minimum intensity to maximum intensity at the centre is
achieved is around 0.5 in this case. The ring shaped intensity profile is maintained
for different analyser orientation to the vertical axis kept in front of CCD as shown
in figure 4.8 (b)-(f).
Figure 4.9(a) shows the fundamental and the doughnut mode at the core for different
effective coupling NA achieved by varying the beam aperture in front of the coupling
objective for the case of the circular polarization states superimposed with the vortex
phase achieved by the phase plate. The mode profile changes from the fundamental
mode to a doughnut mode with a centre intensity null as we open the variable
aperture allowing the higher order modes to couple into the core of the fibre. This
gives us the way to couple the differrent beam profile required for the STED to
couple to the core of the fibre using single objective by varrying the beam diameter of
the two beam path. Figures 4.9(b)-(f) shows the beam shape for the different analyser
angles to the vertical axis kept in front of the CCD. The doughnut beam with the
Chapter 4
81
central null intensity is maintained for any analyser orientation at the fibre output for
the input condition of circular polarization states with a superimposed vortex phase.
Figure 4.9 (g-h) displays the doughnut-like beam mode patterns and their cross-
section at the output face of the HC-PCF at 807 nm wavelength.
To implement a single wavelength optical fibre based STED system both the
fundamental excitation beam and the doughnut beam need to be propagate at the core
of the fibre to ensure spatial overlap of two beams at the sample plane as the
coupling of the doughnut beam coupled to the solid silica cladding could not maintain
its phase and polarization states in the silica medium.The spatial overlap of the two
beams at the sample plane at the output of the fibre is also not ensured if two beams
propagate through the physically displaced and different NA region of the fibre.
During experiments the different input beam conditions are achieved with different
combination of wave plates, polarization states converter, vortex phase plates and
variable aperture in front of the fibre coupling objective used for coupling the beam
to the core of HC-PCF fibre. Though the HC-PCF fibre are endlessly single mode for
and above the designed wavelength we use only short length of 1.5 m of the fibre to
achieve higher order propagation for different coupling conditions achivied by the
varriable aperture in front of the coupling objective.
Chapter 4
82
Fig. 4.9 (a) Intensity ratio of dip and peak of the fibre mode for
different numerical aperture. (b)-(f). Mode profiles for the input
circular polarization states superimposed with the vortex phase and at
different angles of the analyser with respect to the vertical axis . (g)
Doughnut mode profile at the fibre output, (h) cross section intensity
profile. Scale bar 5 µm.
(a)
(b) (c) (d) (e) (f)
(g)
(h)
Chapter 4
83
For other cylindrical polarization states, the linear polarization states, azimuthal
polarization states with and without the superimposed vortex phase, the doughnut
beam shape is not maintained for different orientation of the analyser suggest that for
these input condition mode profile is not pure doughnut but combination of higher
order modes as shown in figures 4.2-4.7.
Comparing figures from 4.2 to 4.9, minimum intensity dip in the centre of
doughnut-like intensity profile achieved after propagation through the fibre is
deeper in case of addition of phase ramp using the vortex phase plate than without
for the circular polarization states. The doughnut intensity profile is maintained for
different orientations of analyser angles with respect to the axis for the both case.
Chapter 4
84
4.4 Conclusion
We characterized the beam propagation for different input polarization conditions (
linear, cylindrical and circular with/ without a superimposed vortex phase ) through
the core of the fibre. Our experimental results suggest that the doughnut mode with
centre intensity null can be propagated through the core of the HC-PCF for circular
polarization states superimposed with the vortex phase. Doughnut mode and
fundamental mode coupling can be achieved simultaneously to the core of the fibre
by varying the beam width through the low NA single coupling objective, which is
the crucial step towards implementing the single wavelength fibre-optic super-
resolution imaging endoscope.
Chapter 5
85
Chapter 5
Design of zero dispersive double-clad
fibre for two color light
5.1 Introduction
A single mode fibre (SMF) was first tried in nonlinear endoscopy for the delivery
of ultra-short pulses to specimens and the collection of nonlinear signal [24].
However, an SMF has limitation in fluorescence signal collection due to its small
core diameter. A double-clad photonic crystal fibre (DCPCF) was sought to
increase the collection of nonlinear signal due to its large inner cladding size [9, 24,
99]. A DCPCF has been reported to have a signal collection efficiency improved
Chapter 5
86
by two orders of magnitude compared to an SMF [9]. Despite the improvement in
signal collection, a DCPCF is mechanically less stable due to its air hole based
structure. System robustness is one of the critical requirements for nonlinear
endoscopy systems where the fibre is mechanically scanned and bent frequently
during imaging. To overcome those drawbacks, a double-clad solid silica fibre was
used in nonlinear endoscopy [10,25]. It has been reported that a double-clad fibre
(DCF) is not only mechanically robust, but also has high nonlinear signal collection
efficiency due to its larger inner cladding diameter [12, 16-18,155]. In addition, a
DCF has higher light confinement than a DCPCF. As a fibre coupler is integrated
in a nonlinear endoscopy system, it can further improve the system compactness
and the signal collection efficiency [23]. The signal collection efficiency of a DCF
coupler is one order magnitude higher than that of a DCPCF coupler.
In spite of these advances in nonlinear endoscopy, fibre based nonlinear endoscopy
systems face a common problem. When femtosecond excitation pulses propagate
through a fibre, they suffer chromatic dispersion from the fibre which broadens the
pulses and reduces the nonlinear signal excitation efficiency. Femtosecond pulses
with a central wavelength of 800 nm are widely used in nonlinear imaging because
the peak of two-photon excitation wavelengths of fluorescein and acriflavine
solutions is near 800 nm. They are the only fluorophores that has been approved for
use in human. All solid silica fibres have a high normal chromatic dispersion at a
wavelength of 800 nm. Currently, nonlinear endoscopy uses grating pairs or prisms
for dispersion compensation of optical pulses with a central wavelength of 800 nm
Chapter 5
87
[3-17, 24-25, 99,156-158], but these units are bulky and are not stable in alignment.
In addition, dispersion compensation using grating pairs or prisms requires a laser
beam to have multiple reflections or transmissions through the units, which induce
high loss to the laser beam power [18, 25].
A higher-order-mode (HOM) fibre has been reported for chromatic dispersion
compensation of optical pulses within a fibre [159-164]. However the HOM fibre
has a single core/cladding structure and only uses its core for transmitting optical
pulses with a single numerical aperture (NA) determined by the core/cladding
index difference at a given wavelength [165]. The core of the HOM fibre supports
multiple modes and the HOM fibre needs to splice two single mode fibres at each
end of the HOM fibre to balance the group delay dispersion, which induces
additional loss. Therefore, it is limited in many applications including nonlinear
endoscopy which requires the operation of two wavelengths of light with different
NA.
In this chapter, we explain our a new DCF design with two values of the NA for
the efficient transmission of two colors of light, near-infrared optical pulses as well
as visible continuous wave (CW) light. The DCF with two inbuilt long period
gratings (LPGs) enables the transmission of near-infrared optical pulses with zero
net chromatic dispersion in its core as well as the high efficiency collection of
visible light coupled by an objective lens. This DCF is promising to be used in a
compact nonlinear fibre-optic imaging system.
Chapter 5
88
5.2 Zero chromatic dispersion
A solid silica DCF is mechanically robust. However, a fundamental mode at a
wavelength of 800 nm is far away from the zero dispersion wavelength (1270 nm)
of a silica fibre and it has high group delay dispersion of ~ -110 ps/nm·km [166].
As femtosecond pulses with a central wavelength of 800 nm in a fundamental
mode transmit through the solid core of a DCF, they suffer from large chromatic
dispersion. Here we shape the refractive index profile of a DCF and place a pair of
inbuilt LPGs inside the fibre core as shown in Fig. 5.1(a) to realize transmit optical
pulses with no chromatic dispersion. The DCF consists of a core, an inner cladding
and an outer cladding section and the refractive index profile of each section is
displayed in Fig. 5.1(b). The coating of the DCF is high refractive index acrylate
coating. Alternatively, a DCF can also be realized by replacing the outer cladding
section with low refractive index coating. Here our design uses the former
approach which is same as the structure of most commercial DCFs.
Chapter 5
89
Fig. 5.1: (a) Schematic structure of a DCF intergrated with a pair of gratings for
achieving zero net chromatic dispersion. (b) Refractive index profile of the DCF.
Chapter 5
90
A pair of long period gratings are used in the conversion from a fundamental mode
to a higher order mode LP02 (Fig. 5.1(a)). The net zero chromatic dispersion to
optical pulses is realized by arranging the dispersion parameters of fundamental
mode pulses and higher order mode pulses into an opposite sign. However a
commercially available DCF has normal dispersion for both a fundamental mode
and a higher order mode. We shaped the refractive index profile of a DCF (Fig.
5.1(b)) to let the higher order mode have anomalous waveguide dispersion
exceeding the material dispersion value, giving a net anomalous dispersion value.
Fig.5.2: Mode evolution for mode LP01 and mode LP02 at wavelengths 725 nm
(blue), 775 nm (green) and 825 nm (red) for the w-DCF. The refractive index
profile is shown in background (black).
Chapter 5
91
Figure 5.2, shows the simulated radial intensity distribution of the LP01 and LP02
modes at different wavelength components of 725 nm, 775 nm and 825 nm using
simulation software Optifibre (Optiwave). As the wavelength component
increases, the fraction of power of the LP02 mode in higher-index regions increases
with the longer wavelength as the mode confines towards the core-center. For a
given wavelength light in higher-index regions travels slower than that in lower-
index regions. The blue components in the LP02 mode travel faster than the red
components, and the LP02 mode has anomalous waveguide dispersion in the fibre.
The core diameter (d1), the thickness of the index dip region outside the core (d2)
and the thickness of the inner cladding region (d3) as shown in Fig. 5.1 affect the
waveguide dispersion of the LP02 mode in the DCF. Figure 5.3 reveals the
simulated waveguide dispersion Dw of the DCF affected by d1, d2, and d3. The
wavelength of the peak waveguide dispersion increases with the increase of d1, d2
and d3.
Chapter 5
92
Fig. 5.3: Waveguide dispersion Dw of a higher order mode in the DCF (a) for
different core diameters d1, (b) for different thicknesses of the index dip region d2,
and (c) for different cladding thicknesses d3.
Chapter 5
93
The core diameter d1, the thickness of the index dip region d2 and the thickness of
the inner cladding region d3 are chosen as 3.κ m, 2.6 m and 14.2 m,
respectively, to obtain the anomalous dispersion peak value at the wavelength of
800nm for the designed core NA. By achieving a high waveguide dispersion value
for an LP02 mode in the fibre, the positive value for the total dispersion at the
designed wavelength of 800 nm can be obtained.
Figure 5.4: (a) Dispersion values for the higher order mode. (b)
Dispersion values for the fundamental mode.
Chapter 5
94
After considering the low index values (that can be achieved through adding
fluorine in silica during fabrication process) in the dip region, we were able to
achieve anomalous waveguide dispersion of +180 ps/nm·km for the wavelength of
800 nm. Figure 5.3 shows the dispersion value for LP02 and LP01.Taking away the
material dispersion -110 ps/nm·km, the total dispersion value for the LP02 mode in
the fibre is ~ +86 ps/nm·km. The waveguide dispersion for the fundamental (LP01)
mode is -197 ps/km·nm, making total dispersion value -324 ps/km·nm. By keeping
the ratio of (L1+L3): L2=1:3.77, the DCF can realize total dispersion of zero to
optical pulses at the central wavelength of 800 nm (Figure 5.5). The dispersion of
mode LP02 of our designed fibre at the wavelength of 800 nm is slightly smaller
than that of a HOM fibre at 770 nm with D=+112.7 ps/(nm·km) [159]. Therefore
our designed fibre requires a longer potion of fibre propagating in mode LP02,
which would lead to the benefit of reducing the fibre nonlinearity since the
nonlinearity of mode LP02 is lower than that of the fundamental mode.
Fig. 5.5: Total dispersion of the fibre under the condition (L1+L3) :L2 = 1:3.77.
Chapter 5
95
In this case, the third order dispersions (TOD) of the fundamental mode and the
LP02 mode are 4.6×10-5
ps3/m and -5.4×10
-4 ps
3/m, respectively. For comparison,
the third order dispersions of a HOM fibre is -0.0002229ps3/m [159] which is
slightly lower than the TOD of the LP02 mode in the DCF. However the third order
dispersion of mode LP01 has an opposite sign to that of mode LP02, thus it can
partially compensate for the third order dispersion of mode LP02 in the DCF. The
net accumulated TODs achieved with a grating pulse compressor or a prism pulse
compressor is ~ +1.1×10-4
ps3/m and ~ -1.3×10
-4 ps
3/m, respectively. The TOD of
the DCF is slightly higher than that from a pair of gratings and a pair of prisms.
However, dispersion compensation using gratings or prisms is bulky and induces
high loss to optical pulses.
The mode effective area of the DCF is found as Aeff=180 µm2, which is 12 folds
larger than that of a HOM fibre [159]. Therefore the DCF induces less nonlinearity
and is more tolerant to the high peak power of optical pulses than HOM fibres. For
optical pulses with a pulse width of 100 fs and a peak power of 1 kW, the
dispersion lengths (Ld) of mode LP01 and mode LP02 are 9 and 34 cm while the
nonlinear length (Lnl) of mode LP01 and mode LP02 are 27 and 232 cm, respectively.
The pulse propagation in the core of the DCF is basically linear and the impact of
nonlinear effects can be negligible.
Chapter 5
96
5.3 Mode conversions
LPGs can induce a single periodic perturbation to a mode in a fibre and perform a
mode conversion from one mode to another mode [167]. Ideally, with an
appropriate set of gratings, any number of modes with arbitrary amplitudes and
phases can be converted into any other modes with arbitrary phase and amplitude
obeying the energy conservation law [167-174]. In an optical fibre, a LPG gives
the efficient mode conversion between two forward propagating modes by
matching the grating period (Λ) to the effective index (neff) difference between two
co-propagating modes at a specific resonant wavelength λres. The period of the LPG
is given by relation [167],
0102 LPLP
res
nn
, (5.1)
where 01LPn and 02LPn are the effective index of the two coupled modes. Here, the
conversion between fundamental and higher order modes (LP01 and LP02) is
performed by a pair of inbuilt LPGs (Fig. 1(a)). For the DCF shown in Fig.1, 01LPn =
1.45526, and 02LPn =1.45515; therefore the period of the LPG Λ is 6.7 mm for the
specific resonance wavelength res=800 nm. The grating refractive index
modulation is chosen to be 0.0005, the maximum grating refractive index
modulation because higher refractive index modulation can induce higher
perturbation to a mode in a fibre and be more effective in mode conversion. The
Chapter 5
97
current LPG fabrication method using a UV laser can have index modulation of
0.0005 for the grating in a silica fibre [175-178]. Using the coupled-mode theory,
the transmission function of the LPG at any wavelength is given by [179],
,
)]1([1
])]1([1[sin
1),(2
22
0
res
resgL
T (5.2)
where T is the transmission function of a LPG filter, Lg is the grating length, is
the coupling constant for the grating,
,),,(4
10201
2
0 dxdyEEzyxnk LPLP (5.3)
),,,(),,(),,( 2
0
22zyxnzyxnzyxn (5.4)
where ),,(2zyxn is the periodic refractive index perturbation of the grating,
),,(2
0 zyxn is the index profile of the waveguide, ),,(2zyxn is the grating index
profile, is the frequency of electric field oscillation, 0 is the permittivity in free
space, 01LPE is the mode field of LP01, and
01LPE the mode field of LP02.
The broad mode conversion bandwidth is achieved through the optimization of the
grating length and the grating period [180-182]. The grating period is calculated by
Equation (5.1) and is determined by the refractive index profile of the DCF.
However the grating period can still be detuned slightly and also keep mode LP02 at
anomalous dispersion.
Chapter 5
98
Fig. 5.6: Mode conversion of the first (a) and the second (b) long
period grating are the intensity of input in mode LP01, input in mode LP02, output in
mode LP01 and output in mode LP02.
Figure 5.6 reveals the transmission spectrum of the two modes by the two gratings
where a refractive index modulation is 0.0005. The mode conversions of the first
and second LPGs are slightly different due to different lengths of the used LPGs.
The length of the first LPG is 16.42 mm while the length of the second LPG is
32.75 mm. Variation of the lengths of two LPGs is due to the optimization of
different transmission peaks of two modes for converting LP01 to LP02 and vice
outoutinin HFHF IIII ,,,
Chapter 5
99
versa. All optical intensity at a wavelength of 800 nm is converted from a
fundamental mode in the core to a higher order mode in the inner cladding by the
first grating and recovered back to a fundamental mode by the second grating as
shown in Fig. 5.6. The two gratings can convert more than 99% of light from a
fundamental mode into a higher order (LP02) mode and back to a fundamental
mode over a bandwidth of 23 nm at the central wavelength of 800 nm. The
bandwidth of 100 fs optical pulses is 15 nm. Therefore the LPGs can sufficiently
convert optical pulses with a pulse width over 100 fs between the fundamental
mode and the LP02 mode.
For LPG fabrication, among different fabrication methods, realization of LPG in
the core of the hydrogen loaded optical fibre by exposing to KrF laser (248nm), is
established method currently in use [93,30,116-119,179-190]. The 2-3% Hydrogen
loading for few hours will increase defects at Ge sites in Germanium doped fibre
core before UV exposure. After laser writing the fibre is annealed to stabilize the
achieved refractive index change by annealing the fibre. This process extract the
hydrogen from the fibre, further stopping hydrogen to increase the defects at Ge
sites that would have raised the average index of fibre resulting the shift in the
resonance peak to longer wavelength. Annealing is generally carried out at 150
degree centigrade for 10 hours. The long period of the gratings is robust to the
bending of the fibre in practical use. The change of the refractive index of grating
period is normally less than 0.1% due to bending, which can be ignored. In
addition, there is a slight change in the resonance peak of the mode conversion for
Chapter 5
100
1-2 % change in length of the grating period due to stretching or bending of fibre,
which can also be ignored.
5.4 High efficient operation of two color light
In many applications including nonlinear endoscopy, an optical fibre is required to
deliver ultra-short optical pulses with a central wavelength of 800 nm as well as to
collect the continuous wave (CW) light with high collection efficiency at other
wavelengths. The design of the DCF, discussed in Figs. 5.1-5.6, can meet those
demands. First, optical pulses are delivered through the core of the DCF. As been
demonstrated in Fig. 5.6, the loss of the DCF to optical pulses, caused by a pair of
inbuilt gratings, is less than 1%. Second, the DCF can also be used for the
backward collection of visible light with a high efficiency. The NA of the core and
the inner cladding of the DCF which are determined by the refractive index
contrast of the core and the trench, and the index contrast of the inner cladding and
the outer cladding, respectively, as shown in Fig. 5.1(b). The NA of the core and
the inner cladding of the DCF are ~ 0.13 and ~ 0.21, respectively. Here, the
wavelength used for calculating the NA of the inner cladding is 521 nm, which is
chosen as the emission wavelength of fluorescein. Figure 5.7(a) shows the
schematic diagram of the DCF for the collection of visible light and dependence of
collection efficiency on the NA of the visible light beam. The collection efficiency
is calculated by calculating the coupling efficiency of the DCF to visible light using
Chapter 5
101
ray optics. The rays with angless less than 120, the corresponding beam angles for
NA of 0.21, can propagate through the DCF through total reflection and the rays
with angless larger than 120 are leaked from the DCF during the propagation and
could not be successfully collected by the DCF. As shown in Fig. 5.7(b), the DCF
can collect 100% of the visible light if the beam NA is less than 0.21. If the optical
pulsed beam from the core of the DCF fills the full aperture of the objective lens,
the NA of the CW beam coupled back by the objective will be equal to the core NA
of the DCF. In this case, the DCF can fully collect all the visible light coupled
through the objective lens.
Fig.5.7 (a) Schematic diagram of the DCF for operating two color light. (b) The
fibre collection efficiency η of the visible light versus the NA of the beam coupled
by an objective.
Chapter 5
102
High NA and large size of the inner cladding can increase the collection of the
visible light. However the size of the inner cladding of the DCF also affects the
dispersion parameter of the LP02 mode (Fig. 5.3). We can maximize the NA and
size of the inner cladding of the DCF without sacrificing anomalous dispersion of
mode LP02.
5.5 Conclusion
In summary, a double-clad solid silica fibre integrated with inbuilt LPGs has been
designed based on the fabrication limitation of the manufacturing process. The
DCF demonstrates two values of the NA for near-infrared and visible beams.
Therefore, it can realize the chromatic dispersion compensation within the optical
fibre at a wavelength of 800 nm as well as provide the high efficiency collection of
visible light through its inner cladding. The loss of the DCF to near-infrared optical
pulses is less than 1%. The designed DCF, which can realize the dispersion
compensation and low loss to the excitation pulses, is important for nonlinear
endoscopy systems to become portable and be able to use a low power and cost
effective femtosecond fibre lasers.
Chapter 6
103
Chapter 6
Conclusion
6.1 Thesis conclusion
Fibre-optic nonlinear imaging is one of the best tools available presently to access
internal organs and tissue surfaces for real time diagnostic imaging and minimally
invasive surgical procedures. The miniaturization of the optical probe and making
whole system more versatile, easy to operate and robust portable has been the
biggest challenge of all times since its beginning. Resolution improvement of these
devices adds their importance in the long term study of the disease development
and the effectiveness of drugs at a sub-cellular level. The low optical power
Chapter 6
104
requirement with highly efficient utilization of excitation beams is important to
make these systems use low cost pulsed femtosecond laser sources. Our
experimental results show that the use of hollow-core photonic crystal fibre (HC-
PCF) in the endomicroscopy system provides options for the broadband excitation
and collection enabling the simultaneous use of multiple fluorescent markers with
different excitation peaks. There is no need of a separate fibre to collect back
fluorescent signal as the cladding of the same HC-PCF collects back fluorescent
signal in the system with the integrated HC-PCF and GRIN lens probe used. The
fibre has a higher nonlinear power threshold than other fibres and the optical power
level required to handle by the fibre is also reduced by the elimination of the
requirement of the use of pre-chirp unit for the chromatic dispersion compensation
that comprises mostly grating pairs which in some cases loses more than 80%
optical power and hinders the device tuneability for different wavelength operation.
Resolution improved fibre-optic nonlinear endoscopy could revolutionize the
clinical research. Super-resolution technology is highly desired features in fibre-
optic based systems in research communities. Towards that direction our study of
the propagation of doughnut beams, to implement STED like super-resolution
imaging features in fibre-optic nonlinear endomicroscopy is valuable. The use of a
single wavelength helps to optimise optics for excitation. Following the successful
demonstration of the two wavelength STED system using a vortex beam in the
fibre-optic endomicroscopy system and the single wavelength bench-top STED in
Chapter 6
105
microscope system we have characterized the propagation of the doughnut beam
for different input polarization states through the HC-PCF fibre at a single
wavelength. Our experimental study shows the way to achieve the doughnut beam
and the propagation by varying the coupling NA for radial and circular polarization
states.
Our new robust solid silica fibre design compensates for the group velocity
dispersion within its designed length by using the mode conversion technique for
femtosecond excitation pulses at the wavelength of 800 nm and also collects the
fluorescent signal through its cladding. The optical loss of less than 1% is achieved
in the mode conversion within the fibre, making it possible to use low power, cost
effective pulsed femtosecond fibre /semiconductor laser in fibre-optic nonlinear
imaging systems.
Chapter 6
106
6.2 Future work
Implementation of a micro-electro mechanical system (MEMS) based probe
scanning system to carry out broadband excitation and collection from biological
samples for multiple fluorescent markers simultaneously will be worth trying.
Further the characterization of newly designed fibre and its use in nonlinear
endoscopy imaging based super-resolution imaging is worth continuing study in
the future. Fibre-optic based single wavelength super-resolution imaging of
biological samples using various long emitting dyes will be the first study of its
kind towards the demonstration of a simple fibre-optic super-resolution imaging
system. Imaging of dyes and bio-samples for different input polarization states with
the HC-PCF fibre is important towards that goal. Detailed study of characteristic of
different dyes particularly long emitting dyes and auto fluorescence for the use in
single wavelength super-resolution imaging is necessary. Pulse characterization
and synchronization using pulse measurement techniques like second harmonic
generation is necessary to confirm the pulse overlap at the focus of the endoscope
probe example that uses a gradient index lens and the scanning mechanism like
using MEMS is necessary to realize miniaturize fibre-optic based super-resolution
endoscopy. Super-resolution imaging study of thick samples using single or
multiple wavelengths is necessary before we can use these techniques for long term
imaging of live suspects.
Bibliography
107
Bibliography
[1] Kobat, D., “Deep tissue multiphoton microscopy using longer wavelength
excitation”, Opt. Express. 17, 13354–13364, (2009)
[2] C.Wang, L.Oiao, Z. Mao, Y.Cheng, and Z.Xu, “Reduced deep-tissue image
degradation in three-dimensional multiphoton microscopy with concentric
two-color two-photon fluorescence excitation”, J. Opt. Soc. Am. B, 25,
976-982, (2008).
[3] D. Brid and M. Gu, “Two-photon fluorescence endoscopy with a micro-
optic scanning head”, Opt. Lett. 28, 1552-1554, (2003).
[4] M. T. Myaing, D.J MacDonald, and X.Li, “Fibre-optic scanning two-
photon fluorescence endoscope” , Opt. Lett. 31, 1076-1078, (2006).
[5] L. Fu and M. Gu, “Fibre-optic nonlinear optical microscopy and
endoscopy” J. Microsc. 226, 195-206, (2007).
Bibliography
108
[6] W.R Zipfel, R. M Williams and W. W. Webb, “Nonlinear magic:
Multiphoton microscopy in the biosciences,” Nat. Biotechno. 21, 1369-
1377, ( 2003).
[7] A. Diaspro, G. Chirico and M. Collini “ Two-photon fluorescence
excitation and related technique in biological microscopy,” Quat. Rev.
Biophys, 38, 2 (2005).
[8] D. Brid and M. Gu, “Resolution improvement in two-photon fluorescence
microscopy with a single-mode fibre”, Appl. Opt. 41, 1852-1856, (2002).
[9] L. Fu, X. Gan, and M. Gu, “Nonlinear optical microscopy based on double-
clad photonic crystal fibres”, Opt. Express. 13, 5528-5534, (2005).
[10] D.Yelin, B. E. Bouma, S.H.Yun, and G.J. Tearney , “ Double-clad fibre for
endoscopy”, Opt. Lett. 29 , 2408-2410, (2004).
[11] S-P Tai, M. C. Chan, T. H Tsai, S-H Guol, L-J Chen, and C-K Sun, “Two-
photon fluorescence microscope with hollow core photonics crystal fibre,”
Opt. Express 12, 6122-6128, (2004).
[12] D. Bird and M. Gu “Compact two-photon fluorescence microscope based
on a single-mode fibre coupler”, Opt. Lett. 27, 1031-1033, (2002).
[13] B. Ha Lee, J. B. Eom, K. S. Park , S. J. Park, and M. J. Ju, “ Specialty
Fibre Coupler: Fabrications and Applications”, J. of the Opt.Soc. of Korea,
14, 326-332, (2010).
Bibliography
109
[14] L. Fu, X. Gan, D. Bird, M. Gu, “ Polarisation characteristics of a 1x 2 fibre
coupler under femtosecond pulsed and continuous wave illumination” , Opt.
and Laser Technol. 37, 494-497, (2004)
[15] L. Fu, X. Gan, and M. Gu “Use of a single-mode fibre coupler for second-
harmonic-generation microscopy”, Opt. Lett. 30, 385-387, (2005).
[16] L. Fu and M. Gu “Double-clad photonic crystal fibre coupler for compact
nonlinear optical microscopy imaging” Opt. Lett. 31, 1471-1473, (2006).
[17] S. Y. Ryu, Hae Y. Choi, M.J.Ju, J. Na, W.J.Choi and B.Ha.Le, “ The
Development of double-clad fibre and double-clad fibre coupler for fibre
based biomedical imaging system”, J. Opt. Soc. Ko. 13 , 310-315 (2009).
[18] H. Bao, S. Y. Ryu, B. H. Lee, W. Tao, and M. Gu, “Nonlinear
endomicroscopy using double-clad fibre coupler ,” Opt. Lett. 35, 995-997,
(2010).
[19] W. Jung, S. Tang, D. T. McCormic, T. Xie, Y-C. Ahn, J.Su,I V. Tomov,
T.B. Krasieva, B. J. Tromberg, and Z. Chen, “Miniaturized probe based on
a microelectromechanical system mirror for multiphoton microscopy,” Opt.
Lett. 33, 1324-1326, (2008).
[20] D. R. Rivera, C. M. Brown, D. G. Ouzounov, I. Pavlova, D. Kobat, W. W.
Webb, and C. Xu, “Compact and flexible raster scanning multiphoton
endoscope capable of imaging unstained tissue,” PNAS, 108, 17598-17603,
(2011).
Bibliography
110
[21] Y. Kremer, J.-F. Léger, R. Lapole, N. Honnorat, Y. Candela, S. Dieudonné,
and L.Bourdieu, “A spatio-temporally compensated acousto-optic scanner
for two-photon microscopy providing large field of view,” Opt. Express 16
,10066-10076, (2008).
[22] S. Kane and J. Squier, “Grating Compensation of Third-Order Material
Dispersion in the Normal Dispersion Regime: Sub- 100-fs Chirped-Pulse
Amplification Using a Fibre Stretcher and Grating-Pair Compressor,”
Quan. Elect. 31, 2052-2057, (1995).
[23] R. Du, R. Jiang and L. Fu, “Enhanced dispersion compensation capability
of angular elements based on beam expansion,” Opt. Express 17, 16415-
16422, (2009).
[24] H. Bao and M. Gu, “ Reduction of self-phase modulation in double-clad
photonic crystal fibre for nonlinear optical endoscopy” , Opt. lett. 34, 148-
150, (2009).
[25] H. Bao, J. Allen, R. Pattie, R. Vance and M. Gu “Fast handheld two-photon
fluorescence microendoscope with a 475 μm - 475 μm field of view for in
vivo imaging” Opt. Lett. 33, 1333-1335, (2008).
[26] C. Xu and W. W. Webb “Measurement of two-photon excitation cross
sections of molecular fluorophores with data from 690 to 1050nm”, Opt.
Express 13, 481-491, (1996).
Bibliography
111
[27] M. A. Albota, C. Xu and W. W. Webb “Two-photon fluorescence
excitation cross sections of bimolecular probes from 690 to 960nm”, Appl.
Opt. 37, 7352-7356, (1998).
[28] Y. R. Shen, The Principles of Nonlinear Optics, John Wiley & Sons, New
York, 1984
[29] X. G. Xu, S. O. Konorov, J. W Hepburn, and V. Milner, “Background free
coherent Raman spectroscopy by detecting the spectral phase of molecular
vibration,” Opt. Lett. 33, 1177-1179, (2008).
[30] U. Utzinger, R. R. Richards-Kortum, “Fibre-optic probes for biomedical
optical spectroscopy”, J. biomed. Opt. 8, 121-147, (2003).
[31] L.Wang, H.Y. Choi, Y. Jung, B.H. Lee, K-T. Kim, “Optical probe based on
double-clad optical fibre for fluorescence spectroscopy”, Opt. Express 15,
17681-17691, (2007).
[32] Y. Wu and X. Li , “Combined influences of chromatic aberration and
scattering in depth-resolved two-photon fluorescence endospectroscopy”, J.
Biomed. Opt. Express 1, 1234-1243, (2010).
[33] F. Helmchen and W. Denk, “Deep tissue two-photon microscopy”, Nature
Meth. 2, 932-940, (2005).
[34] Denk, W., Strickler, J. H. & Webb, W.W. “ Two-photon laser scanning
fluorescence microscopy”. Science 248, 73-76, (1990).
Bibliography
112
[35] J. M Girkin and D. Wokosin, Practical multi-photon microscopy In
confocal and Two-photon microscopy : Foundation, Applications and
Advances (Wiley-Liss & sons Inc., 2001).
[36] M. Gu and L. Fu “Three-dimensional image formation in fibre-optical
second-harmonic-generation microscopy”, Opt. Express 14, 1175-1181,
(2006).
[37] L. Fu, A. Jain, H. Xie, C. Cranfield and M. Gu “ Three-dimensional
nonlinear optical endoscopy” , SPIE 4, 400501 (1-3), ( 2007 ).
[38] M. Gu, “Principles of Three dimensional imaging in confocal
microscopes”, world scientific publishing, 1-274, (1996).
[39] G. Chirco, F. Cannone, Baldini G and Diaspro A “ A two-photon thermal
bleaching of single fluorescent molecules” , J. Biophys. J. 84, 588-598,
(2003b).
[40] W-F. Cheong, S.A.Prahl, and A.J.Welch, “ A Review of the optical
properties of Biological Tissue”, J. Quantum. Electron. 26 , (1990).
[41] R.K Wang, “Modeling optical properties of soft tissue by fractal
distribution of scatterers”, J. Modern Opt., 47 ,103-120, (1999).
[42] C.Mar Blanca and C. Saloma, “Monte Carlo analysis of two-photon
fluorescence imaging through a scattering medium”, Appl. Opt., 37, 8092-
8102, (1998).
Bibliography
113
[43] X. Deng, X. Gan and M. Gu, “Monte carlo simulation of multiphoton
fluorescescence microscopic imaging through inhomogeneous tissuelike
turbid media”, J. Biomed. Opt. 8 ,440-449, (2003).
[44] H.C. Gerristsen, C.J. DeGrauw, “Imaging of optically thick specimen using
two-photon excitation microscopy” Microsc. Res Tech. 47, 206-209,
(1999).
[45] J. W. McLean, J. D Freeman, R. E. Walker, “Beam spread function with
time dispersion”, Appl. Opt. 37, 4701, (1998 ).
[46] G.A Blab, P.H Lommerse , L, Cognet, G. S Harms and T. Schmidt “ Two-
photon excitation action cross-section of the auto fluorescent proteins”
Chem. Phys. Lett. 350, 71-77, (2001).
[47] E. Abbe, Gesammelte abhandlugen ( G. Fischer Jena, 1904)
[48] Q. Li, S. S.H. Wu, K.C.Chou, “Subdiffraction-limit two-photon
fluorescence microscopy for GFP-tagged cell imaging”, Biophys. J. , Vol.
97, 3224-3228, (2009).
[49] T. A. Klar and S. W. Hell, “subdiffraction resolution in far-field
fluorescence microscopy”. Opt. Lett. 24, 954-956, (1999).
[50] P.Dedecker, C.Flors, J-I Hotta, H. Uji-I and J. Hofkens “3D
Nanoscopy:bringing biological nanostructures into sharp focus”,
Angew.Chem Int. 46,8330-8332 ,(2007).
Bibliography
114
[51] E.Betzig, G.H. Patterson, R. Sougrat, O.W. Lindwasser, S. Olenych, J. S
Bonifacino, M. W. Davidson, J. Lippincott-Schwartz, H.F. Hess, “Imaging
intracellular fluorescent proteins at nanometer resolution”,Science
313,1642-1645, (2006).
[52] P.Dedecker, J. Hofkens, and J-I Hotta, “Diffraction-unlimited optical
microscopy”, Mater. Today, 11, 12-21, (2008).
[53] E. Haustein and P. Schwille, “Trends in fluorescence imaging and related
techniques to unravel biological information”, HFSP J., 1(3), 169-180,
(2007).
[54] M. Gu , T. Tannous, and C. J. R. Sheppard, “ Theree dimensional confocal
fluorescence imaging under ultrashort pulse illumination” Opt. commun.
117, 406-412, (1995).
[55] G.cox and C. J.R. Sheppard, “Practical limits of Resolution in confocal and
Non-linear Microscopy”,Microsc. Res. and Tech. 63, 18-22 (2004).
[56] G. donnert et. al ,“Macromolecular-scale resolution in biological
fluorescence microscopy”, PNAS, 103, 11440- 11445, (2006).
[57a] B.Huang, M.Bates, and X.Zhuang, “Super-Resolution Fluorescence
Microscopy”, Annual Rev. Biochem. 78, 993-1016, (2009).
[57b] B. R. Masters, P. T. C. So, “Confocal microscopy and multi-photon
excitation microscopy of human skin in vivo”, Opt. Express 8, 2-10,
(2001).
Bibliography
115
[58a] D.Wildanger, E.Rittweger, L. Kastrup and S.W.Hell, “STED microscopy
with a supercontinuum laser source”, Opt. Express 16, 9614-9621, (2008).
[58b] W. Denk, K. Svoboda, “Photon Upmanship : Why Multiphoton imaging is
more than a Gimmick”, Neuron 18, 351-357, (1997).
[59] V. Westphal and S. W. Hell, “Nanoscale resolution in the focal plane of an
optical microscope, “Phys. Rev. Lett. 94, 143903, (2005).
[60] B.P, Cormack, R.H. Valdivia and S. Falkow, “FACS-optimized mutants of
the green fluorescent protein (GFP)”, Gene, 173, 33-38, (1996).
[61] K. I Willing, S.O. Rizzoli, VWestphal, R. Jahn, and S.W Hell, “STED
microscopy reveals that synaptotagmin remains cluster after synaptic
vesicle exocytosis” , Nature. 400,935-939, (2006).
[62] G. Vicidomini, G. Moneron, C.Eggeling, E. Rittweger, and S.W Hell,
“STED with wavelengths closer to the emission maximum”, Opt. Express.
20, 5225-5236, (2012).
[63] J. R Moffitt, C. Osseforth, and J. Michaelis, “Time-gating imporves the
spatial resolution of the STED microscopy,” Opt. Express 19, 4242-4254,
(2011).
[64] G. Vicidomini, G. Moneron, K.Y. Han, V. Westphal, H. Ta, M. Reuss, J
Engelhardt, C. Eggeling, and S. W. Hell, “ Sharper low-power STED
nanoscopy by time gating, : Nat. Methods 8, 571-573, (2011).
Bibliography
116
[65] S.T Hess, T.P.K Girirajan, M. D.Mason, “Ultra-High Resolution Imaging
by Fluorescence Photoactivation Localization Microscopy” Biophys J., 91,
4258-4272, (2006).
[66] M.J.R. Rust, M.B. Bates, X.Z.Zhuang, “Sub-diffraction-limit imaging by
stochastic optical reconstruction icroscopy (STORM)”,Nat. Methods, 3,
793-796, (2006).
[67] M. Aeschlimann,m.bauer, D.Bayer, T.Brixner, F.J. G.d.Abajo,W.Pfeiffer,
M.Rohmer, C.spindler and F.Steeb, “Adaptive subwavelength control of
nano-optical fields”, Nature 446, 301-304, (2007).
[68] R.Menon, P Rogge and H-Y Tsai, “Design of diffractive lenses that
generate optical nulls without phase singularities”, Opt. Soc. Am. 26, 297-
304, (2009).
[69] D.Wildanger, R. Medda, L.Kastrup and S.W.Hell, “A compact STED
microscope providing 3D nanoscale resolution”, J. Microsc. 236, 33-43,
(2009).
[70] Baer, S., “Method and apparatus for improving resolution in scanned
optical system,” U.S. Patent #5,866,911,(filed July 15, 1994, published Feb.
29, 1996 as International Application WO 9606369A2, issued Feb. 2, 1999)
[71] S. W. Hell and J. Wichmann, “Breaking the diffraction resolution limit by
stimulated emission: Stimulated emission - depletion fluorescence
microscopy”, Opt. Lett. 19, 780-782, (1994).
Bibliography
117
[72] C. Eggeling, C. Ringemann, R. Medda, G.Schwarzmann, K. Sandhoff, S.
Polyakova, V.N. Belov, B. Hein, C. Von. Viddendorff, A. Schonle, and S.
W. Hell, “ Direct observation of the nanoscale dynamics of membrance
lipids in a living cell,” Nature 457 ,1159-1162, (2009).
[73] J. B. Ding, K. T. Takasaki and B.L. Sabatini, “Supraresolution imaging in
brain slices using stimulated-emission depletion two-photon laser scanning
microscopy,” Meirpm 63,429-437 (2009).
[74] T. Grotjohann, I. Testa, M. Leutenegger, H. Bock, N. T. Urban, F. Lavoie-
Cardinal, K. I. Willing, C. Eggeling, S. Jakobs, and S. W. Hell, “
Diffraction-unlimited all-optical imaging and writing with a photochromic
GFP,” Nature 478,204-208, (2011).
[75] J.Huisken, j.Swoger, F.D Bene, J.Wittbrodt, E .H.K.Stelzer, “Optical
sectioning deep inside live embroys by selective plane illumination
microscopy”,Science, 305, 1007-1009, (2004).
[76] G.Cristobal, E.gil, F.sroubek, J.Flusser, D.Miravet, F.B.Rodriguez,
“Superresolution imaging: a survey of current techniques”, Proc. of SPIE
7074, 0C-18, (2008).
[77] Ilan Davis, “ The ‘Super Resolution’ revolution”, Biochem. Soc. Transac.,
37, part 5 (2009).
[78] Gustaffsson. MGL ,“Nonlinear sturctured-illumination microscopy: Wide-
field fluorescence imaging with theoretically unlimited resolution”, PNAS
102, 13081-13086, (2005).
Bibliography
118
[79] G.Zhu, J.V.Howe, M.Durst, W.Zipfel and C.Zu, “Simultaneous spatial and
temporal focussing of femtosecond pulses “, Opt. Express, 13, 2153-2159,
(2005)
[80] D.M.D Juana, J.E.Oti,Vidal F.Canales and M.P.Cagigal, “Transverse or
axial superresolution in a 4Pi-confocal microscope by phase-only filters”,
Opt. Soc. Am. 20 , 2172-2178, (2003).
[81] P.E.Hanninen, S.W. Hell, J. Salo, E.Soini, C.Cremer, “Two-photon
excitation 4Pi confocal microscope: Enhanced axial resolution microscope
for biological research”, Appl. Phys. Lett. 66, 1698-1700, (1995)
[82] Hell, S.W “Towards fluorescence nanoscopy”, Nature Biotechnology, 21,
1347-1355 (2003).
[83] M. Fernandex-suarez and A. Y. Ting, “Fluorescent probes for super-
resolution imaging in living cells”, Nature Rev. Molec. cell Bio. 9 , 929-
9439, (2008).
[84] S. Zhuo, J. Yan, G. Chen, J. Chen, Y. Liu, J. Lu, X. Zhu, X. Jiang and
S.Xie,“Label-free monitoring of colonic cancer progression using
multiphoton microscopy” J. of Biomed. Opt. Express 2, 615-619, (2011).
[85] W. Zheng. , W. Lau, C. Cheng, KC Soo, M Olivo , “ Optimal excitation-
emission wavelength for autofluorescence diagnosis of bladder tumors” ,
Internat. J. Cancer 104, 477-481, (2003).
[86] M.Z. Lin, M. R. McKewon, Ng HL, T. A. Aguilera, N.C. Shaner, R.E.
Campbell, S.R Adams, L.A Gross, W Ma, T Alber , R.Y Tsien , “
Bibliography
119
Autofluorescent Proteins with excitation in the optical window for intravital
imaging in Mammals”, Chem. and Biol. 16, 1169-1179, (2009).
[87] P. Bianchinia, B. Harkea, S. Galiania, G. Vicidominia, and A. Diaspro
“Single-wavelength two-photon excitation–stimulated emission depletion
(SW2PE-STED) superresolution imaging”, PNAS 109, 6390–6393, (2012).
[88] T. scheul, C. D. Amico, I. Wang and J-C. Vial, “Two-photon excitation and
stimulated emission depletion by a single wavelength”, Opt. Exp. 19,
18036-18041, (2011).
[89] D. Baddeley, D. Crossman, S. Rossberger, J. E. cheyne, J. M. Montgomery,
I. D Jayasinghe, C. Cremer, M. B Cannell, C. Soeller “ 4D Super-resolution
microscopy with conventional fluorophores and single wavelength
excitation in optically thick cells and tissues”, PLoS ONE, 6, 1-10, (2011).
[90] I.Iiev , R.Waynant, I.Gannot, A.Gandjbakhche, “Simple fibre-optic
confocal microscopy with nanoscale depth resolution beyond the diffraction
barrier”, Rev. Scient. Inst. 78, 093703, (2007).
[91] Y.Wu,Y.Leng, J.Xi and Z.Li, “Scanning all-fibre-optic endomicroscopy
system for 3D nonlinear optical imaging of biological tissues”, Opt.
Express 17, 7907 – 7915, (2009).
[92] D. Bird and M. Gu “Fibre-optic two-photon scanning fluorescence
microscopy”, J. Microsc. 208, 35-48, (2002).
Bibliography
120
[91] J. C. Knight, J. Arriaga, T.A. Briks, O. Blanch, W.J. Wadsworth and P.St. J.
Russell “ Anomalous Dispersion in Photonic Crystal Fibre”, IEEE Phot.
Technol. Lett. 12, 807-809, (2000).
[92] Y-Z Xu, X-M.Ren, Z.Xia, Y-Q Huang, “ A fully vectroial effective index
method for Accurate dispersion calculation of photonic crystal fibres”,
China Phys. Lett. 23 (2006).
[93] W.Jung, S. Tang, T. Xie, D. T. McCormick, Y.Ahn, J.Su, I. V. Tomov.
T.B.Krasieva, B. J.Tromberg, Z. Chen “Miniaturized probe using 2 axis
MEMS scanner for endoscopic multiphoton excitation microscopy”, Proc.
of SPIE 6851, 0D-7, (2008).
[94] G.A. Stanciu, C. Stoichita, S.G Stanciu, “Scanning laser microscopy : From
Far field to Near Field”, IEEE, ICTON (2009).
[95] B. A Flusberg, E. D Cocker, W. Piyawattanemetha, J. C Jung, E. L M
Chung and M. J Schnitzer , “Fibre-optic fluorescence imaging”, Nature
Meth. 12, 941-949, (2005).
[96] J. C. Knight, J. Arriaga, T.A. Briks, O. Blanch, W.J. Wadsworth and P.St. J.
Russell “Anomalous Dispersion in Photonic Crystal Fibre”, IEEE Phot.
Technol. Lett. 12, 807-809, (2000).
[97] Y-Z Xu, X-M.Ren, Z.Xia, Y-Q Huang, “ A fully vectroial effective index
method for Accurate dispersion calculation of photonic crystal fibres”,
China Phys. Lett. 23 (2006).
Bibliography
121
[98] H. Bao and M. Gu “A 0.4 mm diameter probe for nonlinear optical
imaging”, Opt. Express 17, 10098-10104, (2008).
[99] L. Fu, A. Jain, H. Xie, C. Cranfield and M. Gu “ Nonlinear optical
endoscopy based on a double-clad photonic crystal fibre and a MEMS
mirror” ,Opt. Express 14, 1027-1032, (2006).
[100] H. Kano and H. Hamaguchi, “In vivo multi-nonlinear optical imaging of a
living cell using a super continuum light source generated from a photonic
crystal fibre” Opt. Express 14, 2798-2804, ( 2006).
[101] L. Fu and M. Gu “Polarization states Anisotropy in Fibre-optic second
harmonic generation microscopy” Opt. Express 16, 5000-5006, (2008).
[102] R. L. Harzic, M. Weinigel, I. Riemann, K. König, and B. Messerschmidt,
“Nonlinear optical endoscopy base on a compact two axis piezo scanner
and a miniature objective lens,” Opt. Express 16, 20588-20596, (2008).
[103] C. L. Hoy, O. Ferhanoğlu, M. Yildirim,W. Piyawattanametha, H. Ra, O.
Solgaard, and A. Ben-Yakar, “Optical design and imaging performance
testing of a 9.6-mm diameter femtosecond laser microsurgery probe”, /
Opt. Express 19 , 10536-10552, (2011).
[104] G.P Agrawal, Nonlinear fibre-optics, Academic Press, (2007)
[105] K. Thyagarajan and B.P.Pal , “Modeling dispersion in optical fibres:
applications to dispersion tailoring and dispersion compensation”, J. Opt.
fibre. Commun. Rep.4, 173-213, (2007).
Bibliography
122
[106] J.Kim, D.K, “An efficient dispersion calculation method for axially
symmetric optical fibres”, Fibre and Integ. Opt., 21,13-29, (2002).
[107] W.A.gambling, H.Matsumura and C.M.Ragdale, “Mode dispersion,
material dispersion and profile dispersion in graded-index single-mode
fibres”, Microwaves, Opt. and Acoust., 3, (1979).
[108] Y. Painchaud, C.Paquet and M.Guy, “FBGs enhance dispersion
compensation”, Laser Focus World, 145-147, (2007).
[109] S. Ramachandran, B. Mikkelsen, L.C. Cowsar M. F. Yan, G. Raybon, L.
Boivin, M. Fishteyn, W.A. Reed, P.Wisk, D. Brownlow, R. G Huff and L.
Gruner-Nielsen , “ All-Fibre Grating-Based Hingher Order Mode
Dispersion Compensator for Broad-Band Compensation and 1000-km
Transmission at 40Gb/s” IEEE Phot. Technol. Lett. 13, 632-634, (2001).
[110] W. Yi, W. Ke-jia, W. Xiu-min, Z. Shuai, “The Heterodyne Research on
Transmission Signal of Three-layered Cascaded Long-Period Fibre
Gratings”, Proc. of SPIE 7381, 1R-10, (2009).
[111] X. Chu, S Zhao, Z. Wu, L. Shi, S Zhan, Y. Li, L Ma, “ New method for side
pumping of double-clad fibre lasers using planar grating written in the inner
cladding”, Proc. of SPIE 7382, 0X-7, (2009).
[112] H.Ke, K.S. Chiang, and J.H.Peng, “Analysis of Phase-shifted Long-Period
Fibre Gratings”, IEEE Photon. Technol. Lett. 10, 1596-1598, (1998).
Bibliography
123
[113] M. Hany, M.H.aly and M.Nassr, “Dispersion compensation in Transmission
using AFBG Chirped with Two Beam Interference Fringe Spacing”, J. App.
Sci. Res. 5, 1622-1629, (2009).
[114] B. Y. Kim, J. N. Blake, H. E, Engan, and H. J. Shaw, “All-fibre acousto-
optic frequency shifter”, Opt. Lett 11, 389-391, (1986).
[115] R. C. Youngquist, J. L. Brooks, and H. J. Shaw, “ Birefringence-fibre
polarization states coupler,” Opt. Lett. 8, 656-659, (1983).
[116] Bashir A. T, Jalil A. , Rosly A. R. “ Fabrication of fibre grating by phase
mask and it’s sensing application” , J. Opto.Elec. and Adv. Mat. 8, 1064-
1609, (2006).
[117] Yuki K. ,Kentaro N. and Tsuneo. M , “Fabrication of long-period fibre
gratings by focused irradiation of infrared femtosecond laser pulses”, Opt.
Lett. 24, 646-648, (1999).
[118] Y. Kondo, K. Nouchi, T. Mitsuyu, M. Watanabe, P. Kazansky, and K.
Hirao, “ Fabrication of long period fibre grating by focused irradiation of
infrared femtosecond laser pulses”, Opt. Lett. 24, 646-648, (1999).
[119] P.G. Kryukov, Yu.V. Larionov, A. A. Rybaltovskii, K.A.Aagorul’ko, A.
Dragomir, D.N. Nikogosyan, A.A. Ruth, “Long-period fibre grating
fabrication with femtosecond pulse ration at different wavelength”, Micro-
electron. Eng. 69 , 248-255 (2003).
[120] T. Erdogan, “Cladding-mode resonances in short- and long-period fibre
grating filters”, J. Opt. Soc. Am. 14, 1760-1773, (1997).
Bibliography
124
[121] J. P. Ying, F. Liu, and R.R. Alfano, “Spatial distribution of two-photon-
excited fluorescence in scattering media: erratum” Appl. Opt. 38, 2151-
2156, (1999).
[122] P. Theer, M. T. Hasan and W. Denk “ Two-photon imaging to a depth of
1000 μm in living brains by use of a 32: OAlTi regenerative amplifier”
,Opt. Lett. 28, 1022-1024, (2003).
[123] M. Oheim , E. Beaurepaire, E. Ghaigneau, J. Mertz and S. Charpak “Two-
photon microscopy in brain tissue: parameters influencing the imaging
depth”, J. Neurosci. Meth. 111, 29-37 (2001).
[124] E. Beaurepaire, M Oheim, J. Mertz “Ultra-deep two-photon fluorescence
excitation in turbid media” , Op. Comm. 188, 25-29, (2001).
[125] C.Wang, L.Oiao, Z. Mao, Y.Cheng, and Z.Xu, “Reduced deep-tissue image
degradation in three-dimensional multiphoton microscopy with concentric
two-color two-photon fluorescence excitation”, J. Opt. Soc. Am., 25,
(2008).
[126] Pang, S., Yeh, A. T., Wang, C. Meissner, K. E. “Beyond the 1/Tp limit:
two-photon excited fluorescence using pulses as short as sub-10-fs” J.
Biomed. Opt. 14, 054041, (2009).
[127] Tang, S. Krasieva, T. B., Chen Z., Tempea, G., Tromberg, B. J. “Effect of
pulse duration on two-photon excited fluorescence and second harmonic
generation in nonlinear optical microscopy” J. Biomed. Opt. 11, 020501,
(2006).
Bibliography
125
[128] Larson, A. M. and Yeh, A. T. “Ex-Vivo characterization of sub-10-fs
pulses”, Opt. Lett. 31, 1681–1683, (2006).
[129] D.Wildanger, J.Buckers, V.Westphal, S.W.Hell and L.Kastrup, “A STED
microscope aligned by design” , Opt. Express, 17, 16100-16110, (2009).
[130] D. Kim, H. Choi, S. Yazdanfar, and P. T. C. So, “Ultrafast optical pulse
deliver with fibres for nonlinear microscopy,” Micros. Res. Tech. 71, 887-
896, (2008).
[131] D. G. Ouzounov, K. D. Moll, M. A. Foster, W. R. Zipfel, W. W. Webb, and
A. L. Gaeta, “Delivery of nanojoule femtosecond pulses through large-core
microstructured fibres,” Opt. Lett. 27, 1513-1515, (2002).
[132] Shephard, J.D., Macpherson, W.N., Maier, R.R.J., Jones, J.D.C.,
Hand,D.P., Mohebbi, D.P., George, A.K., Roberts, P.J., and Knight,
J.C.:‘Single-mode mid-IR guidance in a hollow-core photonic crystal
fibre”,Opt. Express 13, 7139–7144, (2005)
[133] P. J Mosley, W. C. Huang, M. G. Welch, B. J. Mangan, W. J. Wadsworth
and J. C. Knight, “Ultrashort pulse compression and delivery in a hollow-
core photonic crystal fibre at 540 nm wavelength”, Opt. Lett. 35, 3589-
3591, (2010).
[134] F. Benabid, “Hollow-core photonic bandgap fibre: new light guidance for
new science and technology,” Philos Transact A Math Phys Eng Sci., 364,
3439-3462, (2006).
Bibliography
126
[135] J. C. Knight, F. Gérôme, and W. J. Wadsworth, “Hollow-core photonic
crystal fibres for delivery and compression of ultrashort optical pulses,”
Opt. Quant. Elect. 39 (12-13), 1047-1056, (2007).
[136] T. Le, M. Hofer, Z. Cheng, and A. Stingl, J. Darmo, D.P. Kelly, and K.
Unterrainer,“Advances in Fibre Delivery of Ultrashort Pulses at 800 nm”,
Proc. of SPIE 6871, 68712C, (2008)
[137] D. H. Kim, J. U. Kang, and I. K. Ilev, “Advance confocal microscope using
single hollow core photonic bandgap fibre design,” Elect. Lett., 43, 608-
609, (2007).
[138] W. Göbel, A. Nimmerjahn, and F. Helmchen, "Distortion-free delivery of
nanojoule femtosecond pulses from aTi:sapphire laser through a hollow-
core photonic crystal fibre," Opt. Lett. 29, 1285-1287, (2004).
[139] C. L Hoy, N. J Durr, P. Y Chen, W. Piyawattanametha, H. Ra, O. Solgaard,
and A. Ben-Yakar, “Miniaturized probe for femtosecond laser microsurgery
and two-photon Imaging,” Opt. Express 16 , 9996-10005, (2008).
[140] Y. C. Wu, J. F. Xi, M. J. Cobb, and X. D. Li, “Scanning fibre-optic
nonlinear endomicroscopy with miniature aspherical compound lens and
multimode fibre collector,” Opt. Lett. 34, 953-955, (2009).
[141] C. J. Engelbrecht, R. S. Johnston, E. J. Seibel, and F. Helmchen, “Ultra-
compact fibre-optic two-photon microscope for functional fluorescence
imaging in vivo,” Opt. Express 16, 5556-5564, (2008).
Bibliography
127
[142] S. K.Vengalathunadakal,V.M. Murukeshan, S. Padmanabhan, P.
Padmanabhan,“Photonics crystal fibre based dual modality probe for
simultaneous sensing and imaging application”, Opt. Eng. 48, 103601-
103604, (2009)
[143] R.S.Pillai, C.Boudoux, g.Labroille, N.Olivier, I.Veilleux, E.Farge, M.Joffre,
and E.Beaurepaire, “Multiplexed two-photon microscopy of dynamic
biological sampless with shaped broadband pulses”, Opt. Express 17,
12741-12752, (2009).
[144] S. W. Hell and J. Wichmann, “Breaking the diffraction resolution limit by
stimulated emission depletion fluorescence microscopy ”, Opt. Lett. 19,
780-782, (1994).
[145] Willig, K. I., Harke, B., Medda, R. and Hell, S. W., “STED microscopy
with continuous wave beams,” Nat.Methods 4, 915-918, (2007).
[146] J. B. Ding, K. T. Takasaki and B.L. Sabatini, “Supraresolution imaging in
brain slices using stimulated-emission depletion two-photon laser scanning
microscopy,” Meirpm 63, 429-437, (2009).
[147] Baer, S. C., “Forming light beams and patterns with zero intensity central
points,” US Patent application 20100315709, (priority filing date 2007)
(2007).
Bibliography
128
[148] Morozova, K. S., Piatkevich, K. D., Gould, T. J., Zhang, J., Bewersdorf, J.
and Verkhusha, V. V., “Far-RedFluorescent Protein Excitable with Red
Lasers for Flow Cytometry and Superresolution STED
Nanoscopy,”Biophys. J. 99, L13–L15, (2010).
[149] Drobizhev, M., Tillo, S., Makarov, N. S., Hughes, T. E., and A. Rebane, A.,
“Absolute Two-Photon Absorption Spectra and Two-Photon Brightness of
Orange and Red Fluorescent Proteins,” J. Phys. Chem. B. 113, 855–859,
(2009).
[150] Leutenegger, M., Eggeling, C. and Hell, S. W., “Analytical description of
STED microscopy performance,” Opt.Express 18, 26417-26429, (2010).
[151] Min Gu, Hong Kang & Xiangping Li , “Breaking the diffraction-limited
resolution barrier in fibre-optical two-photon fluorescence endoscopy by an
azimuthally-polarized beam”, Sci. Rep. 4, 3627-3637, ( 2014).
[152] Baer, S., “Single wavelength stimulated emission depletion microscopy,”
U. S. Patent #7,816,654, (priority filing date Jan. 16, 2005, published July
27, 2006 as International Application WO2006078857, issued Oct.
19,2010) (2005).
[153] A. Small, “Model of bleaching and acquisition for superresolution
microscopy controlled by a single wavelength”, Biomed. Opt. Express 2,
2934-2949, (2011).
[154] S. C. Baer, “Single wavelength STED microscope”, Proc. of SPIE , Vol.
7903, doi:10.1117/12.875005, 1-8, (2011).
Bibliography
129
[155] Z.Xiaoping, T.Zhihong, “Analysis of transmission characteristics of doubly
clad fibres with an inner cladding made of uniaxial crystal materials”, Opt.
Commun. 204, 127-136, (2002).
[156] L. Fu, A. Jain,C.Cranfield, H.Xie and M. Gu, “Three-dimensional nonlinear
optical endoscopy,” J. Biomed. Opt. 12, 040501-040509 (2007).
[157] L. Fu, X. Gan and M. Gu “ Characterization of gradient-index lens-fibre
spacing toward applications in two-photon fluorescence endoscopy” , Appl.
Opt. 44, 7270-7274, (2005).
[158] Chi Wang, Youxin Mao, Chen Fang,Zhi Tang, Yingjie Yu, Bo Qi,
“Analytical method for designing gradient-index fibre probes”, Opt. Engg. ,
50, 094202, (2011)
[159] K. J. Jespersen, T. Le, L. G. Nielsen, D. Jackobsen, M. E. V. Pederesen, M.
B. Smedemand, S. R. Keiding, B. Palsdottir, “A higher-order-mode fibre
delivery for Ti:Sapphire femtosecond lasers” Opt. Express 18, 7798-7806,
(2010)
[160] S. Ramachandran, M. F. Yan, Jasapara, P. wisk, S. Ghalmi, E. Monberg,
and F. V. Dimarcello, “High-energy (nanojoule) femtosecond pulse
delivery with record dispersion higher-order mode fibre”,Opt. Lett. 30,
3225-3227, (2005)
[161] R.Siddhartha, “Dispersion-Tailored Few- Mode Fibres: A Versatile
Platform for In-Fibre Photonic Devices”, J. lightwave tech., 23, 3426-3443,
(2005).
Bibliography
130
[162] J. W. Nicholson, S. Ramachandran and S. Ghalmi “ A Passively-
modelocked, Yb-doped, figure-eight, fibre laser utilixing anomalous-
dispersion highter-order-mode fibre” Opt. Express 15, 6623-6628, (2007).
[163] S. Ramachandran, S. Ghalmi, J.W. Nicholson, M. F. Yan, P. Wisk, E.
Monberg and F. V. Dimarcello “ Anomalous dispersion in a solid silica
based fibre” , Opt. Lett. 31, 2532-2534, (2006 ).
[164] M. Schultz, O. Prochnow, A. Ruehl, D. Wandt, D. Kracht, S.
Ramachandran and S. Ghalmi “ Sub-60-fs yetterbium-doped fibre laser
with a fibre based dispersion compensation”, Opt. Lett. 32, 2372-2374,
(2007).
[165] S. Ramachandran, J.W.Nicholson, S. Ghalmi, M.F. Yan, P. Wisk, E.
Monberg and F. V. Dimarcello, “ Light propagation with ultralarge modal
areas in optical fibres” , Opt. Lett. 31, 1797-1799, (2006).
[166] P Hlubina, “Group velocity dispersion in fused-silica samples measured
using white-light interferometry with the equalization wavelength
determination”,Optik 113, 149-152, (2002).
[167] C-X Shi, T Okoshi, “Analysis of fibre-optic LP01-LP02 mode converter,”
Opt. Lett. 17, 719-721, (1992).
[168] M. Sumetsky and S. Ramachandran, “Conversion and focusing of optical
fibre modes with superimposed long period gratings”,OThV3,
OFC/NFOEC, (2008).
Bibliography
131
[169] H.Shu and M.Bass, “Calculating the guided modes in optical fibres and
waveguides”,J. Lightwave Tech. 25, (2007).
[170] M-Y chen, J Zhou, “Mode converter based on mode coupling in an
asymmetric dual-core photonic crystal fibre”, J. Appl. Opt. 10, (2008).
[171] A.C Turner, C.manolatou, B S Schimdt, M. Lipson, M.A Foster, J. E
Sharping, A.L. Gaeta “Tailored anomalous group-velocity dispersion in
silicon channel waveguides”, Opt. Express 14, 4357-4362, (2006).
[172] D. Gloge, “Dispersion in weakly guiding fibres”, Appl. Opt. 10, 2442-2445,
(1971).
[173] J. Yao, L.Zhan, Zh. Luo, and Y. Xia, “Modeling and Numerical
Demonstration of Femtosecond-pulse Distortion-Free Delivery Using a
Three-Layered Few-Mode Fibre”, Laser Phys., 18, 455-459, (2008).
[174] S. Ramachandran, J.W.Nicholson, S. Ghalmi, M.F. Yan, P. Wisk, E.
Monberg and F. V. Dimarcello, “ Light propagation with ultralarge modal
areas in optical fibres” , Opt. Lett. 31, 1797-1799, (2006).
[175] H. J. Patric, C. G. Askins, R. W. Mcelhanon, and E. J. friebele, “ Amplitude
mask patterned on an excimer laser mirror for high intensity writing of long
period fibre gratings”, Electron. Lett. 33, 1167-1168, (1997).
[176] Oleg V. Ivanov, “Fabrication of long-period fibre gratings by twisting a
standard single-mode fibre”, Opt. Lett., 30 , 3290-3292, (2005).
[177] A. Cusano, D. Paladino, A.Cutolo, I.D,Villar, I.R.Matias and F. J. Arregui,
“Spectral Characteristics in long-period fibre gratings with nonuniform
Bibliography
132
symmetrically ring shaped coating”, Appl. Phys. Lett. 90, 14110-14115,
(2007).
[178] L. Li, W.Fan, S. Fu, Y. Liu, S. Yuan, X.Dong, “Fabrication and
characteristics of double-clad fibre Bragg Gratings”, Front. Optoelectron.
China 2 , 68-75, (2009).
[179] H.Chen, “Design and simulation of broadband IR integrated waveguide
amplifiers with long-period waveguide grating filter”,J Infrared Mili Terahz
Waves 31, 24-30, (2010).
[180] T.W. MacDougall, S.Pilevar, C. W. Haggans, and M.A.Jackson,
“Generalized Expression for the growth of long period gratings”, IEEE
photon. Technol. Lett. 10, 1449-1451, (1998).
[181] C-L.Lee and P.Han, “Optimal design of single resonant and ultra
broadband long-period fibre grating filters”,Opt. Eng. Lett., 48, 080501-3,
(2009).
[182] H.Chen, “Design and simulation of Broadband IR integrated waveguide
amplifiers with long-period waveguide grating filter”,J Infrared Mili Terahz
Waves 31, 24-30, (2010).
[183] F. Bilodeau, K.O. Hill, B.Malo, D.C. Johnson, and I. Skinner, “Efficient,
narrowband LP01<—->LP02 mode converters fabricated in photosensitive
fibre : spectral response”, Elec. Lett. 27, 682, (1991).
Bibliography
133
[184] Y. J. Rao, T. Zhu, Z. L. Ran, Y. P. Wang, J. Jiang, and A.Z. Hu, “ Novel
long-period fibre gratings written by high-frequency CO2 laser pulses and
applications in optical fibre communication”, Opt. Commun., 229 , 209-
221, (2004).
[185] C-L. Lee, R-K. Lee, Y-M Kao, “Synthesis of long –period fibre gratings
with a lagrange multiplier optimization method”, Opt. Commun. 281, 61-
74, (2008).
[186] A.I.Kalachev, D.n. Nikogosyan and G.Brambilla, “Long-period fibre
grating fabrication by high-intensity femtosecond pulses at 211nm”, J.
Lightwave Tech. 23 , 2568-2578, (2005).
[187] K-W Chung , S. Yin, “Analysis of a widely tunable long-period grating by
use of an ultrathin cladding layer and higher-order cladding mode
coupling”, Opt. Lett. 29, 812-814, (2004).
[188] M. Fuumaki, Y. Phki, J. L. Brebner, and S. Roorda “ Fabrication of long-
period fibre gratings by use of ion implantation”, Opt. Lett. , 25, 88-89,
(2000).
[189] N. H. Ky, H.G. Limberger and R. P. Salathe “Efficient broadband intracore
grating LP01-LP02 mode converters for chromatic-dispersion
compensation”, Opt. Lett. 23 , 445-447, (1998).
[190] T. Nielsen , M. Fricke, D. Hellweg and P. Andresen “ High efficiency beam
splitter for multifocal multi-photon microscopy” J. Microsc. 201, 368-376,
(2001).
Appendix
134
Appendix
Dispersion relation for cladding modes
The dispersion relation equations for the cladding modes are given as [120]
)],()()()([
*)]()()()([
)]()()()()([
*)](1
)()()()([
2
1
21
2
2
32
2
1
2
2
2
321
2
32
2
2
2
2
1
21
2
2
2
1
32
2
2
2
1
21
2
2
2
32
21
2
2
2
1
2
2
222
1
2
3
2
21
2
2
322121
2
3
2
2
2
2
2
222
21
2
2
322121
2
ara
uaq
a
uapK
an
nuJ
a
uu
aran
uaq
an
uapK
an
uJ
an
uu
asun
naJraKq
n
nap
aan
uuJK
n
nu
asu
aJraKqapaan
uuJKu
vvv
vvv
vvvv
vvvv
(A.1)
The following definitions are used in above equation.
,/ 01 Zivn cleff (A.2)
Appendix
135
02 Zivn cleff , (A.3)
,11
2
1
2
2
21uu
u (A.4)
2
2
2
3
31
11
uwu , (A.5)
2,1),()/2( 2222 innu cleffii (A.6)
),()/2( 2
3
222
3 nnw cleff (A.7)
,)(
)(
111
11
'
auJu
auJJ
v
v
(A.8)
,)(
)(
233
23
'
awKw
awKK
v
v
(A.9)
),()()()()( 212122 ruYauJauYruJrp vvvvv (A.10)
),()()()()( 212
'
12
'
2 ruYauJauYruJrq vvvvv (A.11)
),()()()()( 2
'
12122
'ruYauJauYruJrr vvvvv
(A.12)
).()()()()( 2
'
12
'
12
'
2
'ruYauJauYruJrs vvvvv
(A.13)
where is the azimuthal number, Z0 is the vacuum impedance , cleffnnnn ,,, 321
represent the refractive indices of the core, cladding ,the surrounding medium and
the effective refractive index of the cladding modes in the cladding region
respectively. Jn(.) and Yn(.) are the Bessel functions of the first and the second kind
respectively. The effective indices of the m-th order cladding modes are obtained
by finding the roots of the dispersion relation in above equation.
Appendix
136
Coupling coefficient (κ) between core mode LP01 and cladding modes with
azimuthal value of 1 is given by [79] for step index fibre.
)](1
)1(
)1(*|)([1
*/)1(
*21(
2
110
11
0
111112
1
02
2
1
22
1
1
2
1
2/1
20
011
auJa
bV
bVJ
bVJauJuE
n
abVu
un
bnZ
b
cl
v
cocl
v
(A.14)
where σ is the step function for the profile of grating, 0 is the dispersion relation
for step index dual clad fibre given by
,
)()()()(
)(1
)()()()(1
2
1
2
1
21
2
1
2
1
32
2
1
2
1
21
2
2
2
32
2
2
2
222
21
2
2
322121
2
2
0
aran
uaq
an
uapK
an
uJ
an
uu
asu
aJraKqapaan
uuJKu
(A.15)
Alternatively,
Coupling coefficient k12 (= - k12 ) between LP01 and LP02 modes can be written as
[95]
,)2
exp(12
q
q
qjajkk
(A.16)
where k is given by,
,)(
)(
)(
)(*
)(
))/(1())/(1(
011
010
02
021
020
012
02
2
01
22
02
22
01
2
0
uJ
uJu
uJ
uJu
uun
VuVunkk
(A.17)
where u01 and u02 is given by,
,)4(1
)21(4/1401
V
Vu
(A.18)
Appendix
137
,1
)1
arcsin()/11arcsin(
exp2
2
02
c
c
c
c
u
V
uu
uu
(A.19)
uc is the cutoff value of the normalized frequency for LP02 mode, V (= nRk02 ) is
normalized frequency for the fibre.
Power conversion between the two modes can be written as [91]
,
2/)(
]}2/)([{sin
2
2
21
2/12
2
21
2
0
k
zkk
T
(A.20)
The power in the core and the cladding in the step index fibre are given by
,)1(
)1()1(1
2
1
112
bVJ
bVJbVJaCP ll
core
(A.21)
where C is the constant made of standard integrals associated with Bessel
functions.
The power fraction propagating in the core is given by
,cladcore
core
PP
P
(A.22)
,1)(
)()(2
1
112
bVK
bVKbVKaCP ll
clad
Appendix
138
The detuning parameter δ is given by
,2
2
101
n
cl
(A.23)
The ratio of power of with nth cladding mode and initial power at LP01 mode is
given by expression, as derived in [91]
,
)(1
)(1[sin
)0(
)(
2
22
01
g
g
gn
cl
L
P
LP
(A.24)
where L is the length of the fibre grating, kg is the coupling constant.
Spectral resonance for this case is given by,
,)(
8.0
0
2
clc nnL
(A.25)
Author’s publications
139
Author’s publications
Journals
Navin Prakash Ghimire 1, Hongchun Bao
1 and Min Gu, “Design of zero dispersive
double-clad fibre for high efficiency operation of two color light”, App. Phys. B
108:295–299 ,( 2012)
Navin Prakash Ghimire 1, Hongchun Bao
1 and Min Gu
“Broadband excitation and
collection in fibre-optic nonlinear endomicroscopy”, App. Phy. Let. 103, 073703
(2013)
Conferences
Navin Prakash Ghimire, Hongchun Bao and Min Gu, Nonlinear optical endoscopy
enabled by fibre-based dispersion compensation, 19 Australian Institute of physics
conference, Melbourne, Australia. 5-9 December 2010.
Navin Prakash Ghimire, Hongchun Bao and Min Gu, Nonlinear endoscopy using a
single zero dispersive hollow-core double-clad fibre, Focus on Microscopy-2011,
Konstanz, Germany, 17-20 April 2011.
Hongchun Bao, Navin Prakash Ghimire and Min Gu, Nonlinear optical endoscopy
for in vivo 3D super-resolution imaging, Focus on Microscopy-2011, Konstanz,
Germany, 17-20 April 2011.
Navin Prakash Ghimire, Hongchun Bao and Min Gu, Single fibre based nonlinear
endoscopy, Koala Conference-2011, Melbourne, Australia, 3-5 November 2011.