UNIVERSITY OF BELGRADE
SCHOOL OF ELECTRICAL ENGINEERING
Marijana R. Gavrilovic´
Interplay of cavitation bubble and plasma
emission during single pulse laser induced
breakdown on submerged target
Doctoral Dissertation
Belgrade, 2017.
UNIVERZITET U BEOGRADU
ELEKTROTEHNIQKI FAKULTET
Marijana R. Gavrilovi
Uzajamno dejstvo kavitacionog mehura i
zraqenja plazme kod proboja indukovanog
jednim laserskim impulsom na meti u
teqnosti
Doktorska disertacija
Beograd, 2017.
Committee members
PhD Jovan Cvetic´, thesis advisor
Professor
School of Electrical Engineering, University of Belgrade
PhD Petar Matavulj,
Professor
School of Electrical Engineering, University of Belgrade
PhD Marko Cvejic´
Research Associate Assistant
Institute of Physics, University of Belgrade
Day of the defense:
To my parents...
Firstly, I would like to express my sincere gratitude to my thesis supervisor pro-
fessor Jovan Cvetic´ and my thesis committee professor Petar Matavulj and dr Marko
Cvejic´ for the knowledge having been communicated and help in improving the final
version of this thesis.
Special thanks to dr Sonja Jovic´evic´, who introduced me into the world of the
experimental physics, and underwater plasma in particular and who conceptualised
the experiment presented in this thesis.
My sincere thanks go to my dear colleagues from the laboratory of the Plasma
Spectroscopy and Lasers at the Institute of Physics, University of Belgrade: dr
Milivoje Ivkovic´, Vladimir Simjanovic´, Stanko Milanovic´, Milica Vinic´ and Biljana
Stankov, for the help in my work and inspirational discussions.
It has been a great honour for me to cooperate closely with acad. Nikola Kon-
jevic´, on which I’m very grateful.
Helpful discussions with dr Violeta Lazic and advices obtained during the course
of the investigations presented here, are greatly acknowledged
Thanks to my friends for the appreciation and support. Last, but not least, thanks
to my family for unselfish support.
Marijana R. Gavrilovic´
Belgrade, March 2th, 2017.
elela bih da izrazim iskrenu zahvalnost mentoru doktorske disertacije
profesoru Jovanu Cvetiu i qlanovima Komisije profesoru Petru Matavulju
i dr Marku Cvejiu na prenesenom znanju i pomoi da ova disertacija do-
bije svoj finalni oblik.
Posebno se zahvaljujem dr Sonji Jovievi koja me je uvela u svet eksper-
imentalne fizike, naroqito u svet plazme u vodi i koja je osmislila
eksperiment predstavljen u ovoj disertaciji.
Iskreno se zahvaljujem dragim kolegama iz Laboratorije za Spektroskopiju
Plazme i Fiziku Lasera Instituta za Fiziku, Univerziteta u Beogradu,
dr Milivoju Ivkoviu, Vladimiru Simjanoviu, Stanku Milanoviu, Milici
Vini i Biljani Stankov na pomoi u radu i inspirativnim diskusijama.
Velika mi je qast xto sam imala prilike za blisku saradnju sa akademikom
Nikolom Konjeviem, na qemu sam veoma zahvalna.
Veliku zahvalnost iskazujem dr Violeti Lazi na korisnim diskusi-
jama i savetima u toku trajanja ovog istraivanja.
Zahvaljujem se prijateljima na razumevanju i podrxci. I na kraju, zah-
valjujem se svojoj porodici na nesebiqnoj podrxci
Marijana R. Gavrilovi
Beograd, 02. mart 2017.
Interplay of cavitation bubble and plasma
emission during single pulse laser induced
breakdown on submerged target
Abstract
Although the experimental setup needed for the plasma and cavitation bubble
creation with the laser is relatively simple, enormous variety of physical processes
take place in this experiment, covering the topics from fluid dynamics, to heat
and mass transfer, to chemical reactions, plasma physics and light emission. Laser
induced plasma (LIP) shows a wide range of characteristics depending on the ex-
perimental parameters used for its creation. In addition to the complexity of the
LIP itself, when such a plasma is created inside the liquid environment several extra
factors come into play, making both the experimental and theoretical investigation
highly challenging tasks. In a liquid environment, besides plasma formation due
to the laser action, the shock wave and cavitation bubble are created, all of them
sharing the input laser energy. Due to the interrelation of the phenomena occurring
after laser induced breakdown (LIB) in liquid environment, shock wave, bubble and
plasma emission, in order to get real insight into the physics of the processes, one
need to study more than one effect at the same time.
In this study main focus was on the last two aspects of the LIB, i.e. cavitation
bubble and plasma emission, while the subject of the shock wave emergence and
propagation was also discussed in short. The mutual impact of these two extremely
interesting phenomena, laser produced plasma and the cavitation bubble, is con-
sidered very important for numerous applications, the production of nanoparticles
being one of the more popular recently. Cavitation bubble plays a crucial role in
determining the size, shape and chemistry of the products of laser ablation in liquid.
Although its importance have been recognized and many research groups are deal-
ing with this topic, a lots of speculations about the role of the cavitation bubble in
nanoparticles production still exist. Another point on which the opinions about the
cavitation bubble are divergent regards the question of double pulse laser induced
breakdown spectroscopy (DP-LIBS). Plasma-bubble relations in DP-LIBS are still
not very well documented, while the relation between the plasma and bubble in sin-
gle pulse laser induced breakdown spectroscopy (SP-LIBS) are almost completely
missing.
In this work precisely those intermediate time scales in SP LIBS at which the
bubble starts to appear and the plasma radiation is present, ranging from nanosec-
onds to millisecond, were studied by several experimental methods. Plasma was
studied using the fast photography and optical emission spectroscopy (OES), the
shock wave propagation was recorded with the Schlieren technique and cavitation
bubble evolution was captured using shadowgraphy and probe beam techniques. In
underwater single pulse laser ablation of alumina and aluminium, a long lasting
optical emission of several hundreds of µs duration was detected for the first time.
Duration of emission on the pure aluminium coincided with the first bubble cycle,
while on the alumina further prolonging of optical emission was detected and it was
explained through the material related properties. Based on the unique observations
of the glowing particles inside the cavitation bubble on aluminium and thin glowing
layer in contact with the target on alumina, the temperature of the bubble was es-
timated to be more than 1000 K throughout the whole bubble lifetime. In the early
phases of the bubble evolution it’s temperature was determined from the molecular
AlO bands and it was around 4000 K, which agrees well with the theoretical models.
Besides having extremely long durations, plasma obtained in this study shows an-
other particular feature, namely it is evolving in the two characteristic phases. The
LIBS spectra from the secondary plasma contain the lines from low excited levels
and have a good quality, with very narrow transitions almost free of the continuum
component. Using these findings it is recommended to record plasma emission af-
ter delay of 1–2 µs. Important conclusion is that with the proposed experimental
approach, SP LIBS detection of underwater plasma is feasible both on metallic and
ceramic samples, also by using a not gated and a relatively cheap detector. In the
process of the sustaining and detecting the plasma emission the role of the cavitation
bubble proved to be the of unique importance.
Keywords:
SP LIBS, underwater plasma, cavitation bubble, OES
Scientific field: Electrical and Computer Engineering
Research area: Nanoelectronics and Photonics
UDC number: 621.3
Uzajamno dejstvo kavitacionog mehura i zraqenja
plazme kod proboja indukovanog jednim
laserskim impulsom na meti u teqnosti
Rezime
Eksperimentalna postavka potrebna za stvaranje plazme i mehura u vodi rel-
ativno je jednostavna, ali je raznovrsnost fiziqkih procesa koji se odigravaju
u ovom eksperimentu ogromna, obuhvatajui polja od dinamike fluida, prenosa
toplote i mase, hemijskih reakcija, fizike plazme i emisije svetlosti. Laser-
ski proizvedena plazma LPP moe imati xirok spektar karakteristika u zav-
isnosti od korixenih ekpserimentalnih parametara. Pored sloenosti same
LPP, kada se takva plazma formira unutar teqnosti, nekoliko dodatnih fak-
tora postaje relevantno, xto qini i teorijska i eksperimentalna istraivanja
izazovnim zadatkom. U teqnom okruenju, pored formiranja plazme dolazi i do
formiranja udarnog talasa i kavitacionog mehura od kojih svaki proces troxi
deo energije. Usled meusobne povezanosti ovih pojava, kako bi se dobio pravi
uvid u fiziku procesa, potrebno je prouqavati vixe procesa istovremeno.
U ovom radu glavni akcenat istraivanja je na kavitacionom mehuru i zraqenju
plazme, dok je pitanje pojavljivanja udarnog talasa i njegovog prostiranja disku-
tovano u kratkim crtama. Uzajamno dejstvo ova dva veoma interesantna fenom-
ena, laserski proizvedene plazme i kavitacionog mehura, veoma je vano za
veliki broj primena, od kojih je u poslednje vreme proizvodnja nanoqestica
najpopularnija. Kavitacioni mehur ima presudnu ulogu u odreivanju veli-
qine, oblika i hemije proizvoda laserske ablacije u teqnosti. Iako je vanost
mehura prepoznata i mnoge istraivaqke grupe se bave ovom tematikom i dalje
postoji dosta nejasnoa oko uloge kavitacionog mehura u proizvodnji nanoqes-
tica. Druga tema oko koje takoe postoje neslaganja oko kavitacionog mehura
je spektroskopija laserski indukovanog proboja sa dva laserska impulsa. (DP
LIBS). Odnosi izmeu plazme i mehura u DP LIBS-u jox uvek nisu dovoljno
dokumentovani, dok je takav odnos kod laserskog proboja jednim impulsom (SP
LIBS) apsolutna nepoznanica.
U ovom radu prouqavane su upravo te vremenske skale kod SP LIBS na ko-
jima se mehur pojavljuje, a zraqenje plazme jox uvek postoji, od mikro do mili
sekundi, pomou vixe eksperimentalnih tehnika. Plazma je prouqavana pomou
brze fotografije i optiqke emisione spektroskopije, udarni talas pomou xliren
metode dok je evolucija kavitacionog mehura razluqena uz pomo xadografije
i tehnika sa probnim snopom. U podvodnoj ablaciji alumine i aluminijuma,
prvi put je detektovana dugotrajna optiqka emisija trajanja od nekoliko stotina
mikrosekundi. Trajanje zraqenja na qistom aluminijumu se poklapa sa prvim
ciklusom mehura, dok je kod alumine detektovano dalje produenje emisije koje
je objaxnjeno preko osobina materijala mete. Na osnovu jedinstvenih opaanja
svetleih qestica unutar mehura na aluminijumu i tankog svetleeg sloja u
kontaktu sa metom od alumine, procenjeno je da je temperatura u mehuru vea od
1000K tokom qitavog prvog ciklusa evolucije mehura. U ranim fazama evolu-
cije, temperatura mehura odreena je iz aluminijum oksid molekulskih traka i
iznosi oko 4000K, xto je u skladu sa rezultatima teorijskih modela.
Pored ekstremno dugog ivota plazme jox jedna nova karakteristika plazme
je otkrivena, naime plazma se razvija u dve faze. SP LIBS spektar sekundarne
plazme sadri linije koje potiqu sa pobuenih stanja niskih energija i imaju
dobar kvalitet, jer su detektovane linije gotovo bez kontinuumske komponente.
Koristei ova saznanja preporuqeno je snimanje emisije plazme sa kaxnjenjem
od 1–2 mikrosekunde. Vaan zakljuqak je da je sa predloenim eksperimen-
talnim pristupom, mogua SP LIBS detekcija pod vodom i na metalnim i na
keramiqkim uzorcima, qak i uz korixenje relativno jeftinih detektora bez
opcije vremenskog odabira. U procesu odravanja plazme i detektovanja njenog
zraqenja, uloga kavitacionog mehura se pokazala kao izuzetno bitna.
Kljuqne reqi: SP LIBS, plazma pod vodom, kavitacioni mehur, OES
Nauqna oblast: Elektrotehniqko i raqunarsko inenjerstvo
Ua nauqna oblast: Nanoelektronika i fotonika
UDK broj: 621.3
Contents
List of Figures xiii
1 Introduction 1
1.1 Cavitation bubble . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 General aspects of the laser induced plasma . . . . . . . . . . . . . . 3
1.3 Mutual impact of the cavitation bubble and plasma emission . . . . . 5
2 Laser beam propagation in the liquid environment 7
2.1 Influence of matter on laser light propagation . . . . . . . . . . . . . 7
2.2 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3 Shock wave production and bubble formation 17
3.1 Laser induced shock wave . . . . . . . . . . . . . . . . . . . . . . . . 17
3.1.1 Physics of shock waves . . . . . . . . . . . . . . . . . . . . . . 18
3.1.2 Techniques of studying shock waves in LIB . . . . . . . . . . . 19
3.2 Cavitation bubble . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.2.1 Process of cavitation . . . . . . . . . . . . . . . . . . . . . . . 20
3.2.2 Simple model of spherical bubble dynamics . . . . . . . . . . . 22
3.2.3 Laser induced bubble on a solid target . . . . . . . . . . . . . 26
3.2.4 Interaction of light with bubbles . . . . . . . . . . . . . . . . . 29
4 Plasma formation in liquid ambient 32
4.1 Plasma creation through the LIB . . . . . . . . . . . . . . . . . . . . 33
4.2 Properties of the laser produced plasma in liquid environment . . . . 39
4.2.1 Optical emission spectroscopy of laser-induced plasmas in liq-
uid (LIBS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
xi
CONTENTS
5 Experimental techniques for underwater Laser Induced Breakdown
diagnostics 44
5.1 Fast imaging techniques . . . . . . . . . . . . . . . . . . . . . . . . . 47
5.1.1 Plasma imaging . . . . . . . . . . . . . . . . . . . . . . . . . . 47
5.1.2 Shadowgraphy and Schlieren techniques . . . . . . . . . . . . 47
5.2 Transmission and scattering with the probe beam . . . . . . . . . . . 49
5.3 Optical emission spectroscopy (OES) . . . . . . . . . . . . . . . . . . 50
5.4 Data acquisition and processing . . . . . . . . . . . . . . . . . . . . . 52
6 Results 55
6.1 Plasma evolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
6.2 Bubble dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
6.2.1 Plasma and bubble relations . . . . . . . . . . . . . . . . . . . 62
6.2.2 Optical effects of the laser induced bubble . . . . . . . . . . . 63
6.2.3 Bubble sizing . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
6.2.4 Probe techniques for investigation of the bubble’s dynamic . . 70
6.3 Shock wave propagation . . . . . . . . . . . . . . . . . . . . . . . . . 78
6.4 Optical emission spectroscopy results . . . . . . . . . . . . . . . . . . 78
6.4.1 Primary plasma phase . . . . . . . . . . . . . . . . . . . . . . 79
6.4.2 Secondary plasma phase . . . . . . . . . . . . . . . . . . . . . 86
6.5 Influence of the target material on the laser induced breakdown in
liquid medium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
7 Conclusion 100
Bibliography 105
Biography 129
List of publications 130
xii
List of Figures
1.1 Scheme of the physical processes associated with optical breakdown . 2
1.2 Main stages occurring during the LIB on submerged target . . . . . . 3
2.1 Interaction of light with medium . . . . . . . . . . . . . . . . . . . . . 8
2.2 Reflection and refraction . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.3 Optical properties of the water . . . . . . . . . . . . . . . . . . . . . . 10
2.4 On site trials on the Mediterranean Sea . . . . . . . . . . . . . . . . . 15
2.5 Deep sea in-situ multi-element chemical analysis . . . . . . . . . . . . 16
3.1 Schematic of a spherical bubble in an infinite liquid. . . . . . . . . . . 22
3.2 Influence of the target shape on the bubble evolution . . . . . . . . . 28
3.3 Collapsing bubble and jet formation . . . . . . . . . . . . . . . . . . . 28
3.4 Different scattering regimes depending on the particle size . . . . . . 30
4.1 Dependence of the LIB on submerged target on the focusing conditions 38
4.2 Initiation of plasma by laser and subsequent relaxation and recombi-
nation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
5.1 Optimizing the focusing conditions . . . . . . . . . . . . . . . . . . . 45
5.2 Photograph of the interaction chamber and the target holder . . . . . 46
5.3 Experimental setup for fast imaging techniques . . . . . . . . . . . . 48
5.4 Experimental setup for probe techniques . . . . . . . . . . . . . . . . 49
5.5 Experimental setup for OES with lens . . . . . . . . . . . . . . . . . 50
5.6 Wavelength calibration of the Jarrel-Ash spectrometer with the mounted
iCCD camera . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
5.7 Intensity calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
5.8 Example of the image cropping . . . . . . . . . . . . . . . . . . . . . 53
5.9 Spectral data manipulation . . . . . . . . . . . . . . . . . . . . . . . 53
xiii
LIST OF FIGURES
6.1 Fast photography of plasma evolution in air and water . . . . . . . . 56
6.2 Fast photography of plasma evolution . . . . . . . . . . . . . . . . . . 57
6.3 GUI for plasma integral intensity calculations . . . . . . . . . . . . . 59
6.4 Gain calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
6.5 Temporal dependence of integral plasma intensity . . . . . . . . . . . 60
6.6 Early plasma decay . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
6.7 Position of the maximum intensity plasma region . . . . . . . . . . . 62
6.8 Superimposed images of plasma photography and shadowgraphy-pure
Al. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
6.9 Superimposed images of plasma photography and shadowgraphy-alumina. 64
6.10 Bubble and plasma size relation . . . . . . . . . . . . . . . . . . . . . 65
6.11 Optical effect of the bubble . . . . . . . . . . . . . . . . . . . . . . . 66
6.12 Formation of the glare circle from the N=3 rays . . . . . . . . . . . . 66
6.13 Steps in determining bubble size . . . . . . . . . . . . . . . . . . . . . 68
6.14 Size determination for bubbles with irregular shapes . . . . . . . . . . 69
6.15 Output of the procedure for the bubble sizing . . . . . . . . . . . . . 69
6.16 Probe beam techniques on alumina sample . . . . . . . . . . . . . . . 71
6.17 Transmission techniques on aluminium sample . . . . . . . . . . . . . 72
6.18 Scattering technique on aluminium sample . . . . . . . . . . . . . . . 73
6.19 Transmission with the slit . . . . . . . . . . . . . . . . . . . . . . . . 74
6.20 Transmission with the slit . . . . . . . . . . . . . . . . . . . . . . . . 75
6.21 Calibrated scatter for calculating the bubble size . . . . . . . . . . . . 76
6.22 Images of the shock wave on Al target obtained using the Schlieren
technique . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
6.23 Overall spectra on alumina with fitted continuum contribution . . . . 79
6.24 Illustration of the continuum component for the pure Al target . . . . 80
6.25 GUI for the blackbody subtraction and temperature evaluation based
on the Planck’s law . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
6.26 Absence of line emission in primary plasma phase on aluminium . . . 82
6.27 Hα and oxygen I emission in the first plasma phase on the alumina . 83
6.28 Example of the Hα line fit and determined Ne . . . . . . . . . . . . . 84
6.29 Example of the oxygen I line fit and determined Ne . . . . . . . . . . 87
6.30 Temperature determination from vibrational AlO band . . . . . . . . 87
6.31 Determination of the temperature inside the bubble . . . . . . . . . . 88
6.32 Temporal evolution of vibrational temperature . . . . . . . . . . . . . 89
xiv
LIST OF FIGURES
6.33 Spectra from the secondary plasma phase on the pure Al target . . . 90
6.34 Intensity and SNR evolution of the excites species on the pure Al . . 91
6.35 Spectra in the secondary plasma phase from the impurities in alumina 92
6.36 Long lasting emission of the secondary plasma . . . . . . . . . . . . . 93
6.37 Plasma evolution on different targets . . . . . . . . . . . . . . . . . . 95
6.38 Comparison of LIBS emission in primary plasma phase on two differ-
ent materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
6.39 Images of the shock wave on alumina target obtained using the Schlieren
technique . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
6.40 Bubble evolution on different targets . . . . . . . . . . . . . . . . . . 97
6.41 Bubble size on different materials . . . . . . . . . . . . . . . . . . . . 98
6.42 Comparison of LIBS emission in secondary plasma phase on two dif-
ferent materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
7.1 Graphical comparison of the timescale of LIB processes . . . . . . . . 102
7.2 Sketch of the processes occurring after LIB on the Al target . . . . . 103
xv
Chapter 1
Introduction
While the subject of a laser produced plasma is relatively new area of the research
dating from the invention of the laser in 1960, cavitation bubble has been in the
research focus almost two centuries. This lasting interest in the dynamics of the cav-
itation bubbles follows from the early recognition of the potential damage caused by
the cavitation bubble collapse and destruction of pressure pipes, hydraulic machiner-
ies and turbine structures due to the presence of the bubbles in liquid. Relevance of
both research fields dramatically increased with the widespread use of the lasers in
medicine(Niemz, 2004,2007),(Vogel and Venugopalan, 2003)(Shangguan et al., 1997)
especially in ophthalmology(Vogel, 1998)(Kennedy, 1995)(Kennedy et al., 1995). Al-
though the experimental setup needed for the plasma and cavitation bubble creation
with the laser is rather simple, enormous variety of physical processes take place in
this simple experiment, covering the topics from fluid dynamics, to heat and mass
transfer, to chemical reactions, plasma physics and light emission. Scheme of the
physical processes associated with laser induced breakdown (LIB) on the tissue is
given in the Fig.1.1, where the fraction of the input energy spent on the different
processes is given in the percents, while the typical timescale at which these events
occur on as solid target is given in the Fig. 1.2
Due to the interrelation of the phenomena occurring after LIB in liquid environ-
ment, shock wave, bubble and plasma emission, in order to get real insight into the
physics of the processes, one need to study more than one effect at the same time. In
this study main focus was on the last two aspects of the LIB, i.e. cavitation bubble
and plasma emission, while the subject of the shock wave emergence and propa-
gation was also discussed in short. In the following, brief overview of the research
dealing with one of the phenomena of interest in this study (cavitation bubble or
1
1.1 Cavitation bubble
Figure 1.1: Scheme of the physical processes associated with optical break-
down(Niemz, 2004,2007)
plasma) is given in sect.1.11.2, while in the sect.1.3 the research where both effects
are simultaneously considered is reviewed.
1.1 Cavitation bubble
Interest in the cavitation bubble dynamics arose due to their destructive effects
on the solid surfaces, and it was studied first by lord Rayleigh(Rayleigh, 1917) in
an effort to explain the damage of propellers of high speed boats and submarines.
Acoustic cavitation has been an important area of research as a potential candi-
date responsible for kidney stone comminution and tissue damage in shock-wave
lithotripsy (SWL).(Zhong et al., 2001) High intensity sound can induce cavitation
in the soft tissue and bubbles formed in this way can affect to a great extent the
distribution of the ultrasound energy. It is of crucial importance to understand the
behaviour of these bubbles in order to improve the quality of results of medical
ultrasound in clinics.
To study the cavitation erosion laser produced bubbles are employed frequently
(Isselin et al., 1998)(Vogel et al., 1989)(Takada et al., 2010) due to the ease of
production and different attainable sizes. Severity of the damage produced by the
bubble is determined by its collapse phase in which the micro jet and collapse shock
wave can be formed, and it depends on the parameter γ=d/Rmax, where d is the
distance between initial location of the bubble center and the wall and the Rmax is
2
1.2 General aspects of the laser induced plasma
Figure 1.2: Main stages occurring during the LIB on submerged target,(Lam et al.,
2014)
the maximum bubble radius. General understanding of the bubble beahaviour is
useful for characterisation and identification of underwater explosions using seismic
and/or acoustic signature based on spherical models of explosion bubbles.(Krieger
and Chahine, 2005). Cavitation in the liquid has also been used for control of
nanoparticles synthesis with the ultrasound. (Bang and Suslick, 2010) Studies of
bubbles in liquid are crucial for explaining the striking phenomena of sonolumines-
cence, discovered relatively recently and studied intensively from then on.(Gaitan
and Holt, 1999)(Brenner et al., 2002)(Margulis and Margulis, 2006)(Chu et al., 2009)
Although the majority of application include cavitation in the water, bubble dynam-
ics in other liquids has also been studied. In (Jomni et al., 2009) bubble dynamics
in viscous liquids used for electrical insulators for high voltage applications were
studied. Comparison of bubble formation by laser in water and different types of
alchohol is given in (Kovalchuk et al., 2010) . Encapsulated gas microbubbles found
applications in medical ultrasound as contrast agents for imaging and as a ther-
apeutic tool for drug delivery.(Delale (Ed.), 2013) More applications of the laser
produced cavitation bubble are given in the Chap.2.2
1.2 General aspects of the laser induced plasma
Laser induced plasma (LIP) shows a wide range of characteristics depending on the
experimental parameters used for its creation. Influence of the laser wavelength, en-
ergy, pulse duration(Bai et al., 2013),(Detalle et al., 2003) focusing conditions and
the focal spot size (Ohata et al., 2002), different background gases (Harilal et al.,
2006)(De Giacomo et al., 2012), target materials (Hwang et al., 1991) on plasma
formation and properties are in addition intertwined, increasing the complexity of
experimental studies. Large variety of experimental parameters that can be changed
3
1.2 General aspects of the laser induced plasma
are at the same time the advantage and the drawback of the LIP. On the plus side,
by changing some or all parameters listed above, plasmas of significantly different
conditions can be obtained. As a drawback, it is hard to perform any systematic
comparison between the results of different research groups, and hence to pinpoint
the influence of one specific parameter when, at the same time, large number of
them are affecting plasma’s behaviour.
Due to the large potential for variations of experimental parameters, LIP is not
studied thoroughly and its properties are still interesting topic for many investiga-
tions. Its transient nature with complex propagation of the ablation plume into the
background gas,(Harilal et al., 2002)(Harilal et al., 2003)(Wen et al., 2007a)(Wen
et al., 2007b) where electron number density and temperatures change very fast,
make hard to obtain the same quality of experimental results as in the case of the
arcs or some other well studied plasma source. But, on the other hand, it enables
excitation of almost all elements in periodic system, which are hard to introduce
in conventional plasma sources. Furthermore, by using time resolved studies, range
of electron number density and temperatures that can be obtained in LIP is larger
than with any other source of plasma radiation, which lead to the large popularity
of LIP as a source for Stark parameters measurements. (Cvejic´ et al., 2013)(Cirisan
et al., 2014)(Cvejic´ et al., 2014) Plasma created by short and intense laser radia-
tion focused on a solid, liquid or gaseous sample is the essence of the laser induced
breakdown spectroscopy (LIBS) technique. Plasma emission detected after passing
through some dispersing element, enables determination of elements present in the
material without any sample preparation and with quite simple experimental con-
figuration which is the main advantage of LIBS.
Due to the numerous applications of the LIBS technique , a lot of research is going on
in this area, which resulted in several books on the subject(Cremers and Radziem-
ski, 2006)(Miziolek et al., 2006)(Singh and Thakur, 2007)(Noll, 2012)(Musazzi and
Perini, 2014). Still, as pointed out in several review papers(Konjevic´ et al., 2010)
(Hahn and Omenetto, 2010)(Hahn and Omenetto, 2012)(Gaudiuso et al., 2010)(Fortes
et al., 2013), great deal of research is devoted purely to the applications, while the
research of fundamentals of the technique is lagging behind. This is partially, a con-
sequence of the extreme complexity of the laser induced breakdown (LIB) in media
and extremely short timescales on which the events are occurring after laser action.
4
1.3 Mutual impact of the cavitation bubble and plasma emission
1.3 Mutual impact of the cavitation bubble and
plasma emission
In addition to the complexity of the LIP itself, when such a plasma is created in-
side the liquid environment several extra factors come into play, making both the
experimental and theoretical investigation extremely complicated. In a liquid en-
vironment, besides plasma formation due to the laser action, the shock wave and
cavitation bubble are created, all of them sharing the input laser energy. Formation
of the shock waves happens also in the gaseous surrounding, but these mechanical
effects of the laser action are more pronounced when the interaction is occurring
in the liquid. Although the cavitation bubble was detected in some experiments
performed in the gaseous hydrogen(Kielkopf, 2000), the cavitation bubble is usually
considered to be the unique property of the LIB in the liquid. The mutual im-
pact of these two extremely interesting phenomena, laser produced plasma and the
cavitation bubble, is considered very important for numerous applications, some of
which are listed in the Sect. 2.2. Recently the most emphasized is the production
of nanoparicles by laser ablation in liquid, during which the cavitation bubble plays
a crucial role in determining the size, shape and chemistry of the products. (Yang,
2012)(Amendola and Meneghetti, 2013)DellAglio et al. (2015)(Ibrahimkutty et al.,
2015)(Sasaki and Takada, 2010). Although its importance have been recognized and
many research groups are dealing with this topic, a lots of speculations about the
role of the cavitation bubble in nanoparticles production still exist. Another point
on which the opinions about the cavitation bubble are divergent regards the ques-
tion of double pulse laser induced breakdown spectroscopy (DP-LIBS). DP-LIBS is
applied both in the gaseous and liquid surrounding, but its importance in liquids
is tremendous. Namely, the first laser pulse in liquid or target in liquid, serves to
produce the cavitation bubble as a more suitable gaseous surrounding inside which
the second laser pulse is then sent after certain delay.(Nyga and Neu, 1993)(Michel
and Chave, 2008) (Pu and Cheung, 2003). Depending on the conditions inside the
cavitation bubble, the quality of the spectra obtained after the second laser pulse
differs, and thus it is necessary to optimize the delay between the pulses and the
spatial alignment of the beams, so as to obtain the largest enhancement of the use-
ful signal. (Cristoforetti et al., 2012)(Lazic et al., 2013a)(De Giacomo et al., 2007).
Plasma-bubble relations in DP-LIBS are still not very well documented, while the
relation between the plasma and bubble in single pulse laser induced breakdown
5
1.3 Mutual impact of the cavitation bubble and plasma emission
spectroscopy (SP-LIBS) are almost completely missing. Example to the contrary
is the study of the very early times of bubble appearance in (Tamura et al., 2015),
where the authors compared the plasma-bubble relations for two different pulse du-
rations. Possible explanation for the lack of the studies of interplay of the cavitation
bubble and plasma emission in SP-LIBS is short plasma duration, as usually quoted
in the literature.(De Giacomo et al., 2007)
In this work precisely those intermediate time scales at which the bubble starts to
appear and the plasma radiation is present, ranging from nanoseconds to millisecond,
will be studied by several experimental methods. The structure of the thesis is as
follows. In the Chapter 2 brief overview of the light-liquid interaction is given
and some applications where laser-liquid interact are listed. Chapter 3 gives short
theoretical background on the phenomena of the shock wave and cavitation bubble,
with little accent on the specific case of the laser produced shock wave and bubble.
In the Chapter 4 characteristic of the breakdown in pure liquid and on a submerged
target are closely examined and implication on the optical emission spectroscopy of
such plasmas are explained. In the chapter 5 complete description of all experimental
methods used in this study is given, and in the Chapter 6 the results of different
applied techniques are laid out and compared. At the end, in Chapter 7 the
conclusions are given.
6
Chapter 2
Laser beam propagation in the
liquid environment
Matter can act on light (electromagnetic radiation in general) in manifold ways.
Light propagation through the media can be affected by three main mechanisms:
reflection and refraction, absorption, and scattering. These mechanisms of inter-
action will be described in the following, with special emphasis on the specifics of
the water impact on the light propagation. Besides, they are also relevant during
interactions of light with cavitation bubbles which are described in Sections 3.2.4
and 6.2.2 At higher values of energy density, e.g. when focused laser beam propa-
gates through the medium, the breakdown and plasma formation occurs. Plasma
formation is accompanied by shock wave emergence and cavitation bubble forma-
tion in the case of the water (or any other liquid) as a propagation medium. In
these situations, it can be stated that interaction is going in the opposite direction,
namely light acts on the matter and induces changes inside of it. These phenomena
are described in Sect.3 and Sect. 4
2.1 Influence of matter on laser light propagation
When laser beam propagates through any type of medium, it suffers several types
of losses inside of it. Which of the losses - reflection, absorption, or scattering is
dominant determines the type of medium and the laser wavelength. The sketch of
the ray of light interacting with the slice of matter is given in Fig.2.1
Reflection – Reflection is defined as the returning of electromagnetic radiation
by surfaces upon which it is incident.(Niemz, 2004,2007). According to the simple
7
2.1 Influence of matter on laser light propagation
Figure 2.1: Mechanisms of light interaction with the medium during its propaga-
tion,(Niemz, 2004,2007)
law of reflection, the wave normals of the incident and reflected beams and the
normal of the reflecting surface all lie within the plane of incidence. It also applies
that the reflection angle is equal to the angle of incidence θ = θ′, see Fig.2.2.
Refraction – Refraction is the consequence of the different propagation speeds of
light wave in different media and it occurs on the surfaces which separate two media
of different index of refraction, see Fig. 2.2. The correlation between the angles in
the reflection is given by the Snell’s law
sinθ
sinθ′′
=
v′
v′′
(2.1)
where θ” is the angle of refraction, and v and v’ are the speeds of light in the
media before and after the reflecting surface, respectively. Since the corresponding
indices of refraction can be related to the propagation speed through the relations
n=c/v and n’=c/v’ the Snell’s law 2.1 can be written in different form
nsinθ = n′sinθ′′ (2.2)
Total reflection will occur when sin θ > n’/n.
Absorption – Absorption arises due to the partial conversion of light energy into
heat motion of the molecules of the absorbing material, which results in decreasing
intensity of an incident light beam. If light propagates through the perfectly trans-
parent medium no absorption will occur. Opposite of the transparent are opaque
media, inside which the incident radiation is almost completely absorbed. Of course,
not a single material is completely transparent or completely opaque. In addition,
those properties depend very strongly on the wavelength of the incident radiation.
8
2.1 Influence of matter on laser light propagation
Figure 2.2: Reflection and refraction of the incident beam,(Niemz, 2004,2007)
The ability of a medium to absorb the radiation is usually expressed through the
Beer-Lambert law
I(z) = I0exp(−αz) (2.3)
where z denotes the optical axis, I(z) is the intensity at a distance z, I0 is the
incident intensity and α is the absorption coefficient of the medium.
Scattering – Scattering is occurring when frequency of the electromagnetic radia-
tion impinging on the elastically bound charged particles doesn’t correspond to the
natural frequencies of particles. Resulting forced vibration will have the same fre-
quency and direction as that of the electric force of the incident wave, but with much
smaller amplitude compared to the case of the resonance, i.e. the same frequencies.
In addition, the phase of the forced vibrations is different, which causes the photon
to slow down when penetrating into denser medium. Two types of scattering can
be distinguished, elastic and inelastic, depending if the part of the incident photon
energy is converted during the process. One of the forms of the elastic scattering, in
which the photon doesn’t change its energy, is Rayleigh scattering which is inversely
proportional to the fourth power of wavelength.(Niemz, 2004,2007) In order that
the Rayleigh scattering be valid, the scattering particles must be smaller than the
wavelength of the light. If the size of the scattering particles becomes comparable
with the wavelength Mie scattering occurs. Mies scattering and geometrical optics
approximation is frequently used to describe the scattering of light by bubbles and
it is mentioned briefly in Sect.3.2.4
Looking again the Fig. 2.1, only the photons that have not undergone reflection
9
2.1 Influence of matter on laser light propagation
and absorption, and forward scattered photons can be detected on the other side of
the slice. In the case of LIBS on submerged target this implies that only the portion
of the laser energy actually reaches the target, while large part of the input energy
is spent on the optical path trough the liquid. Wavelength of the input radiation
is extremely important parameter, since it determines the index of refraction and
absorption and scattering coefficients. This is illustrated on the example of the water
as one of the most commonly used liquid in underwater LIBS and almost all laser
material processing applications, see Fig. 2.3. If using the laser with the wavelength
inside the water transmission window, which lies in the blue-green spectral region,
the penetration depth of the laser beam in clear water can be ∼10 m.(Musazzi and
Perini, 2014).
Figure 2.3: Refractive index (top) and absorption coefficient (bottom) of pure water
at room temperature and pressure as a function of wavelength; the vertical lines
indicate the wavelengths of Nd:YAG laser: 1064 nm (red), 532 nm (green),(Lazic and
Jovic´evic´, 2014)
By observing the wavelength dependence of water absorption it can be concluded
that the absorption is the highest in the infrared part of the spectrum, which makes
this wavelength favourable for the plasma formation on the water surface. On the
other side, in order to decrease the losses of laser energy in the water it is neces-
10
2.2 Applications
sary to use the shortest possible optical paths. When choosing the optical paths
and suitable optics one has to bear in mind that focal length of the lenses increases
inside the water due to its higher refractive index (∼1.33) compared to the air ∼1.
In real applications however, the water is not of the highest purity and there are
always suspended particles and gas bubbles present which scatter the light strongly,
making the use of long range in underwater applications impossible. For example,
frequently studied is the sea water which contains sea salts in part per million (ppm)
concentrations. These concentrations for the visible light don’t have an influence
on the absorption or refraction, but increase the molecular scattering significantly
(Lazic and Jovic´evic´, 2014) thereby decreasing the beam transmission and aggra-
vating the plasma formation. This also occurs in natural waters which are rich with
the inorganic particulates and different types of organic materials with sizes in the
range from 0.1 µm to 100 µm (from small colloids to large phytoplanktons). Since
they scatter the incoming light strongly, general recommendation is to work with the
shortest possible optical paths in turbid water. The scattering coefficient decreases
with water pressure and grows with its temperature. In laser surgery it is very
important to know the absorbing and scattering properties of the tissues in order
to perform optimally. Knowledge of the refraction index is usually necessary when
applying the laser radiation on the highly reflective surfaces. The refractive index
determines the overall reflectivity and it is usually wavelength dependent only in the
regions of high absorption. Scattering on the other hand, can scale universally with
the fourth power of wavelength (Niemz, 2004,2007). Important thing to remember
when performing LIBS measurements in the water is to correct the detected spec-
trum for the wavelength dependent absorption of the water before determining the
plasma parameters from it.
2.2 Applications
Application in which the laser beam propagates through the liquid are numerous,
and some of them are described in the following. Several main groups of applications
are elaborated, namely medical, nanoparticles production, material processing, ma-
terial characterisation, environmental ecological and marine applications.
Medical applications
11
2.2 Applications
One of the main causes for huge interest in the area of laser induced breakdown
LIB in liquid is the ophthalmic surgery, which started to develop since 1980s mainly
for cataract surgery or for relieving intraocular pressure as a treatment for glau-
coma, see (Hammer et al., 1997) and references therein. Numerous studies have
been devoted to understand and quantify LIB in ocular media, not only because of
the application but also due to the potential ocular damage.(Kennedy et al., 1997)
Great deal of the research was devoted to determination of the breakdown thresh-
old with nanosecond and picosecond pulses at 1064 nm, motivated by their use in
opthalmology These wavelengths are favoured since they don’t dazzle the patient,
they are well transmitted in the eye, and they posses the minimal risk to the retinal
damage since the absorption coefficient of the retinal pigment epithelium is small.
(Vogel, 1997) In the following years, the transition towards the shorter laser pulses
is continued to the femtosecond regime, in order to minimize the collateral tissue
damage and obtain the same surgical effect with the lower threshold pulse energies
(Heisterkamp et al., 2002)(Nuzzo et al., 2010), but nanosecond and picosecond laser
pulses are still in use in clinical practice.
After the opthalmology, one of the oldest medical applications of the laser is in the
dentistry, but with much less success, and discussions about its usefulness is still
going on, where possible enhancement of the results can be expected with the use
of very short pulse durations(Muoz et al., 2010). One of the most significant disci-
plines for laser application is gynecology, since the CO2 lasers have proven to be very
effective in removing unnecessary cervical tissue. In urology CO2 laser is also very
frequently applied for high precision cutting, while argon ion and Nd:Yag lasers are
used for the coagulation of highly vascularized tumors or malformations. Nd:Yag
lasers are in addition the most common choice in lithotripsy and photo dynamic
therapy.
Besides the studies on the breakdown threshold, the physical after effects produced
by the LIB were intensively studied for all pulse durations, since they largely de-
termine the degree of the tissue damage. These after effects include plasma, shock
wave, bubbles, jets etc.(Vogel et al., 1990) (Vogel et al., 1996)(Docchio and Sacchi,
1988)(Vogel and Venugopalan, 2003) (Niemz, 2004,2007). Studies of the underlying
mechanisms of the bubble and plasma formation during laser action on the tissue
are very useful when determining optimal parameters of the laser for the specific
applications. For example, bubbles in absorbing liquids and on submerged targets
were investigated in the context of controlled drug delivery with microsecond laser
12
2.2 Applications
pulses.(Shangguan et al., 1998) or to estimate the damage on the tissue during the
laser microsurgery in vivo.(Hutson and Ma, 2007)
More extensive list of the different laser applications in medicine can be found in
(Niemz, 2004,2007)
Nanoparticles production
Laser ablation in the liquid as a route for nanoparticles (NPs) production is
experiencing increasing popularity over the last two decades. Number of publica-
tions on the subjects also speaks in the favour of this statement.(Barcikowski et
al., 2009) Production of different types of NPs and use of different experimental
conditions during the production is described extensively in the recent book(Yang,
2012). Also several review articles give excellent overview of the current achieve-
ments of the technique (Zeng et al., 2012)(Sasaki and Takada, 2010)(Venugopal Rao
et al., 2014)(Barcikowski and Compagnini, 2013) It can be seen from the literature
survey that this method of the NPs production is quite versatile, clean or so called
”green” process and particles produced by ablation in a liquid are free of chemical
side products or specific ligands which can compromise the further applications of
NPs.(Salminen, 2013) But, at the same time, there remain a lot of unknowns about
the process, leaving place for improvement of the efficiency and yield, by gaining
insight into the physics of the underwater ablation. In recent years, the lot of atten-
tion was devoted to the use of femtosecond laser for NPs production. Complexity of
the parameters governing the femtosecond laser ablation efficiency for nanoparticles
production is discussed in (Mene´ndez-Manjo´n et al., 2011)(Povarnitsyn et al., 2013).
The importance of the optimal focusing and liquid levels is emphasized for maximum
ablation rate, while several factors which complicate the optimization of the process,
e.g. self-focusing, vaporization of liquid, refraction at the air-liquid boundary and
optical breakdown in the liquid are analysed. The general recommendation of the
authors of (Mene´ndez-Manjo´n et al., 2011) for effective fabrication of colloids via
ultrashort laser ablation in liquid is to use tightly focused beams.
The importance of the bubble as a reservoir of particles has been recognized in sev-
eral independent investigations(Wagener et al., 2013)(Soliman et al., 2010)(Chen et
al., 2017)(Reich et al., 2017)(Ibrahimkutty et al., 2015)(Amendola and Meneghetti,
2013)(Lam et al., 2016) It has also been documented that the target shape and
rigidity can largely determine the yield of NPs through its influence on the bubble
13
2.2 Applications
dynamics. (De Giacomo et al., 2013)(Kohsakowski et al., 2016) In (Al-Mamun et
al., 2013) several physical and chemical factors which influence the size of obtained
NPs are investigated, among which the relative position (inclination) of the target
was concluded to contribute to smaller size of obtained NPs. Many open questions
and large number of independent parameters that can be changed during the NPs
production, which on the other hand have increasing number of applications, make
every research dealing with the fundamentals of underwater ablation relevant and
important.
Material processing
The potential of processing material with lasers have been recognized almost as
soon as the lasers were invented. Optimization of the processes demanded thorough
dedicated studies which resulted in great deal of knowledge, systematised in sev-
eral great books (Gladush and Smurov, 2011)(Kannatey-Asibu, 2009)(Peter Schaaf,
2010). The wide variety of laser material processing in which the interaction of the
laser and material is occurring through the layer of liquid is described in great de-
tail in two sequential publications (Kruusing, 2004a)(Kruusing, 2004b), which have
formed the basis for the subsequently published book (Kruusing, 2008). Systematic
overview of advantages and disadvantages of liquid assisted laser processing can be
found there. Huge interest in under liquid material processing arose from several
advantages that such processing is offering. During processing in liquid, significantly
high pressures and temperatures can be achieved which introduces additional me-
chanical component to the process and increase the amount of the material removed
from the surface of material(Hopp et al., 2000)(Zhu et al., 2001), e.g eight times
for the case of aluminium (Wook Kang— et al., 2008). Besides increasing the ab-
lation efficiency , the liquid serves to cool the surface of the liquid and for efficient
debris removal, leading to the less recast of material in the working zone.(Garcia-
Giron et al., 2016) This has additional beneficial impact since the emission in the
atmosphere is reduced and the process in much safer for the operator due to the
cleaner working area. Material present in the liquid is also useful product, e.g. NPs
discussed in previous section. Purpose with which the liquid is introduced in the
process of laser processing are numerous. Liquid has been used as a photo mask
during backside etching of transparent materials, as a waveguide, for photochemi-
cal processes, removal of particles from the surface, subtractive processing (cutting,
14
2.2 Applications
drilling)(Tsai and Li, 2009) (Krstulovic´ et al., 2013), shock processing (peening, den-
sification)(Mart´ı-Lo´pez et al., 2011)(Peyre et al., 1998). Beside the use of the shock
wave impact, the cavitation bubble is sometimes also used, e.g for laser induced
forward transfer for fabrication of diverse functional devices such as microbatteries,
solar cells, organic light-emitting diodes, or biosensors (Duocastella et al., 2009)
Figure 2.4: Pictures taken during the on-site trials on the Mediterranean Sea: A)
display of the remote LIBS instrument on the research vessel and B) the diver working
at a 30 meter depth.(Fortes et al., 2013)
Material characterisation and applications in environmental, ecologi-
cal and marine surveys
LIBS measurements are often performed on water solutions or inside the same.
Due to the specific properties of the plasmas in liquid, usually only continuum
emission and few very broad lines are detected in the spectrum when applying single
laser pulse, i.e. in SP LIBS. To overcome the limitations of working underwater, a
gas flow over the submerged target was deployed in (Beddows et al., 2002) and also in
on-site trials in the Mediterranean sea (Guirado et al., 2012), see Fig. 2.4, or ablation
of liquid jets was preferred (Rai and Rai, 2008)(Yueh et al., 2002). Archaeological
materials underwater were also the subject of study in (Lazic et al., 2005) (Lazic,
Colao, Fantoni, Spizzicchino and Jovic´evic´, 2007) but there the DP LIBS technique
was used to obtain the usable spectra. This method of the sensitivity increase
(DP LIBS or multiple pulse LIBS) was used by many research groups, both for the
submerged target and for bulk liquid, which resulted in rich and resolved spectra
(Koch et al., 2004)(Michel and Chave, 2008) (De Giacomo et al., 2005)(Pichahchy
et al., 1997) (De Giacomo et al., 2006)
Recently, effort was made to extend the applicability of the LIBS technique
for the stand off applications underwater (Fortes et al., 2015) which is extremely
15
2.2 Applications
Figure 2.5: (A) ChemiCam mounted on the ROV Hyper-Dolphin 3000 about to be
deployed in the Okinawa trough, and (B) a close-up of the linearly actuated focusing
probe held by the ROVs manipulator.(Thornton et al., 2015)
complicated task due to the large energy losses of the laser beam, presence of the
impurities and bubbles, and dispersion of the useful optical signal in the long optical
path. The successful application of the SP LIBS was only documented for long
duration nanosecond pulses (Matsumoto et al., 2013a)(Sakka et al., 2006) which
resulted in impressive ChemiCam apparatus(Thornton et al., 2015), dedicated for
deep sea research, see Fig.2.5. In the construction of the ChemiCam the greatest
attention was devoted to building the long-pulse laser, which have also significantly
increased the cost of the project. The commercial applications of this and similar
devices would be significantly easier and readily available if the standard commercial
laser can be used. One of the main tasks of the present thesis is to discover whether
or not is possible to obtain the spectra of sufficiently good quality for material
characterisation using the SP LIBS technique.
16
Chapter 3
Shock wave production and
bubble formation
High energy density reached in the focal volume during the interaction of the laser
with the material in the liquid environment leads to sudden increase of the tem-
perature and the pressure, which causes powerful expansion of the plasma. Plasma
expansion at supersonic speeds leads to the formation of the shock front surrounding
the plasma. Afterwards, a cavitation bubble is generated, and it oscillates several
times until all of its energy is spent. Basic theory behind the shock wave and bub-
ble formation are given in the following, while the characteristics of laser plasma
produced in liquid are described in the Chap.4
3.1 Laser induced shock wave
Fundamentals on the phenomena of the shock wave and the physics behind it are
given in e.g. (Zel’dovich et al., 1966). Brief survey, based on more specific consid-
erations of shock wave emergence in the cases of laser ablation in liquid where also
the cavitation bubble emerges ((Vogel, 1998)), are given in 3.1.1. Afterwards, the
overview of the experimental techniques used for the shock wave characterisation is
given, with more emphasis on the optical and imaging techniques, some of which
were also employed in this study 6.3.
17
3.1 Laser induced shock wave
3.1.1 Physics of shock waves
Since it is quite hard to calculate the shock profile in the vicinity of the plasma,
simple formulation of an idealized shock front are used instead, as a simpler method
for determination of the pressure at the shock front. Shock front, in its essence,
represents discontinuity in the fluid, where the pressure, density, particle velocity
and internal energy of fluid are experiencing abrupt changes. Still, the conservation
laws for mass, momentum and energy need to be fulfilled for the quantities behind
and in the front of the shock front. For a medium at rest in front of the shock front
(up=0 ) it can be written
usρ0 = (us − up)ρs (3.1)
ps − p0 = usupρ0 (3.2)
and
psup =
1
2
ρ0usup
2 + ρ0us(εs − ε0) (3.3)
Here ρ0 and ρs are the densities and p0 and ps are the pressures in front of and
behind the compression, us is the shock speed, up is the particle velocity behind the
shock front, and (εs−−ε0) is the increase in the specific internal energy of the fluid
as it crosses the shock front.
Combination of the eq. 3.1 and eq. 3.2 eliminates the us and up from the eq.
3.3 and the Rankine-Hugoniot equation is obtained
εs − ε0 = 1
2
(
1
ρ0
− 1
ρs
)(ps + p0) (3.4)
Equations 3.1-3.3 coupled with the material-specific equation,provide complete
description of the relationships between all of the shock front parameters (ps, ρs,
us, up). This makes possible determination of all the parameters by measuring just
one of them. Material specific equation can be either equation of state of medium,
which connects the ps and ρs or the Hugoniot curve for the medium, which describes
the relationship between us and up. If the Hugoniot curve determined by Rice and
Walsh is used as a material-specific equation
up = c1(10
(us−c0)/c2 − 1) (3.5)
18
3.1 Laser induced shock wave
relation between the pressure and the velocity of the shock front is obtained
ps = c1ρ0us(10
(us−c0)/c2 − 1) + pstat (3.6)
where c0 = 1483 m/s is the sound speed and ρ0 = 998 kg/m
3 the density of the
undisturbed water. c1 = 5190 m/s and c2 = 25 306 m/s are empirical constants
determined from Rankine-Hugoniot data for water by Rice and Walsh, and pstat is
the static pressure in the liquid(Delale (Ed.), 2013)(Tagawa et al., 2016)(Vogel et
al., 1996).
3.1.2 Techniques of studying shock waves in LIB
Different experimental techniques used to characterize the shock waves and possi-
bility to obtain various information about the shock wave characteristics in laser
induced breakdown in liquid are discussed in (Vogel, 1998) and (Delale (Ed.), 2013).
Usual path to reveal the propagation of the shock waves is to study the pressure
amplitude ps at the shock front and the shock wave profile as a function of the
distance r from the emission centre of the shock wave. If the ps(r) is determined,
energy transfer from the shock wave to the surrounding can be assessed, which is
especially important in medical applications where the shock wave impacts the tis-
sue. Direct measurements of the pressure in the vicinity of the emission centre are
much harder than the far field measurements, and due to this fact, it is important
to obtain knowledge of the scaling laws for the shock pressure with increasing dis-
tance from the emission centre. This enable one to calculate the pressure in the
vicinity of the plasma based on the far field measurements. Shock pressure in the
far field can be measured very simply using the piezoelectric pressure probes with
sufficient fast response times(Vogel and Lauterborn, 1988). Numerous studies of the
subject have shown that the shock speed decreases very fast in a first few hundreds
of nanoseconds and over a distances of a few hundreds of micrometers(Alloncle et
al., 1994)(Mart´ı-Lo´pez et al., 2011)(T.Tsuji et al., 2007). The largest energy deposi-
tions is actually occurring before the speed of the shock wave decreases to the sound
speed, in the region very near to the emission center and in the very short time,
thus measurements of high temporal and spatial resolution are required to resolve
the properties of the shock wave in this early phase.
Unfortunately, the shock pressure close to the plasma can not be measured with
hydrophones since they can easily be damaged due to the strong pressure transients
19
3.2 Cavitation bubble
or due to the cavitation events. Much better option are optical diagnostic meth-
ods, either photographic or some probe beam technique. The spatial resolution of
these methods is limited only by the resolution of the imaging optics or the size of
the probe beam focus, which can be as little as few µs. Fundamental principle on
which all optical techniques for shock wave detection are relying is the change of
refractive index of a fluid as a consequence of the pressure change(Abraham et al.,
2000). The change of refractive index results in a phase shift or change in direction
of the light that is proportional to the pressure gradient.(Alloncle et al., 1994) The
optical techniques in which the deflection of light trough the shock wave is used
qualitatively for detection of the shock front location can be also applied for mea-
surements of the shock pressure in the vicinity of the plasma. These techniques
actually measure the velocity of the shock front and based on the equations of state
of the fluid, the pressure jump at the shock front is then determined (Alloncle et
al., 1994)(T.Tsuji et al., 2007). The disadvantage of the probe techniques is that
they provide one-dimensional picture of the shock propagation, and any fluctuations
in the breakdown position (which can not be recognized) increase the inaccuracies
in the measurements. This problem can be overcome by employing photographic
techniques, where the shock wave propagation is determined from the series of pho-
tographs, taken at different delays after the laser pulse. This procedure was also
applied in this experiment,5.1.26.3. Further improvement can be obtained with a
streak camera, which make possible to record complete evolution of the shock front
from one laser shot.(Noack et al., 1998)
3.2 Cavitation bubble
3.2.1 Process of cavitation
The process of cavitation represents the emergence of the vapour cavities inside the
liquids, or alternatively it can be defined as a breakdown of a liquid medium under
very low pressures.(Franc and Michel, 2004). The process can occur both in the
static liquid and in the liquid flows. Essential requirement for cavitation is a region
of liquid in which the absolute pressure is less than or equal to the vapour pressure
of the liquid. Very thorough description of the fundamental physics behind the
cavitation process can be found in the (Brennen, 1995). There, the description of
the processes starts from the very pure liquids and clean environments, to separate
20
3.2 Cavitation bubble
the effects of the liquid itself and different impurities and inceptions that can be
found in the industrial and engineering systems. Unclean environments are also
found in the conditions of the laser ablation in the liquid, where a large amount
of the ablated material is present.(Cristoforetti et al., 2012)(Chen et al., 2017).
Usual practice in the engineering applications is to separate the two mechanisms
of bubble creation in liquid: cavitation and boiling. A simple way to distinguish
between these two routes of bubble emergence in the liquid is to define the cavitation
as a process of nucleation when the pressure falls below the vapour pressure, and
boiling as a process of nucleation when the temperature is raised above the saturated
vapour/liquid temperature. But in its essence, there is not much difference between
the two processes and they are mostly reviewed simultaneously.
In the case of the laser created bubble on a submerged target, both the pres-
sure and the temperature conditions are experiencing large and sudden changes(De
Giacomo et al., 2004) and, irrespective of the thermodynamic route leading to the
bubble creation, the bubble will be referred as cavitation bubble as is usual done in
the literature(Petkovsˇek et al., 2007)(Sakka et al., 2014)(Ibrahimkutty et al., 2012).
The problem of the spherical cavity in an infinite medium was solved first for a
non-viscous liquid by (Rayleigh, 1917) to interpret the phenomenon of cavitation ero-
sion, while viscous and non-compressible liquid was considered later on in (Plesset,
1949)(Plesset and Prosperetti, 1977). Although rather simple, Rayleigh’s analysis
proved as quite useful in many practical cases such as bubble collapse, bubble for-
mation from a nucleus, bubble oscillations, and also for laser produced bubbles of
interest here. Very often, this simple model is used in describing the dynamics of the
laser induced bubble (Peel et al., 2011)(Gregorcˇicˇ et al., 2012)(Tomita and Kodama,
2003)(Vogel et al., 1989) or with some additional conditions to account for different
effects(Gregorcˇicˇ et al., 2007), the bubble content(Akhatov et al., 2001) and often
compared with other models(Nath and Khare, 2011). Extensions of the Rayleigh-
Pleset equations3.16 have also been used in explanations of highly interesting effect
of single bubble sonoluminescence.(Brenner et al., 2002)(Margulis and Margulis,
2006)(Ohl et al., 1999). In (Matula et al., 2002), modified Rayleigh–Plesset bubble
dynamics equation of motion with incorporated mass and heat transfer across the
bubble boundary and with chemical reactions included was applied to describe the
behaviour of individual bubbles exposed to shock wave lithotripsy. This approach
was used also for describing the phenomenon of sonoluminescence.
In the following simple model of the dynamic evolution of a spherical bubble in an
21
3.2 Cavitation bubble
infinite domain of liquid at rest far from the bubble and with uniform temperature
far from the bubble will be considered.(Franc and Michel, 2004)(Brennen, 1995)
More thorough explanation can be found in the aforementioned books. Later on,
some modification of the theories of bubble evolution, suited better for the case of
underwater laser ablation will be described3.2.3.
3.2.2 Simple model of spherical bubble dynamics
Starting point of the simple model is a spherical bubble of radius, R(t) in an infi-
nite domain of liquid whose temperature and pressure far from the bubble are T∞
and p∞(t) respectively, see Fig. 3.1. The temperature, T∞, is assumed to be a
simple constant while the pressure, p∞(t), is assumed to be a known input which
regulates the growth or collapse of the bubble. The liquid density ρL and viscosity
µL are considered constant and uniform (incompressible liquid), while the content
of the bubble is considered to be homogeneous, with uniform pressure pB(t) and
temperature TB(t) inside the bubble.
r
p ( r , t)
T ( r , t)
u ( r , t)liquid
vapour/gas
pΒ ( t), ΤΒ (  t)
bubble surface
R ( t )
far from bubble
p∞ ( t), Τ∞  (  t)
Figure 3.1: Schematic of a spherical bubble in an infinite liquid.
The primary goal of the analysis is to obtain the expression for the temporal
dependence of the bubble radius, R(t). Conservation of mass leads to the relation:
u(R, t) =
F (t)
r2
(3.7)
22
3.2 Cavitation bubble
where F(t) is related to R(t) by a kinematic boundary condition at the bubble
surface. When the mass transport across the interface (bubble surface) is zero, radial
outward velocity can also be expressed as u(r,t) = dR/dt, leading to the:
F (t) = R2
dR
dt
(3.8)
Although eq.3.8 is derived under the assumption of the zero mass transport, it
is often fairly good approximation in the cases where evaporation and condensation
occur at the interface. This is supported by the following reasoning. The rate at
which the bubble size grows must be equal to the volume rate production of the
vapour, 4pi R2 dR/dt, hence mass rate of evaporation is equal to ρV (TB)4piR
2dR/dt,
where ρV (TB) is the saturated vapour density at the bubble temperature TB. On
the other hand, this must be balanced by inward mass flow of the liquid relative to
the interface, which yields for the inward velocity of the liquid ρV (TB)(dR/dt)/ρL.
Resulting outward velocity is then given by:
u(R, t) =
dR
dt
− ρV (TB)
ρL
dR
dt
(3.9)
which gives
F (t) = (1− ρV (TB)
ρL
)R2
dR
dt
(3.10)
Since in the most of the practical situations the ρV (TB)  ρL the use of the
eq.3.8 is justified even in the cases when there exist some evaporation and conden-
sation on the surface. Having this in mind, in continuance of derivation of simple
model for spherical bubble dynamics it will be assumed that eq.3.8 is valid.
Motion in the r direction can be written using the Navier-Stokes equation for
the Newtonian liquid as
− 1
ρL
∂p
∂r
=
∂u
∂t
+ u
∂u
∂r
− νL[ 1
r2
∂
∂r
(r2
∂u
∂r
)− 2u
r2
] (3.11)
where ν is kinematic viscosity. After substituting u with eq.3.7 in eq.3.11 it
follows
− 1
ρL
∂p
∂r
=
1
r2
dF
dt
− 2F
2
r5
(3.12)
If the eq.3.12 is integrated and the boundary conditions p → p∞ as r → ∞ it
obtains the following form:
23
3.2 Cavitation bubble
p− p∞
ρL
=
∂1
∂r
dF
dt
− 1
2
F 2
r4
(3.13)
Interesting thing to note that viscous term vanishes from the equation 3.12. The
only viscous term that still needs to be taken into account comes from the dynamic
boundary condition on the bubble surface. To find this boundary condition thin
volume on the interface is observed. The net force on this thin volume in the
outward radial direction per unit area is given as
(σrr)r=R + pB − 2S
R
(3.14)
or, since the stress tensor σrr is σrr =-p+2 µL∂u/ ∂r, the force per unit area can
be written as
pB − (p)r=R − 4µL
R
dR
dt
− 2S
R
(3.15)
If, again, the evaporation and condensation across the interface do not occur the
net force given by eq.3.15 needs to be zero. Susbstitution of the value (p)r=R from
eq.3.13 with the value of F given by eq.3.8 the generalize form of the Rayleigh-Plesset
equation is obtained:
pB(t)− p∞(t)
ρL
= R
d2R
dt2
+
3
2
(
dR
dt
)2 +
4µL
R
dR
dt
+
2S
ρLR
(3.16)
If we neglect the liquid viscosity eq.3.16 is known as a RAYLEIGH equation.
Since in the problem of the bubble collapse bubble behaviour is mostly determined
by inertial forces, liquid viscosity and surface tension can be neglected. Then much
simpler RAYLEIGH equation can be solved to obtain the bubble collapse time. To
be able to do that, it is necessary to estimate also the contents of the bubble. Under
general assumptions it is usually considered that the bubble contains some quantity
of the gas with the partial pressure pG0 at reference size R0 and temperature T∞.
pB(t) = pV (TB) + pG0(
TB
T∞
)(
R0
R
)3 (3.17)
This relation for the bubble pressure eq.3.17 is substituted in the 3.16, which
gives Rayleigh-Plesset equation in the following general form:
pV (T∞)− p∞(t)
ρL
+
pV (TB)− pV (T∞)
ρL
+
pG0
ρL
(
TB
T∞
)(
R0
R
)3 = R
d2R
dt2
+
3
2
(
dR
dt
)2+
4µL
R
dR
dt
+
2S
ρLR
(3.18)
24
3.2 Cavitation bubble
The first term in eq.3.18 is the instantaneous tension or driving term determined
by the conditions far from the bubble. The degree to which the bubble temperature,
TB, departs from the remote liquid temperature, T∞, can have a major effect on the
bubble dynamics. To determined this departure it is necessary to solve heat diffusion
equation, and there is no exact solution. But, in the cases when the thermal effects
need to be taken into account, the approximate solution of Plesset and Zwick is
often good enough. This is beyond the simple model presented here, and in the
further discussion, the thermal term (term 2 in the eq.3.18) will be neglected. This
is so called ”inertially controlled” bubble behaviour. In addition, the behaviour of
the gas in the bubble will be assumed polytropic so that
pG = pG0(
R0
R
)3k (3.19)
where k is approximately constant. With k=1 the bubble temperature would be
constant, while k=γ would correspond to adiabatic expansion, where γ is the ratio
of specific heats. Introducing these approximations, the Rayleigh-Plesset equation
becomes:
pV (T∞)− p∞(t)
ρL
+
pG0
ρL
(
R0
R
)3k = R
d2R
dt2
+
3
2
(
dR
dt
)2 +
4µL
R
dR
dt
+
2S
ρLR
(3.20)
In order to perform the integration, the viscous term, tension term and noncon-
densable gas will be neglected. With these simplifications, and having in mind that
RR¨ + 3
2
R˙2= 1
2R˙R2
d
dt
[R˙2R3] the integration of eq. 3.20 gives
ρLR˙
2R3 = −2
3
(p∞ − pV )(R3 −R03) (3.21)
Since r˙ is negative during the collapse, it follows:
dR
dt
= −
√
2
3
p∞ − pV
ρL
[
R0
3
R3
− 1] (3.22)
The numerical integration of equation 3.22 gives the characteristic collapse time
or Rayleigh time:
τ =
√
3
2
ρL
p∞ − pV
∫
0
R0 dR√
R0
3
R3
− 1
∼= 0.915R0
√
ρL
p∞ − pV (3.23)
25
3.2 Cavitation bubble
The constant 0.915 is the approximate value of
√
pi
6
Γ(5/6)
Γ(4/3)
where Γ is the factorial
gamma function.
The equation 3.23 is very often employed in laser induced breakdown on a sub-
merged target studies to estimate the maximum bubble radius from the measured
time of the bubble collapse from optical or acoustic measurements(Lazic, Jovievic,
Fantoni and Colao, 2007)(Huber et al., 1998)(Vogel et al., 1989). Eq.3.23 will also
be employed in this study in order to assess the amount of the input laser energy
stored in the bubble, see chap.6.2.3.
3.2.3 Laser induced bubble on a solid target
The phenomenon of laser-induced cavitation over a target surface differs from the
ideal case of a spherical bubble produced in bulk water since the ablation products
influence the evolution of the bubble, which can ,in addition, interact with the target
surface via heat or mechanical energy transfer. Although dynamics of the cavitation
bubble formed after laser ablation in liquid is more complex than this simple model
described in 3.2.2, as pointed out in (Cristoforetti et al., 2012), it is often reproduced
satisfactorily with it and, in the absence of dedicated experimental and theoretical
investigations about the dynamic of the bubble over target surface, it is often ap-
plied in the literature. This might appear incorrect, since the bubble produced by
laser ablation in liquid is of approximately hemispherical shape and simple model
in3.2.2 was derived for spherical bubble in infinite domain of liquid. But, due to
the symmetry of the problem, if the surface tension effects at the rigid boundary
are neglected, the hemispherical bubble should collapse in the similar manner as in
the spherical case away from all boundaries. To enhance the agreement between
theoretical predictions and measured bubble parameters, usually several additional
effects are included in the models. In (Casavola et al., 2005) the Rayleigh-Plesset
equation is generalized to account for the liquid compressibility and Keller-Miksis
formulation is used instead(Brenner et al., 2002). For the estimation of the condi-
tions inside the bubble, an adiabatic Van der Wals equation of state is assumed and
solved together with the Keller-Miksis equation. This model of bubble dynamics
had a goal to determine the optimal timing between the pulses in DP-LIBS, since
the characteristics of the spectra produced by the second laser pulse are largely
dependent on the temperature and pressure conditions inside the bubble. The op-
timal timing between the successive laser pulses in DP-LIBS has been the subject
26
3.2 Cavitation bubble
of discussion in preceding years, where several research groups came to slightly dif-
ferent conclusions(Sakka et al., 2012)(Cristoforetti et al., 2012)(De Giacomo et al.,
2007)(Lazic et al., 2013a)(Lazic et al., 2013b)
When dealing with the cavitation bubble formed over a target surface it is supposed
that, due to the inherent asymmetry of the bubble collapsing on the surface of the
target, some deviation from experimental observation during the bubble collapse
phase will occur. Indeed, when the bubble is created near the boundary or on the
surface, the shrinking phase of the bubble is difficult to model due to the breaking
of some assumptions used to derive the Rayleigh-Plesset equation and their im-
proved versions. Importance of the bubble collapse phase is recently emphasized in
the context of the nanoparticles production in the liquid(Ibrahimkutty et al., 2015)
(Ibrahimkutty et al., 2012) The question of whether the material is trapped inside
the bubble and if it is released during the collapse are still not completely understood,
although recent experimental results suggest that the bubble confines material inside
and also partially redeposit material back to the target surface(Ibrahimkutty et al.,
2015). The influence of different target shapes on the successive bubble evolution
and hence the nanoparticle yield was investigated in (De Giacomo et al., 2013). The
authors applied multiple experimental techniques, Optical Emission Spectroscopy
for the plasma characterization, fast shadowgraph for plasma and cavitation bubble
dynamics and laser scattering for the mechanisms of delivery of the produced ma-
terials in the liquid. They discovered unusual arrow–bow ejection phenomena when
the cavitation bubble was formed on the wire shaped target, which they connected
with the suppressed material back-deposition and more efficient ejection of ablated
matter into the liquid, see fig. 3.2
On the other side, according to the recent investigations of the bubble dynamics
in the frame of the nanoparticles production through the laser ablation in the liquid,
system evolution is found to be mostly inertial and adiabatic(Lam, 2015). In addi-
tion, the authors also dealt with the ”hot-topic” regarding the content of the bubble,
arriving to the conclusion that the inside of the bubble is mostly occupied by sol-
vent molecules whose number is not changing dramatically. These several examples
illustrate the high interest and still large number or unresolved issues regarding the
relation of the cavitation bubble and ablated material, which will be to some extent
addressed here.6.10 although the primary interest of this investigation are relations
between the bubble and the plasma.
Collapse of the cavitation bubble and formation of jets or additional shock waves
27
3.2 Cavitation bubble
Figure 3.2: Influence of the target shape on the bubble evolution: comparison be-
tween laser ablation of the wire and bulk target(De Giacomo et al., 2013)
upon the bubble collapse is another interesting subject that has been investigated
over the years(Ellis, 1953)(Benjamin and Ellis, 1966)(Vogel et al., 1989)(Ohl et al.,
1999) and example of the emergence of the jet is given in the fig.3.3.
Figure 3.3: Example of the jet formation during the collapse of the laser induced
bubble near solid boundary(Vogel and Venugopalan, 2003)
These research are mostly relevant for medical application of laser ablation (Bru-
jan et al., 2001) and laser-material processing where the impact of the cavitation
bubble is very often undesired side-effect. Today’s held view is that the most of
the cavitation damage is induced primarily due to the asymmetrical bubble col-
28
3.2 Cavitation bubble
lapses, where distinctive jet and counter propagating jet are formed (Takada et al.,
2010). The strong influence of the bubble collapse will also be documented in this
experiment, through the second shock wave and ,in some cases, intensity emission
enhancement detected in the period of the bubble collapse.6.5
3.2.4 Interaction of light with bubbles
Other than its great influence on the material and plasma trapped inside, cavitation
bubble impacts quite significantly rays that are encountering the bubble surface.
(Langley and Marston, 1991) This is of big importance for optical diagnostics tech-
niques, since the information obtained might be modulated by the presence of the
bubble in liquid, which has significantly different refractive index (∼1) from the
surrounding liquid(∼1.33 for distilled water). The bubble inside the liquid acts as a
negative lens, whose focal length measured in (Lazic et al., 2012) at the moment of
the maximal expansion is only f=-11 mm at the wavelength of 1064 nm. Defocusing
and refraction of the light by vapour cavity increases at shorter wavelengths. This
defocusing power of the bubble needs to be taken into account when performing DP
LIBS experiment, since it enlarges the ablation craters and decreases the ablation
efficiency.(Lazic and Jovic´evic´, 2014). Apart form defocusing effect on the incoming
laser beam, the optical effects of the bubble also lead to the spatial redistribution of
the plasma radiation created by the second laser pulse. This can lead to very large
signal losses, since more than 50% of the plasma radiation can escape the detector
depending on the optical collecting system employed. In addition, the bright central
spot in the proximity of the bubble center is formed due to the multiple reflections
from the bubble wall, which can partially recover the signal intensity Optical effects
of the bubble, relevant for the DP LIBS, will be analysed here 6.2.2 also in the view
of the SP LIBS for the first time, since the duration of the plasma emission in this
experiment is of the same order of magnitude as the bubble duration, which enables
to study their mutual influences.
Interaction of the spherical bubble and light has been studied intensively due to
numerous applications in which the measurements of the radius and lifetime of the
bubble were of interest. In the beginning they were mostly dealing with the sta-
tionary bubbles in water and refined over time to suit better the measurement of
nonlinear pulsation of bubbles, as those encountered in the sonoluminescence(Barber
and Putterman, 1992)(Crum et al., 1999). When the diameter of the scattering ob-
29
3.2 Cavitation bubble
ject is small compared to the wavelength of the probing laser beam, the dominant
contribution to the scattering signal is Rayleigh scattering and in the 90◦ direction
to the probe beam. As the diameter of the scattering object increases scattering law
starts to change to Mie’s scattering, and the preferential direction of the scattering
is in the forward direction.(Huber et al., 1998). Since the laser produced bubbles are
known to have a diameter in the range of 0.01 to few mm (Kennedy et al., 1997))
the scattering from them can be described by Mie’s theory,see Fig.3.4.
Figure 3.4: Size parameter x as a function of wavelength λ of the incident radiation
and particle radius r.(Wallace and Hobbs, 2006)
If the size, shape and composition of the scatterer are assumed, classical electro-
dynamics can be used to determined the size of the bubble based on the scattered
signal intensity.(Yu et al., 2008). An exact solution for the scattering of electromag-
netic waves from a spherical dielectric body was first obtained by Mie(Mie, 1908),
and hence the term Mie’s scattering. In (Lentz et al., 1995) laser-scattering tech-
nique has been developed to size sonoluminescing bubbles in water by the use of
Mie scattering lobe clusters. Comparison of the measured and calculated intensities
of the scattered light according to the Mie’s scattering theory is given in (Holt and
Crum, 1990). Scattering of electromagnetic radiation from a sphere, so-called Mie
scattering, requires calculations that can become lengthy, hence great deal of work
was devoted to improvement and optimization of the Mie scattering algorithms ,see
30
3.2 Cavitation bubble
e.g.(Wiscombe, 1980)(Ma¨tzler, 2002). If the diameter of the scatterer is further in-
creased the process of the scattering can be described with much simpler geometrical
optics approximation, see Fig.3.4. This is especially convenient since the numerical
calculations of the Mie coefficients becomes time consuming and difficult as the size
of the scatterer is increased. For a large particles, geometrical optics approximation
in which the the scattered light intensity is obtained as the superposition of the
reflected, refracted and diffracted rays, is considered advantageous since it avoids
the difficulties with Bessel functions of high orders in the Mie coefficient calcula-
tions. General characteristics of the geometrical light scattering by spheres is given
in (McK. Ellison and Peetz, 1959). Theoretical investigation of the scattering from
the large air bubbles in water (Yu et al., 2008) in which the total reflection was taken
into account proved that geometrical optics approximation approach agree well with
the results obtained from Mie theory for those particles with relative refractive index
m<1 and dimensionless particle size parameter α >50, where m=m2/m1 is the rela-
tive refractive index, viz. the ratio of the refractive index of the particle m2 to that
of the medium m1, and α=2pia/λ is the dimensionless particle size parameter,where
a is the radius of spherical particle and λ is the wavelength of the incident light in
the medium. Comparison of the results of the Mie theory and geometrical optics
approximations yielded good agreement for the case of the spheroidal bubble in (He
et al., 2012) and for critical angle scattering (Dean and Marston, 1991).
Amplitude of the measured scattered signal is largely dependent on the angle be-
tween the laser and the detector. In the geometrical optics approximation, intensity
of the light scattered by the spherical bubble or radius R is proportional to R2 for all
scattering angles(Barber and Putterman, 1992)(Matula et al., 2002)(Lazic, Jovievic,
Fantoni and Colao, 2007). Thus, if the intensity of the scattered probe beam on
the bubble is measured on the opposite side with suitable detector (photomultiplier
or photo diode), the bubble radius will be proportional to the square root of its
voltage. As radius varies, the exact Mie’s result for the light scattered into a fixed
angle also follows approximately the square dependence but with many additional
fringes. This problem can be easily avoided in the measurement by using the short
focal length lenses with large aperture which averages the result across the large
angle and thus the fringes disappear It was found in (Barber and Putterman, 1992)
that the deviation from the R2 dependence is less than 10 % for bubbles with radius
larger than 1 µm. This property was used for the scattering measurements described
in 5.2 and the results are given in the Sect.6.2.4
31
Chapter 4
Plasma formation in liquid
ambient
The fourth state of the matter called plasma, can be created in a number of ways, one
of which is the laser induced breakdown (LIB). During the LIB, plasma emerges as
a consequence of the complete or partial ionization of solid, liquid or gaseous sample
following the dielectric breakdown occurred due to the absorption of electromagnetic
energy. Plasma represents the gas mixture which contains the neutral particles
(atoms and molecules), positive ions, and free electrons able to interact among
themselves and with photons emerging from different energy levels and excitation
states. In gaseous surrounding the plasma formation is usually initiated through the
multiphoton absorption or cascade ionization of the ’free electrons’ which are always
present in the gases. In contrast, inside solids and liquids electrons are either bound
to the particular molecule or lattice site or are ’quasi-free’, i.e. they posses enough
kinetic energy that they can move through the liquid or solid lattice without being
trapped by localized potential wells.(Kennedy et al., 1997). Having that difference in
mind, transition between the bound and quasi-free state during the breakdown in the
condensed matter corresponds to the molecular ionization in gases. LIB mechanisms
in liquid media are far less studied compared to the LIB in gaseous and solid media.
Besides few exceptions, e.g. (Barnes, 1969), studies of LIB in liquid have been
rarely mentioned before 1980’s, when wide variety of laser applications in medicine
draw attention to this topic. (Docchio et al., 1988)(Docchio et al., 1988) Since the
great deal of medical (and other) applications as a breakdown medium use water,
correspondingly more research was devoted to the studies of LIB in water. But liquid
water represents extremely complicated medium with complex physical structure,
32
4.1 Plasma creation through the LIB
molecular bonding and band structure, due to which many questions about complex
sequence of interrelated phenomena produced by LIB in water remained open. In
the following, basic mechanisms of the processes responsible for the breakdown in
the liquid media and on submerged target are explained, after which the properties
of plasmas created in the process are outlined with an accent on the emission spectra
of interest for LIBS.
4.1 Plasma creation through the LIB
1. LIB in the bulk of liquid
Breakdown in aqueous media is defined experimentally through detectable ef-
fects of laser radiation in media, like plasma emission, bubbles and shock waves.
Theoretical modeling of LIB thresholds requires defining a minimum free electron
density which corresponds to a breakdown, usually referred to as the critical density.
Laser radiation in general can lead to a breakdown in two different ways: thermal
breakdown or optical breakdown. The former case is mainly connected to the long
duration pulses or high power repetitive pulses applied on materials with high lin-
ear absorption coefficients whilst optical breakdown involves use of short duration
pulses, from micro to femto second regimes. Optical breakdown is of greater im-
portance for LIBS underwater since it is the primary mechanism of laser-induced
breakdown when dealing with transparent medium. Main principles of plasma for-
mation in the case of interaction of short pulse lasers and water rely on electron
cascade formation or direct ionization by multiphoton absorption (Kennedy et al.,
1997). Both mechanisms of plasma formation can occur also in solids and gases. The
threshold irradiance required for LIB is a function of both the medium character-
istics (ionization energy, impurity level) and the beam characteristics (wavelength,
pulse width, beam diameter at focus). Threshold measurements are the first step in
any experimental characterization of LIB. Similarly, calculation of the breakdown
threshold is a key goal of any theoretical model. This is not an easy task, since
liquids are very complex media, (liquid water in particular), therefore main body of
theoretical calculations of LIB threshold values are based on fitting of experimen-
tal data with different simple formalisms, such as lucky electron model (Shockley,
1961). This model was the first evidence of cascade ionization in liquids as a primary
mechanism for LIB for nanosecond to picosecond pulses, similarly like in solids for
33
4.1 Plasma creation through the LIB
the same pulse width range. However, this model can not account for multiphoton
ionization nor to include impact of present impurities, and for that reason it can
not be used for the calculations of the threshold values in a general case. Except
lucky electron model, common alternative approach to cascade breakdown model-
ing is rate equation model. Using the rate equation formalism and Drude model
(uniform scattering rate for all free electrons) cascade ionization threshold can be
easily calculated. For multiphoton ionization in liquids, most frequently employed is
Keldysh model, originally developed for semiconductors, but applicable in the case of
liquids if considered as amorphous semiconductors.(Williams et al., 1976) Kennedy
provided first order model for estimation of LIB threshold (Kennedy, 1995) that
accounts both multiphoton and cascade ionization regimes. In relation to femto
and pico second pulses significant interest was devoted to self-focusing effect on LIB
threshold. It was found by comparing experiments, first order model and numeri-
cal simulation that this effect plays important role in the case of short laser pulse
regimes (pico and femto). (Feng et al., 1997)
Cascade ionisation – To start the process of cascade ionisation (also called avalanche
ionisation) seed electrons must be produced (since they are not present in liquids
like in the case of gases) and then by multiplication, high number of free electrons
(critical density) is achieved. Cascade ionization is initiated when primary elec-
tron absorbs laser photons during collisions with heavy particles, process named as
inverse bremsstrahlung absorption. Role of the heavy particles is to preserve the
momentum and energy. After photon absorption free electron has enough energy
to ionize bound electron by collision, leading to the creation of two free electrons
with lower energies. Repetition of this process leads to the multiplication of the
free electrons or electron cascade (breakdown). Nevertheless, in order to have an
electron cascade, rate of energy loses has to be smaller than rate of energy produced
in focal volume. Energy loses are mainly due to inelastic collisions. Considering
the fact that energy gain is proportional to irradiance but energy losses are not, it
follows that threshold irradiance value exist for which breakdown process will occur.
Multiphoton ionisation – On the other hand, multiphoton ionisation is in its
basics much simpler process. It does not require an initiation and it is much faster,
therefore the lost mechanisms can be ignored. Every electron in the process of
multiphoton ionization is ionized by absorption of multiple photons from the field,
therefore particle-particle interactions and collisions are not needed. In this way,
process of multiphoton ionization doesn’t depend on the presence of impurities and
34
4.1 Plasma creation through the LIB
it can take place in very diffuse media in which multiple collisions can not occur.
In respect to the cascade ionization, it is much faster process and it can happens
even for the very short laser pulse durations. On the other hand, energies required
for multiphoton ionization are much higher than energies for cascade ionization,
since multiphoton ionization is nonlinear process for which very high irradiances
are needed. Because of that, more common way of plasma formation is cascade
avalanche ionization, except in the case of short laser pulses (femtosecond) where
the field in the focal volume doesn’t exist long enough to allow formation of the criti-
cal density of free electrons. Also in the case of diffuse gases, multiphoton ionization
is favourable process in respect to avalanche ionization since collision rates are very
low to establish significant number of free electrons through collision processes. Ex-
cept as an independent method of breakdown in liquids, multiphoton ionization is
often considered as initiation mechanism for avalanche ionization, especially for me-
dia with little amount of impurities or very diffuse media, where no other way for
starting the cascade process exists.
Effect of impurities – Presence of impurities significantly lowers breakdown thresh-
old in liquids and makes it probabilistic process with defined probability of a break-
down. In the case of water with impurities, initial seed electrons for cascade ion-
ization mainly come from the thermal excitation of those impurities, which produce
initial free electron density in the focal volume. For pure water, initial free electrons
must be produced through the process of multiphoton absorption, but for that to
happen much higher irradiances are needed. In the later case, there will actually
be two thresholds associated with the cascade breakdown. (Kennedy, 1995)(Shen,
1984) The first threshold is related to the irradiance needed to initiate a cascade
by producing minimal critical density through multiphoton ionization. The other
threshold is the irradiance needed to sustain a cascade to breakdown with initial
critical density of free electrons. Since cascade breakdown threshold is usually much
lower than multiphoton ionization threshold, measured threshold values actually
correspond to either multiphoton ionisation or impurity-initiated cascade break-
down. In pure water where two threshold values need to be surpassed, measured
threshold value should be the higher, in this case multiphoton breakdown thresh-
old. On the contrary, when liquid contains enough amounts of the impurities that
multiphoton initiation of cascade process is not needed, measured threshold value
corresponds to the cascade breakdown threshold. With the shortening of the pulse
duration (other parameters remain constant) relation between these two threshold is
35
4.1 Plasma creation through the LIB
changing: both threshold increase as pulse duration decrease, but cascade threshold
increase at much bigger rate. Based on that fact, influence of impurity presence in
liquid should gradually decrease when moving to the short pulse regime, which is
experimentally confirmed (Docchio et al., 1986). Importance of the impurities for
the LIB in water and alcohol is recently emphasized (Toker et al., 2009)(Kovalchuk
et al., 2010). It was proven that mechanism of the breakdown is initiated by in-
clusion particles and that laser spark column has discrete structure. Instead of one
plasma, the authors explain the process through the formation of numerous plasma
events that consist of micro-plasma balls, micro-bubbles and spherical shock waves.
Because plasma is formed around inclusion particles, initial characteristics and size
of micro-plasma balls will depend on the nature and concentration of suspended
particles.
Effect of focusing conditions – Dependence of LIB threshold on various con-
ditions is rather complex, since numerous parameters are combined and mutually
dependent. Further complication is introduced since the two main mechanisms for
breakdown in liquids (avalanche and multiphoton ionisation) also tend to intersect
in some regimes. Wavelength dependence of the two breakdown mechanism is the
opposite: longer wavelengths favour the breakdown through the cascade ionisation,
and in short wavelength range primary mechanism for LIB is multiphoton ionisa-
tion since inverse bremsstrahlung absorption needed for the cascade ionization has
low efficiency. One of the laser beam parameter with biggest influence is focusing
and spot size in medium. Spot size can influence cascade breakdown threshold in
two ways: through free electron losses due to diffusion during the buildup process,
or variation in minimal density required for initiation of avalanche. For pulses of
longer duration, smaller spot size will mean higher threshold, because one has to
include the influence of diffusion, primarily in the cases of microsecond regimes or
longer. When working with the smaller beam diameters, initial minimal densities
required to initiate a breakdown are higher, in order to ensure enough free elec-
trons in the focal volume. This means that less number of cascade multiplications
is needed to reach the breakdown threshold, but still, the spot size dependence is
weak. Muiltiphoton initiated breakdown threshold should also experience small in-
fluence of spot size but in the opposite direction: smaller spot size, higher threshold.
This means that for the pure water, where threshold for breakdown is determined
by multiphoton ionisation, decreasing the spot size will lead to higher thresholds
for breakdown. For very impure media there should not exist dependence on spot
36
4.1 Plasma creation through the LIB
size. When level of impurity is low enough or spot size small enough multiphoton
initiation of breakdown process is required, just like in the case of the pure water. In
addition, the optimal focusing is extremely important whenever the stable position
of the LIB and plasma is of importance, as is the case for LIBS technique. Reasons
for fluctuating position of LIB and multiple plasma formation as those observed in
(Toker et al., 2009) can be multiple breakdown on impurities, probabilistic nature
of the breakdown and finding seed electrons in the focal volume, interaction of the
later part of the laser pulse with the formed plasma causing forward movement of
the plasma towards the laser, ’hot-spots’ in the focal region and self-focusing, etc.
Recent papers (Tian et al., 2016)(Tian et al., 2015) report well localized, stable
plasma position with one core obtained with tightly focused laser beam and large
focusing angles.
2. LIB on a submerged target
The process of LIB in the presence of the solid target in the liquid is somewhat
different then previously described LIB in the bulk liquid. The LIB threshold is
much lower than for the bulk liquids, since the initial free electrons are more eas-
ily obtained from the solid target. During LIB on the submerged target, part of
the input laser radiation transmitted through the liquid column reaches the sample
and heats it locally up to the boiling or decomposition temperatures in order to
achieve efficient evaporation. (Noll, 2012) The process is largely dependent on the
sample material in the question, since the amount of the temperature rise will be
influenced by the efficiency of the laser-material coupling, which on the other side,
depends on the thermal conductivity of the material. If the laser energy surpasses
the characteristic threshold which depends on the sample material, the electrons
are cut free from the atomic nuclei by multiphoton ionisation. Once the initial free
electrons are created, they can continue to absorb the laser photons through the
process of inverse Bremsstrahlung (for ns pulses). Through this process, electrons
increase their kinetic energy, which leads to the more efficient ionisation through
the impact collisions. As a result, dense plasma is formed together with two strong
shock waves, one on the liquid side and laser–induced stress wave (LSW) on the
solid side.
Due to the interaction of the pulsed laser beam with the target, ablation and expul-
sion of the material occurs. This ablated mass is partially evaporated and ionized,
forming the plasma state. The remaining part of the ablated material is removed
37
4.1 Plasma creation through the LIB
from the target in different forms (vapour, particles, liquid droplets) and later on it
can form particulates in the interaction region. If the concentration of this partic-
ulates is too high, great deal of the laser energy will be spent before reaching the
target. In addition, particles can also act as a preferential breakdown sites which
provide initial seed electrons, and produce parasitic breakdown in the liquid before
the target. For that reason, it is necessary to frequently exchange the liquid or to
use flowing systems when performing the LIBS measurements on the submerged
targets. (De Giacomo et al., 2007)(Musazzi and Perini, 2014) Comparison of the
laser ablation process in the pure liquid and colloid is reported recently (Chen et
al., 2017), concluding that much less energy was reaching the target in the case of
the colloid, resulting in the smaller size of the nanoparticles. Besides keeping the
impurity concentration low, for laser ablation inside liquid optimal focusing is very
critical. Optimal focusing conditions with respect to the stable and reproducible
LIB event on the target surface, are achieved when the target is placed above the
nominal focal point of the focusing element. Images obtained with photoelastic-
ity imaging technique developed by (Nguyen et al., 2013) to observe the LSW in
the solid target, shows that placing the target at the exact focal position results in
elongated plasma of very poor reproducibility, see Fig. 4.1
Figure 4.1: Dependence of the LIB on submerged target on the focusing condi-
tions,(Nguyen et al., 2013)
This position also reduces the advantages of the laser processing underwater,
since the intensities of the generated LSWs are much weaker, based on the number
of fringes detected in the target. This is the consequence of the columnar plasma
shape, whose front efficiently absorbs the remaining part of the laser pulse energy
38
4.2 Properties of the laser produced plasma in liquid environment
and increases its volume, thus decreasing the confining effect of the liquid.
4.2 Properties of the laser produced plasma in
liquid environment
Laser produced plasma is transient system that undergoes fast changes during its
evolution. When LIP in the liquid is created with the nanosecond pulse duration
laser, initially formed vapour plume can gain energy form the remaining part of the
laser pulse. This produces heating of the electrons through the process of the inverse
Bremsstrahlung absorption and photoionisation of the heavy species. Absorption of
the laser radiation depends on the laser wavelength and it can be calculated from the
Maxwell equations and equation of motion for free non relativistic electrons(Noll,
2012). Taking the typical conditions of a laser-induced plasma for LIBS into account
simple expression for the absorption coefficient can be obtained
α =
ν
c
ω2p√
1− (ωp/ω)2
(4.1)
where ω is the angular frequency of the laser radiation, c is the vacuum speed
of light, ν is the collision frequency of electrons with ions and atoms and ωp is the
plasma frequency given by the equation
ωp =
√
nee2
ε0me
(4.2)
LIP produced in liquid or on the submerged target has initially very high elec-
tron number density Ne >10
19 cm−3 due to the strong confinement which limits
the expansion rate. Such a high electron number density strongly decreases the
ionisation energies of the species, due to the Debye effect on the potential energy
of the bounded electrons. (De Giacomo et al., 2012) As a consequence, the ioni-
sation degree is almost unity, i.e. ablated material is fully ionised and almost all
excited energy levels are forbidden due to the ionisation potential depression. Bal-
ance of the processes in this kind of plasma is moved from excitation-deexcitation
to recombination-ionisation pair.
High electron number density coupled with the high ionization degree leads to the
very strong continuum radiation due to the prevalence of the recombination pro-
cesses. Such an environment is suitable for the intensive chemical reactions through
39
4.2 Properties of the laser produced plasma in liquid environment
which the molecular species are easily formed, and relatively early on from the on-
set of the plasma (few hundreds of nanoseconds), compared to the times of the
molecular formation in the LIP in gaseous surrounding (several µs). This is also
the explanation for the widespread usage od the laser ablation (LA) in the liquid
for nanoparticles production. Efficiency and yield in the nanoparticles production is
largely dependent on the ablation mechanisms and material removal during LIB on
submerged target. Material removal during LA in the liquid occurs in several distinc-
tive phases. (Lazic and Jovic´evic´, 2014) Up to ∼100ns the material is evaporated by
the laser and hot plasma, and afterwards, intensive material expulsion begins due to
the impact of the compression wave, which moves towards the target during several
hundreds of ns. In some experiments (Lazic et al., 2013a)(Gavrilovic´ et al., 2016)
(Gavrilovic´ et al., 2017), additional phase of material removal is detected, where slow
target evaporation occurs in the interval when the cavitation bubble is already well
developed. In this late stage of LIB, cavitation bubble plays significant role in sus-
taining the plasma emission and thus producing additional material ablation, since
the thermal exchange between the target and bubble is significantly smaller than it
would be between the target and the liquid, which enables high temperature state
ot the material to lasts longer. Additional possibility for material ablation is the
collapse of the cavitation bubble, in the cases when the impact jet is formed.(Takada
et al., 2010)
4.2.1 Optical emission spectroscopy of laser-induced plas-
mas in liquid (LIBS)
Laser induced breakdown LIBS (Cremers and Radziemski, 2006) is a powerful tool
for in-situ elemental analysis in different surroundings (gases and liquids) and at
different ambient pressures. The technique is based on a plasma generation by an
intense laser pulse with duration in nanosecond range or shorter. From spectrally
resolved emission of the plasma containing excited atoms and ions belonging to sam-
ple constituents, the material composition can be determined, see Fig. 4.2. LIBS is
practically the only existing technique for in-situ chemical analysis of bulk liquids
and of submerged solid samples. (Guirado et al., 2012) Complete and thorough
characterization of LIB in liquid has posed a difficult scientific problem, both for
experimental and theoretical investigations, because of the number of involved com-
plex phenomena and complexity of the media itself (water or other liquid). Despite
40
4.2 Properties of the laser produced plasma in liquid environment
its wide range of applications, due to its peculiar structure, water still represents
especially complicated and interesting case of liquid with lots of unknowns during
the interaction with the high-irradiance laser beams. (De Giacomo et al., 2012)
Figure 4.2: Laser-induced plasma initiation and the recombination and relaxation
of emission lines,(Koch, 2012)
High acoustic pressure and long duration shock wave in liquids with respect to gas
environment lead to a manifold increase of the LA efficiency and also to a lowering
of the ablation threshold(Wook Kang— et al., 2008). Although the ablation rates
inside liquids are higher than in a gas environment, the LIBS signal at equivalent
laser excitation is significantly lower than in the presence of sample-air interface(De
Giacomo et al., 2007)(De Giacomo et al., 2005). Liquid environment has several
disadvantages for LIBS signal, among them are short lived optical emission, fast
plasma quenching, broadening of the very few emission lines, intense continuum ra-
diation and pronounced effect of the radiation self-absorption. Plasmas in the liquid
emit strong continuum radiation in the range from ultraviolet to infrared due to the
bremsstrahlung from free electrons and electron-ion recombination. This continuum
is the strongest immediately after the laser pulse impacts the target but after it
diminishes, instead of intense atomic and ionic lines, as usually seen during LIB in
gaseous surrounding, in underwater LIBS very often no analytical useful signal can
be detected. This is the consequence of the short plasma duration in the liquid sur-
rounding compared to the gaseous one. In some of the experiments on the SP LIB
in liquids the continuum was the only spectral feature that could be recorded(De
Giacomo et al., 2011) (Suzuki et al., 2002)(Ushida et al., 2007) Although the con-
tinuum can also be used as a spectroscopic tool to study the characteristics of the
plasma, see. e.g. (De Giacomo et al., 2010), it doesn’t provide elemental specific
41
4.2 Properties of the laser produced plasma in liquid environment
information which can only be obtained from the well resolved atomic spectra.
Besides high electron number densities, the underwater plasmas have substantially
lower temperatures compared to the LIP in the gases. This is a consequence of the
better thermal dissipation capacities of water compared to the air and vacuum and
large energy losses in mechanical effects and in liquid evaporation. Also a presence
of hydrogen from water strongly contributes to an effective thermalisation and cool-
ing of the plasma. Besides rapid thermalisation, oxidation is also very intensive in
underwater plasma, since the ionization energy of the water molecule is much lower
than oxygen molecule, and hence in water surrounding abundance of oxygen atoms
is increased compared to the gaseous surrounding. But, despite the abundance of the
oxygen and hydrogen atoms in LIP under water, low plasma temperature disables
the excitation of the highest excited energy levels, and thus, surprisingly, hydrogen
and oxygen emission are rarely observed in the LIP in the water, with few exceptions
(Escarguel, Ferhat, Lesage and Richou, 2000)(Kumar and Thareja, 2010) As men-
tioned previously, LIP in liquid is very suitable environment for chemical reactions
through which the diatomic molecules are formed, and their emission is easily ob-
tained due to the relatively low temperatures.(Sakka et al., 2005) (Lam et al., 2014)
Therefore one of the suggestions for characterisation of the LIP plasma in water is
to use this molecular band emission to probe the temperature conditions and tem-
poral evolution of the chemical composition. (Lam et al., 2016) These poor spectral
properties of the SP LIBS require the use of optimal gated and delayed acquisition,
which is usually reported to be in the order of 100 ns. (Lazic and Jovic´evic´, 2014)
The specific combination of high density ( 10 18cm−3) and low temperatures
(∼6000-15000 K) makes the underwater plasma extremely hard for characterisa-
tion, due to the breaking of many assumptions used in the spectroscopic theories
due to the plasma non-ideality. The usual parameters that are determined with spec-
troscopic measurements are plasma temperature, obtained usually from the Boltz-
mann plot method, and electron number density determined usually using Stark
effect(R.Griem, 1974). In the conditions when plasma deviates from ideality and
correlation effect are important, corrections to the Stark line broadening will also
be needed. This area of plasma research is far more complex and less understood
than the ideal plasma case (Fortov and Iakubov, 2000). Convenient parameter for
describing plasma deviation from ideality is the coupling parameter Γ (Stambulchik
et al., 2007), defined for a single-species gas of charged particles with charges zs and
42
4.2 Properties of the laser produced plasma in liquid environment
temperature T as
Γs =
z2se
2
rskBT
(4.3)
where e is the electron charge, kB is the Boltzmann constant, and rs is mean
interparticle distance given by rs=(4piNs/3)
−1/3 Eq. 4.4 can be rewritten using the
expression for Debye length λs=
√
(kBT/4piNse2z2s) in the following form
Γs =
1
3
(
rs
λs
)2 (4.4)
Due to the correlation effects in the plasma, corrections to the Stark line param-
eters are expected to be the same order of magnitude as coupling parameter value,
in so called, weakly-coupled plasmas. In weakly-coupled plasmas only the coupling
between the radiator,high Z-ion and the rest of the plasma is not negligible. But
situations is even more complicated in the case of low temperature dense hydrogen
plasma , where the interactions between the perturbers and corresponding correla-
tion effects can not be neglected. Hydrogen in dense, non-ideal plasmas have been
studied by many research groups and many opposing opinions and theories have been
developed(Kielkopf, 1995)(Stambulchik et al., 2007)(Kielkopf and Allard, 2014) (Vi-
tel et al., 1996)(Bu¨scher et al., 2002) (Bo¨ddeker et al., 1993)(Griem, 2000)(Calisti et
al., 2007)(Alexiou and Leboucher-Dalimier, 1999). Especial controversy arose about
the measurements of shifts and widths of the Hα line given in (Escarguel, Ferhat,
Lesage and Richou, 2000) (Escarguel, Oks, Richou and Volodko, 2000) for the un-
derwater laser produced plasma, which were discussed in several publications(Stehle
et al., 2000)(Alexiou, 2005)(Griem, 2001)(Halenka, 2004) (Griem et al., 2005)
Despite all the difficulties connected with the SP LIBS and complexity of the plas-
mas produced inside liquid environment, applications which demand rapid, in-situ,
non-contact analysis, all of which are the advantages of the LIBS, make that interest
in the subject of the laser produced plasma under water continuously grows. This
produces the need for the knowledge about the physics behind the process hence,
the less examined interrelations between the bubble and plasma emission during SP
LIB on submerged target are closely examined in this study.
43
Chapter 5
Experimental techniques for
underwater Laser Induced
Breakdown diagnostics
The laser induced breakdown on a solid target was generated using a Nd:YAG laser
(Molectron MY34) operated at 1064 or 532 nm , with 20 ns pulse duration and
a pulse energy of 40 mJ. The laser was run in low repetition rate (up to 5 Hz)
to ensure proper ablation event with every laser pulse and avoid time overlapping
of successive breakdown events. The laser beam was focused perpendicular to the
target plane by two quartz planoconvex lenses L1 ( f1 = 100 mm) and L2 ( f2 =
68 mm), where the second lens was mounted directly on the chamber wall. The
advantage of the lens embedded in the chamber wall is that the focused laser beam
propagates entirely inside the breakdown medium, thus avoiding aberrations on
the air-glass-liquid interface (Kennedy et al., 1997). Additional advantage is that
there is no splashing of the liquid which is known to pose a problem in underwater
laser ablation (De Giacomo et al., 2007) and the possibility of damaging the optical
components is thus minimized. The combination of the two lenses enabled smaller
focusing angles and tighter focusing on the target. Also, the additional lens in front
of the embedded one provided more flexibility during optimization of the focusing
conditions. Simply by changing the separation between the two lenses location of
the combination focal point given by the equation
s”2 =
f2(f1 − d)
f1 + f2 − d (5.1)
44
could be changed. The fact that the lens focal length becomes longer in water due
to the higher refractive index compared to air was taken into account. The limiting
factor was only the chamber’s dimension, or more precisely, the dimension of the
viewing lateral window (25.4 mm). The chamber was made in the laboratory from
optically transparent polycarbonate sheets (12 x 8 x 0.5 cm3). The chamber was
equipped with two quartz windows w1 and w2, used for the lateral observation of the
plasma and the vapour bubble. Several chamber dimensions and focusing conditions
were tested before the optimal geometry for stable and reproducible breakdown event
was found. Under the optimized conditions,the focal plane of the focusing system
was slightly below (∼0.5 mm) the surface of a target. This position of the focus
was already reported by (Nguyen et al., 2013), as optimal in respect to the the
most reproducible plasma with the stable position above the target surface. In this
way, undesirable breakdown on the suspended particles before the target surface
(Cristoforetti et al., 2012) was avoided, Fig.5.1
Figure 5.1: Photograph of the breakdown event with non-optimized (left) and opti-
mized focusing (right)
Different target materials were investigated, metals, ceramics and semiconduc-
tors, but the main part of the results regards the aluminium containing targets,
namely pure Al (99.999% ) and a ceramic α-alumina target. Both target materials
have very widespread use and significant part of the research connected to the un-
derwater laser ablation was performed with this type of materials. The target was
placed vertically inside a chamber filled with 700 ml of the liquid in the final cham-
ber version. The target was mounted on the target holder and shifted by 1 mm after
each laser shot to exclude the influence of the formed crater on subsequent plasma
formation. Vast majority of the results were obtained for distiled water, while some
experiments were performed in the tap water and etil-alcohol. The optical path
length of the laser beam through water in final configuration was ∼3 cm. To esti-
mate the losses of the input laser energy through the water column BeerLambert
45
law was used 2.3 as in (Lazic and Jovic´evic´, 2014) with water absorption coefficient
α = 0.072 cm−1 derived from the imaginary part of the refractive index k (Kou et
al., 1993) using the equation
α =
2pik
λ
(5.2)
By applying this approximation estimated laser energy reaching the target sur-
face was 60% of the input laser energy. Based on the crater diameter on the target
surface (250 ± 25 µ m), the estimated fluence within an error of 12% was about
45 Jcm−2(2.25 GWcm−2). Water inside the chamber was periodically exchanged to
avoid scattering on the particles produced by LA of the sample. The appearance
and mutual position of the chamber, target and associated optics is given in Fig.5.2.
Figure 5.2: Photograph of the interaction chamber and the target holder
Since the processes occurring after the laser target interaction in liquid surround-
ing are numerous and extremely fast, the whole picture and full understanding can
be obtained only by applying different experimental techniques. Most of the used ex-
perimental techniques in this work belong to the class of so called fast imaging tech-
niques, from which we can distinguish plasma imaging, shadowgraphy and Schlieren
techniques. Those techniques enabled insight in the spatio-temporal evolution of the
plasma, cavitation bubble and shock wave, respectively. The disadvantage of these
techniques is that by using single frame iCCD camera one can record only one time
46
5.1 Fast imaging techniques
delay from one laser shot, so in order to obtain full time evolution consecutive laser
pulses need to be applied. This implicitly calls for high stability and reproducibility
of the breakdown event, the condition less stringent for the case of the submerged
target than for the pure liquid, but still pretty demanding. To check the correctness
of such analysis complementary techniques of the transmission and scattering were
applied, where with one probe beam and from one pulse whole time evolution is
obtained, but without spatial resolution. Optical emission spectroscopy was used
as irreplaceable tool for elemental analysis of the material’s composition and char-
acterisation, and for the plasma diagnostics. In following brief explanation of these
experimental techniques will be given and used experimental setups explained.
5.1 Fast imaging techniques
5.1.1 Plasma imaging
Fast imaging was performed through the lateral window w1 by using a biconvex
achromatic lens L3 ( f3 = 100 mm) and a photographic objective lens OL (focal
length 75 mm, f 1.5) mounted with macro-bellows on an iCCD camera (Andor
Technology, model DH734I-18U-03, with 1024 x 1024 pixels, 13 x 13 m size, 18
mm intensifier diameter), see Fig.5.3. The optical magnification of the system was
determined using a reference object and the dispersion (0.01176 mm per px) was
calculated for subsequent recordings. The iCCD was controlled using a pulse gener-
ator (DG-535, Stanford Research Systems), triggered optically by a fast photodiode
that captures the laser beam partially reflected on mirror M2. The acquisition gate
width was varied between 15 ns and 50 µs. Shorter acquisition widths at earlier
delays were favourable to be able to track fast changing plasma shape and size. Ac-
cumulation of the signal was necessary at longer delays due to the very weak signal,
but was nevertheless performed also for the early delays to exclude the influence
of the eventual shot-to-shot changes in appearance. Images of the plasma were ac-
cumulated over 5 laser shots, except for long delays (hundred of µs from the laser
pulse) where mainly 10 pulses were used.
5.1.2 Shadowgraphy and Schlieren techniques
Shadowgraph and Schlieren techniques allow us to see the optical inhomogeneities
in transparent media, or schlieren, normally not accessible to our vision (Settles,
47
5.1 Fast imaging techniques
Figure 5.3: Experimental setup for fast imaging techniques: L1-L3 – lenses, M1–M3
– flat mirror, M4 – concave mirror, WLS – white light source, KE – knife edge, OL –
objective lens, and w1 and w2 – quartz windows.
2006),(Vasil’ev, 1971). Schlieren exists in liquid, solid and gaseous surroundings.
It can emerge due to the temperature changes, fast flows or mixing of different
materials. Both the shock wave and the bubble represent significant disturbances
of the refractive index of the surrounding media and as such are perfect candidates
for the Schlieren and shadowgraphy studies.More about the basic concept of this
techniques can be found in the ref. (Settles, 2006),(Vasil’ev, 1971). The same optical
setup previously described in Section 5.1.1 was also used for shadowgraphy adding
illumination using a white light source (WLS). The light from the WLS (halogen
lamp, 80 W) was guided by a fiber bundle (diameter 6 mm) to the chambers lateral
window w2, providing evenly distributed illumination over the observation volume.
The same setup was used for the Schlieren imaging, where a vertically mounted knife
edge KE was positioned in the focus of lens L3, thus monitoring the refractive index
gradient perpendicular to the sample. For shadowgraphs and Schlieren images only
one laser pulse was applied and the gate width was varied between 20 ns and 2 µs.
Due to very bright plasma emission and not enough intense back illumination the
earliest delays of the bubble and especially shock wave formation couldn’t be clearly
48
5.2 Transmission and scattering with the probe beam
discerned with the used experimental setup.
5.2 Transmission and scattering with the probe
beam
Light propagation through the region in which the interaction of the laser beam
with the target in the liquid has occurred is highly affected by the optical and
physical properties of the evolving bubble. To study this more closely the probing
techniques of transmission and scattering were applied. Transmission measurements
were performed under illumination of 2 mW HeNe laser, Fig 5.4. The laser beam was
expanded to ∼5 mm diameter and sent through the chamber windows parallel to the
target surface. On the opposite side of the chamber photomultiplier tube (PMT)
EMI 9659QB with interferential filter (IF) for He-Ne wavelength was placed and
used for the detection of transmitted radiation. For the scattering measurements
probe was green laser diode beam sent through the beam expander and directed on
the target with incidence angle of ∼60◦, Fig5.4b. The scattered light was collected
through chamber window and focused with lens L3 (f=5 cm, diameter 5cm) onto
the entrance slit of PMT equipped with suitable IF. The signal from the PMT was
monitored on the oscilloscope (model TDS2024C) and saved for the later processing.
Figure 5.4: Experimental setup for probe techniques: a) transmission b) scattering
L1-L3 – lenses, BE – beam expander, HeNe – helium-neon laser, LD – green laser
diode, IF1,IF2 – interference filters, PMT – photomultiplier tube
49
5.3 Optical emission spectroscopy (OES)
5.3 Optical emission spectroscopy (OES)
Two setups were evaluated for optical emission spectroscopy measurements. In the
first configuration plasma emission was recorded using setup shown in Fig5.5. Image
of the plasma plume was projected onto the entrance slit (10 m width and 2.5 mm
height) of the spectrometer (Shamrock sr-303i, Czerny-Turner type, focal length
303 mm, with the three gratings of 2400 grooves/mm, 1200 grooves/mm and 300
grooves/mm, f/4) using lens L3( f3= 170 mm). The plasma radiation was recorded
with the iCCD detector(Andor Technology, model DH720-18F-63, with 1024 × 256
pixels, 26 × 26 µm pixel size, 18 mm intensifier diameter) mounted on the exit slit
plane of the spectrometer.
Figure 5.5: Experimental setup for optical emission spectroscopy with lens
To avoid problem of chromatic aberrations introduced by the lens and to extend
wavelength optical range to the UV part of the spectrum, setup with mirrors was
tested and afterwards used for the rest of the OES measurements. By using mirrors
instead of the lenses for image projection the dependence of the focal length on the
thickness of the liquid layer is also excluded. In this setup the 1 : 1 image of the
plasma plume was projected, by means of optical mirrors M3 and M4 (see Fig.5.3) on
the entrance slit (100 µm wide) of a 0.5 m Ebert-type spectrometer, f/8.6 equipped
with a grating of 1180 grooves per mm. The iCCD detector, previously used for the
fast imaging techniques (Andor Technology, model DH734I-18U-03), was mounted
at the spectrometers exit plane and operated in an imaging mode. Advantage of
this camera in respect to the other one used for OES in lens configuration is in
extended wavelength range that covers due to the quartz entrance window (180-850
nm), smaller pixel size and the fact that image intensifier covers the whole matrix of
pixels. After construction of the suitable adapters to fit the camera onto the one of
50
5.3 Optical emission spectroscopy (OES)
the exit ports of the spectrometer, it was necessary to do the wavelenght calibration
of the system. This calibration and instrumental profile recording was performed
using low pressure Hg and Ne spectral calibration lamps. Resulting calibration
curves are shown in the Fig.5.6.
Figure 5.6: Wavelength calibration of the Jarrel-Ash spectrometer with the mounted
DH734I-18U-03 iCCD camera
Figure 5.7: Calibration curve - Intensity calibration: Without and with the water
in the chamber
The sensitivity of spectrometer, ICCD system and all optical elements in the
setup with mirrors was calibrated against standard tungsten coiled-coil filament
quartz halogen lamp (EGG, model597-1). The calibration curve was made for the
two cases, with and without the water in the chamber. The whole spectral range
from 200 nm to 900 nm was covered, but only the calibration curve for the wave-
lengths larger than 300 nm was used, since for the proper calibration at wavelengths
51
5.4 Data acquisition and processing
Table 5.1: Instrumental apparatus and components – final configurations
Component Specification
Nd:YAG laser Molectron MY34, λ = 1064 nm, τ = 20 ns, E=40 mJ
Spectrometer 0.5–m Ebert-type (Jarrell Ash 82-025)
Pulse generator DG-535, Stanford Research Systems
iCCD camera Andor Technology, model DH734I-18U-03
Chamber Made of optically transparent polycarbonate sheets
Target Alumina (α -Al2O3) and pure Al (99.999%)
Nd:YAG Dual laser Line
mirrors M1 and M2
BK7, diameter 25.4 x 6 mm, 450, 99.5%, 532 + 1064
nm, YAG 1H + 2H
Mirrors M3 and M4 Plain with aluminized surface M3, concave mirror focal
length 50 cm with aluminized surface M4
Lenses L1 , L2 and L3 Quartz plano-convex lenses L1 focal length 100 mm
and L2 focal length 68 mm, biconvex achromatic lens
L3 focal length 100 mm
Objective lens OL focal length 75 mm, f1.5
Windows Quartz windows w1 and w2
Light source WLS White light source with fiber bundle (diameter 6 mm)
IF1, IF2 Interference filters for 525 and 650 nm wavelength
lower than 300 nm deuterium lamp is needed. Spectral recordings were made with
large overlapping , i.e. for each recording wavelength was changed for 10 nm, so
the percent of overlapping points in two successive recording was 50. Resultant
calibration curve is shown in Fig.5.7.
The spectra were reconstructed by averaging over spatial axis in the spectral
recordings. Each LIBS spectrum was an accumulation over 160 laser pulses, where
each laser shot was delivered to a fresh spot on the target. Reproducibility of the
breakdown events - LIP - was monitored with photo-diode, simultaneously with
ICCD recordings. In order to improve reliability of the measurements, shots which
had signals from the photo-diode that deviated from some average signal were not
taken into account. Summary of instrumental apparatus and components used in
the final configurations for all experimental techniques is given in the table 5.1.
5.4 Data acquisition and processing
Data obtained from the fast imaging techniques are all in the matrix form. On the
image matrices some basic techniques for image manipulation and quality enhance-
52
5.4 Data acquisition and processing
Figure 5.8: Example of the image cropping
ment were applied (manipulating the contrast, histogram, colormaps etc.). Routine
written in the Matlab R© was used for correlating the pixel values and position in
respect to the target, based on the predetermined dispersion value. To be able to
measure the sizes of the plasma, bubble and position of the shock front, slight ad-
justments in the null position (target surface) between successive recordings were
applied (±3 px). After centering and transferring from pixels to millimeters, image
cropping was performed. In the program there was the additional option from image
symmetrisation and smoothing. The example of basic image cropping procedure is
shown in the Fig.5.8.
Figure 5.9: Spectral data manipulation: a) raw image b) image after background
subtraction and conversion in wavelength c) spectra obtained after summing the pixels
(vertically) with the useful signal
Regarding the processing of the spectroscopic measurements, the routine was
53
5.4 Data acquisition and processing
made to assign wavelength vector to matrix based on the name of the recorded file.
The file name had to be in the format xxxx xyyyyyyy.sif where xxxx x is central
wavelength in angstroms read from the spectrometer and the rest of the characters
are chosen arbitrarily. Bu using the determined values of dispersion and shift (see
Fig.5.6) and central wavelength from the input file name, the wavelength vector was
calculated and added to the matrix. For each recorded spectral range and every set
of experimental conditions (different gate widths on the sensor or different number
of accumulations applied) the image in the absence of the breakdown event was
recorded and subtracted from the plasma image. After that step, due to the low
signal intensity and strong continuum the binning of the pixels with the useful signal
(pixels whose intensity represent spectral emission) was performed. The care was
taken that spectral image falls approximately on the center of the sensor since in
that part sensor has the best linearity of the response. By selecting only the part of
the sensor with the useful signal and by averaging over that part, the level of noise
was significantly reduced compared to the case of the Full Vertical Binning (FVB),
where the whole spatial direction is integrated. The procedure is illustrated in the
Fig.5.9.
Few additional procedures used for specific tasks during image and spectral data
processing will be explained in Section 6 .
54
Chapter 6
Results
6.1 Plasma evolution
Peculiarities of the plasma formation and evolution inside the liquid environment
were captured using fast photography. Striking differences are immediately observed
when compared to the plasma evolution in the air at atmospheric pressure. Example
of the plasma evolution and formation on α-alumina in air and in distilled water
are given in Fig. 6.1. The most notable difference is reflected in the dimensions
of the two plasmas and their expansion speed into the background environment.
When laser induced plasma (LIP) is formed in gaseous surroundings it can expand
for several millimeters away from the target, depending among other factors, on the
pressure of the background gas . In contrast, water has significantly higher density
and orders of magnitude higher compressibility, both leading to very effective plasma
compression and its small size. As a consequence, a typical size of the plasma inside
liquid doesn’t go over 0.5 mm. Besides its small size, intensity of the plasma radia-
tion is decaying much faster compared to the plasma in gaseous surroundings, due
to the pronounced energy losses into mechanical effects, bubbles and shock waves.
This rapid energy loss leads to the fast cooling of the plasma, accompanied by high
recombination rate which contributes largely to the poor spectral characteristics of
the plasmas in liquid.
Besides being extremely small in size, plasma produced by laser ablation un-
derwater is usually quoted to have very short duration of emission (De Giacomo
et al., 2007)(Sakka et al., 2002)(T.Tsuji et al., 2007)(Saito et al., 2002)(Kim et al.,
2015)(DellAglio et al., 2015), and this is another big difficulty for experimental stud-
55
6.1 Plasma evolution
ies. Returning again to the Fig.6.1, although the small plasma size is also detected
in this experiment, plasma persistence is substantially longer than usually stated in
SP LIBS 1µs. To put in evidence extremely long plasma persistence following laser
ablation in liquid, series of photographs captured after laser ablation of aluminium is
given in Fig. 6.2. Similar long emission also exists on alumina target, and difference
between the plasmas on different target materials will be discussed in Sect. 6.5
z (mm)
 
 
0 1 2
−1
−0.5
0
0.5
1
z (mm)
 
 
0 1 2
−1
−0.5
0
0.5
1
z (mm)
 
 
0 1 2
−1
−0.5
0
0.5
1
z (mm)
 
 
0 1 2
−1
−0.5
0
0.5
1
z (mm)
 
 
0 1 2
−1
−0.5
0
0.5
1
z (mm)
 
 
0 1 2
−1
−0.5
0
0.5
1
z (mm)
 
 
0 1 2
−1
−0.5
0
0.5
1
z (mm)
 
 
0 1 2
−1
−0.5
0
0.5
1
z (mm)
 
 
0 1 2
−1
−0.5
0
0.5
1
z (mm)
 
 
0 1 2
−1
−0.5
0
0.5
1
z (mm)
 
 
0 1 2
−1
−0.5
0
0.5
1
z (mm)
 
 
0 1 2
−1
−0.5
0
0.5
1
z (mm)
 
 
0 1 2
−1
−0.5
0
0.5
1
z (mm)
 
 
0 1 2
−1
−0.5
0
0.5
1
z (mm)
 
 
0 1 2
−1
−0.5
0
0.5
1
z (mm)
 
 
0 1 2
−1
−0.5
0
0.5
1
z (mm)
 
 
0 1 2
−1
−0.5
0
0.5
1
z (mm)
 
 
0 1 2
−1
−0.5
0
0.5
1
z(mm) z(mm) z(mm) z(mm) z(mm) z(mm) z(mm) z(mm) z(mm)
D 0.1 µs D 0.2 µs D 0.5 µs D 1 µs D 2 µs D 4 µs D 6 µs D 10 µsD 0.8 µs
Figure 6.1: Fast photography of the plasma evolving from the target (placed on
the left) captured at different delays D from the laser pulse in air at a atmospheric
pressure (upper row) and in distiled water (lower row). Laser energy was 40 mJ.
From the photographs it can be seen that detected plasma emission changes sig-
nificantly during the time and part of the explanation for this long plasma lifetime
lies exactly in this change of the plasma’s shape. In the first microsecond after the
laser pulse usual picture of underwater plasma - very bright and very small plasma
confined by surrounding liquid to only 400µm diameter, is seen. At about 400 ns
from the laser pulse the plasma expansion perpendicular to the target slows down
drastically; its further expansion occurs mostly in the direction parallel to the target
surface. In this stage the plasma has a flat shape where its length parallel to the tar-
get is almost twice the height (direction normal to the target). The flat shape of the
plasma is dominant until ∼10 µs and afterwards expansion is more pronounced and
characterized by the formation of glowing centres and their clustering. A flat plasma
shape on submerged metallic targets was observed by different authors,(Thornton
and Ura, 2011) (Oguchi et al., 2007), but a long lasting emission (well beyond 1 µs)
was reported only recently.(Lazic et al., 2013a).
However, there are some studies that discuss about light emission on a longer
timescale of a few microseconds. (Lam et al., 2014) gave a temporal dependence
of the intensity of atomic Al lines and molecular AlO bands for a time period of
56
6.1 Plasma evolution
Figure 6.2: Fast photography of the plasma evolving from the target (placed on the
left) captured at different delays D from the laser pulse with acquisition gate width
W.
57
6.1 Plasma evolution
6 µs after the laser pulse. Although in ref. (Lam et al., 2014) the plasma plume
formation and shape were not imaged, the experimental data clearly demonstrate
that some emission still exists after several microseconds from the laser pulse. In
ref.(Oguchi et al., 2007), a visible plasma emission was documented up to delays
of 1000 ns and 2000 ns, for laser pulse durations of 20 ns and 100 ns, respectively.
To date, a long lasting plasma emission after a SP laser excitation underwater was
reported only by (Lazic et al., 2013a) where the plasma plume was detected up to
30 µs from the laser pulse. Besides the long duration of emission similar two-stage
plasma evolution is also reported in (Lazic et al., 2013a), where plasma shape is
changed from the spherical to flat at slightly longer delay of 1 µs. The first stage
lasts approximately 1 µs and is characterized by the rapidly expanding plume and
by ejection of macroscopic hot particles from the sample surface. In the second
stage, the newly forming plasma expands slowly and smoothly from the target into
the growing vapour bubble. This secondary plasma formation was explained by
complex interaction of the hot plasma front with the surrounding, which leads to
the propagation of a high temperature and high pressure region back to the target.
Similar phenomena were reported for LA in air(Zhou et al., 2011), (Tao et al., 2012)
where a newly formed emissive plume detected at 3 µs after the laser pulse was
attributed to the expulsion of the material by the backward reheated and melted
sample layer. Slow target evaporation through the backward reheated target into
the growing bubble was recently explained in the context of material removal during
LA in liquids.(Lazic and Jovic´evic´, 2014). This particular phase of LA is not always
observed and it depends on the experimental conditions.
Starting from the plasma photographs Fig. 6.2 for each acquisition delay the
integral plasma intensity was calculated. The appearance of the graphical user
interface (GUI) of the program used for this purpose is given in the Fig. 6.3.
The first step is to import the experimental matrix, which is then displayed as
an image in the lower part of the GUI. Afterwards, coordinates of the area over
which the summation of the plasma intensity needs to be performed is defined in the
section Define Rectangle and displayed over the plasma image as a white rectangle.
The parameters of the rectangle can be changed interactively. In the same way,
coordinates of the rectangle that represents background emission and noise that
needs to be subtracted from the signal are then inputed and shown as the pink
rectangle. Position of the background rectangle is chosen to be the furthest from
58
6.1 Plasma evolution
Figure 6.3: Graphical user interface for the calculations of the integral plasma in-
tensity
59
6.1 Plasma evolution
the plasma emission and its size is the same as the rectangle for the summation of
the plasma intensity. As a final step, subtraction of the background intensity from
the plasma intensity is performed. This procedure was repeated for plasma images
recorded at different delays.
In the next step, the obtained values were normalized on the gate width, the number
of accumulations and on the nominal signal intensity as a function of the applied
iCCD gain. The influence of the iCCD gain on the signal intensity was derived
from measurements using an external light source. Example of the images used for
calibration on iCCD gain is shown in the Fig. 6.4. All other acquisition parameters
were kept constant, only the values of the gain were changing, taking care not to
saturate the detector in the process and in this way the calibration curve for different
applied gains was obtained. The normalized integral plasma intensity as a function
of the acquisition delay for the pure Al target is shown in Fig.6.5.
G 0 G 20 G 40 G 80 G 120 G 160 G 190
Figure 6.4: Sequence of the images recorded under the same experimental conditions
only with varying applied gain used for the gain calibration
Figure 6.5: Temporal dependence of integral plasma intensity for early delays (main
figure) and later delays (inset)
High intensity plasma emission immediately after the laser pulse is followed by
rapid decrease of emission intensity in a few hundreds of nanoseconds as already
reported in the literature (De Giacomo et al., 2013). Temporal behaviour of integral
plasma intensity up to delays of 500 ns is well described with the equation 6.1
60
6.1 Plasma evolution
I(t) = a1e
[−(t−t0)/τ1] + a2e[−(t−t0)/τ2] (6.1)
,often used to fit the temporal evolution of the continuum intensity (De Giacomo
et al., 2013)(Kennedy et al., 1997) and is shown in the Fig.6.6
Figure 6.6: Early plasma emission intensity drop fitted with the exponential decay
according to the eq. 6.1
After this initial fast decay, sudden increase in total intensity is observed at
delay time of 500 ns . This phenomenon is ascribed to the coincident change in the
plasma shape and size, which is at that point well localized on the target and with
increased contact surface due to the flattening, thus producing more efficient heating
and additional material evaporation. This heating remains effective as long as the
plasma temperature is sufficiently high to produce target melting and evaporation.
Looking at the inset of the fig. 6.5 it can be seen that some weak residual emission
was detected up to delay of ∼475 µs.This time interval, as it will be shown later,
correspond to the time of the first bubble collapse.
In Fig. 6.7, the position of the maximum intensity plasma region above the
target is shown as a function of time. After the initial fast forward movement of the
most luminous plasma region (up to 250 µm from the target), the same moves back
to the target after less than 100 ns. This is visible in the inset of Fig.6.7. Similar
phenomena were previously observed in the laser ablation of targets in air.(Zhou et
al., 2011),(Tao et al., 2012). Since the ablation in liquid environment has additional
unique properties, other mechanisms driving this backward motion of the intense
plasma emission towards the target (formation of the cavitation bubble, rebound
of the internal shock wave, etc.) are also possible and additional investigations are
needed.
In addition to the plasma flattening, the position of the most intense emission
61
6.2 Bubble dynamics
Figure 6.7: The position of the maximum intensity plasma region with respect to
the target for later delays (main figure) and early delays (inset).
region remains almost constant for a few microseconds and it is situated up to 200
µm from the target, Fig.6.7.
Results presented in this Section show that high intensity, well localized and
stable plasma over a few microseconds is formed which is advantageous for opti-
cal emission spectroscopy analysis of single pulse laser induced plasma (Gavrilovic´
et al., 2016), (Gavrilovic´ et al., 2017), since it enables the use of large number of
accumulations and larger gate widths with respect to the initial, fast decaying emis-
sion. In contrast to this results, some papers(Oguchi et al., 2007),(Matsumoto et al.,
2013a),(Matsumoto et al., 2013b) reported the plasma images with a dark region
immediately above the target. In this way, heating of the target with the plasma is
less efficient, which can cause shorter (in the order of 100 ns) plasma persistence,
as reported by authors. Throughout the course of these studies, very long lasting
plasma emission ( > 100 µs) was detected. Appearance and properties of this late
emission coming from within the bubble will be discussed in the Sect.6.2.
6.2 Bubble dynamics
6.2.1 Plasma and bubble relations
As already mentioned in Sect.6.1, duration of the optical emission in these set of
experiments on a single pulse laser ablation events is in the order of hundreds of
µs. At this long time scales, cavitation bubble is formed and well expanded, leading
to the completely different physical conditions inside which the plasma is evolving.
62
6.2 Bubble dynamics
During the ablation of the pure Al target, some optically active species persist dur-
ing the whole bubble cycle i.e. until its first collapse. Spatial relations between
light emitting centres and vapour bubble was checked using superimposed images
obtained by fast photography and shadowgraphy. Since the optical setup was the
same for both imaging techniques, with addition of the WLS for the shadowgraphy,
using dimensionally calibrated images precise position of the light emitting centres
inside expanding bubble could be determined. From the Fig. 6.8, it seems that the
unique temperature and pressure conditions inside confining vapour bubble favour
the long lasting optical emission, since no emitting centers were observed outside
the vapour bubble. Similar set of photographs for the ceramic alumina was also
made and it is shown in the Fig.6.9. Also in this case, complete emission is coming
from within the bubble, but dispersion of glowing points was not observed. These
differences in appearance will be discussed in the Sect.6.5. Recent research about
position and origin of nano particles (NPs) in underwater ablation claiming that all
NPs are trapped inside the cavitation bubble (T.Tsuji et al., 2007)(Wagener et al.,
2013)(Soliman et al., 2010) seem to be supported by these findings, since all the
emitting species are contained inside the bubble.
From the results presented here it can not be claimed that there are no par-
ticles outside the bubble but only the ones trapped inside still hold temperature
high enough to emit some light. This surmise is further justified by temperature
measurement with AlO molecular band up to delays of 10 µs, that clearly shows
that early bubble holds temperature as high as 3500 K in employed experimental
conditions. (more about the temperature measurements can be found in the section
6.4). These temperatures agree with the values reported by (Casavola et al., 2005)
obtained by modeling of bubble dynamics.
6.2.2 Optical effects of the laser induced bubble
Furthermore, optical emitting region seems to never fill the entire bubble volume.
Spatial relation of bubble and optically active area is illustrated graphically in the
Fig. 6.10, where plasma radius normal and parallel to the target is expressed in
percents of the corresponding bubble radius. Due to the bright plasma emission
clear observation of the bubble and hence the bubble radius measurement, was not
possible before ∼3 µs delay for the pure Al target. Radius of the plasma in direction
normal to the target is 92% of the bubble radius at 3 µs delay, falls down to the
63
6.2 Bubble dynamics
Figure 6.8: Superimposed images of plasma photography (false colour) and shad-
owgraphy for the pure Al target
Figure 6.9: Superimposed images of plasma photography (false colour) and shad-
owgraphy for the alumina target
64
6.2 Bubble dynamics
minimum of approximately 62% for delay of 20 µs, and afterwards starts enlarging
again to approximately 70% .At the earlier delays of bubble expansion most of the
glowing centres are located in proximity of the target. As the bubble grows and
its internal pressure decreases these trapped particles start to expand more freely
inside the bubble volume. Still, no optical emission was observed from the upper
part of the bubble near the boundary with the liquid. The reason for this additional
confinement to approximately 2/3 of the bubbles dimension can be just apparent
reduction of the plasma size due to the optical properties of the bubble, which acts
as a negative lens (Lazic et al., 2012)(Lazic et al., 2013a). Rays originating near
the bubble edge and traveling parallel to the axis of the collecting lens are deviated
strongly by expanded and cooled bubble so that they never reach the detector,
creating the image of the emitting region with dark ring next to the bubble’s wall.
Figure 6.10: Percentage of the bubble volume occupied with the plasma for the pure
Al target
Optical effect of the bubble are even more pronounced when observing the im-
ages obtained with shadowgraphy in the phases when the bubble is fully expanded.
Imaging of the spherical bubbles and their effects on the light propagation have
been studied widely in the literature, (Bongiovanni et al., 1997)(Bongiovanni et al.,
2000)(Tiwari et al., 2010). Different interaction orders of the bubble and the WLS
used for the back lighting of the bubble are visible as high intensity points known
as a glory (Langley and Marston, 1991) or glare points (Dehaeck and van Beeck,
2007). Glare circle and a bright spot formation due to the multiple reflections at the
water-vapour interface during the imaging of the hemispherical and relatively cold
gas bubble is shown in the Fig.6.11. Bright spot in the center is formed from the
rays that have suffered N=2 interactions with the bubble wall, while the glare circle
65
6.2 Bubble dynamics
is formed from the rays that have experienced N=3 interactions with the bubble
wall.
Figure 6.11: Formation of the bright spot and glare circle due to the optical effects
of the expanded bubble on the pure Al target
Figure 6.12: Formation of the glare circle from the N=3 rays
The ratio between the glare circle diameter and the shadow contour was used
in the (Dehaeck and van Beeck, 2007) to calculate the refractive index of the liquid
with the accuracy reaching the third decimal. In the most of the studies involving
gas bubbles in liquid the refractive index of the bubble was considered to be constant
and equal to the refractive index of the air (nb=1). But, since in the course of the
bubble growing and shrinking, large variations of the pressure and temperature are
occurring, it is not realistic to assume that this condition will be fulfilled for the
laser induced bubble on the solid target, as discussed in (Lazic et al., 2012). In order
to assess the changes of the refractive index inside the bubble similar approach as
in (Dehaeck and van Beeck, 2007) was used and it is explained in the following .
Namely, by using geometrical optics approximation, the expression for the calcula-
tion of the refractive index inside the vapour bubble is derived. Starting from the
illustration of parallel rays traversing the bubble, see Fig.6.12, the Snell’s law for
66
6.2 Bubble dynamics
the incoming ray can be written in the form
cos(τ3)
cos(τ ′3)
=
nbub
nliq
(6.2)
while the dimension of the glory ring D3 can be expressed as D3=|AB|=Rcos τ3.
From the Fig.6.12 it can also be written τ3=2τ
′
3. Combining these two relations with
the Eq.6.2 and after some mathematical manipulation, relation between the bubble
diameter and the glory ring diameter can be obtained
D3 =
n2bub +
√
n4bub + 8n
2
liqn
2
bub
4n2liq
Db (6.3)
, where Db represents the determined bubble radius. Introducing a new dimen-
sionless parameter η=D3/Db and solving the Eg.6.3 two expressions can be obtained,
either to express the refractive index of the liquid or refractive index of the bubble as
unknown variable. If constant and known refractive index of the liquid is assumed,
and the dimensions of the bubble shadow DB and glory ring D3 are measured from
the images, transient refractive index of the laser induced bubble can be calculated
using the following equation
nbub =
√
2η2
η + 1
nliq (6.4)
Using the Eq.6.4, the value of the refractive index of the bubble in the maximum
expansion phase is unity, as previously obtained from the ray tracing models (Lazic
et al., 2012). The other measured values of refractive index in the time interval
of 100 µs up to 450 µs are in the range 1 - 1.07, which also agree with the values
obtained with the ray tracing in (Lazic et al., 2012).
The important thing to notice here is that this method doesn’t require any cal-
ibration, which is always advantageous. Although the accuracy of the method is
not known (bubble is hemispherical, and initial assumption is spherical bubble) and
it could only be applied for relatively late delays when the bubble is sufficiently
expanded, it provides experimental evidence of refractive index change. The preci-
sion of the measurement is mainly determined by the accuracy of the bubble size
determination since the glare circle has much sharper defined edges. Bubble sizing
methods are described in the 6.2.3.
67
6.2 Bubble dynamics
6.2.3 Bubble sizing
The procedure for determining the bubble size from dimensionally calibrated images
was written in the Matlab R© and the main steps are illustrated in the Fig.6.13
50 100 150 200 250 300 350
100
200
300
400
500
600
50 100 150 200 250 300 350
100
200
300
400
500
600
50 100 150 200 250 300 350
100
200
300
400
500
600
Figure 6.13: Steps in determining bubble size: Target position, bubble size along
the target, bubble size normal on the target, from left to right
The first step in determining the bubble size is to determine the position of
the target surface. This position (and the bubble edge also) is defined at 50% of
the difference between the maximum and the minimum intensity values along the
horizontal direction in the image 6.13. This way of the edge determination was
defined since the intensity falls from the maximum or ”white” to the minimum or
”black” over considerable amount of pixels (>10). Although the edge is located
somewhere among these gray pixels, it is impossible to obtain any more precise
information, as pointed out in (Bongiovanni et al., 1997) where it was concluded
that the intensity level at the mathematical bubble edge depends on the illumination
source, the location and the size of the bubble. Once the position of the target is
determined and displayed as magenta line, the bubble size is determined and marked
with the cyan line, see Fig.6.13, middle figure- size along the target, and right figure-
size normal to the target. This procedure for the bubble size determination works
well also in the cases when the bubble has less regular shape, which happens very
often during the collapse phases of the bubble. Extraction of the bubble size from
the images where the shape of the bubble is deformed is illustrated in the Fig. 6.14.
The procedure for the bubble sizing is coded in the way that it imports all the
experimental files with the extension .sif from the chosen folder, determines both
bubble sizes and displays them in the separate figures. For each imported file, the
procedure extracts from the name of the file the delay at which the bubble was
recorded and stores that information with the corresponding values of the bubble
size along and normal on the target. As a final step, based on the inputed value
68
6.2 Bubble dynamics
50 100 150 200 250 300 350
100
200
300
400
500
600
50 100 150 200 250 300 350
100
200
300
400
500
600
50 100 150 200 250 300 350
100
200
300
400
500
600
Figure 6.14: Size determination for bubbles with irregular shapes
of the spatial dispersion in mm/px graph of temporal evolution of the bubble size
is displayed, see Fig.6.15 where the outcome of the procedure is illustrated on the
example of the pure Al target in the final experimental configuration (see Sect.5).
0 200 400 600 800
0
1
2
3
4
 normal on the target
 along the target
target: pure Al
B
u
b
b
l
e
 
r
a
d
i
u
s
 
(
m
m
)
Delay ( s)
Figure 6.15: Output of the procedure for the bubble sizing
From the Fig.6.15 it is found that the maximum bubble expansion occurs 260 µs
after the laser pulse, and its radius is∼3.157 mm and ∼3.371 mm in the direction
parallel and normal to the target surface, respectively. The bubble remains slightly
elongated towards the laser source during the expansion. Differently, in the shrinking
phase a part of the bubble just above the target starts to deform, increasing the
contact angle, as evident in Fig.6.8 at a delay of 400 µs. Similar phenomena were
also observed by (Ibrahimkutty et al., 2015) who speculated that still high surface
temperature can be responsible for the observed change in the contact angle between
vapour, liquids and solids.
In this experimental setup on the pure aluminium, the bubble collapses after
about 475 µs from the laser pulse, evidenced by the release of the second shock wave
69
6.2 Bubble dynamics
in Schlieren images,see Sect.6.3. Based on the experimentally determined radius and
time of the bubble’s collapse , the energy contained in the bubble was estimated
using formula 6.5 widely used in the literature(Sasaki et al., 2009)(Cristoforetti et
al., 2012) and adapted to the cases of hemispherical cavity produced by laser ablation
in liquid(Soliman et al., 2010)(Sasaki et al., 2009). The difference between static
pressure in liquid and pressure of the vapour inside the bubble ( pstat - pv) is inferred
using Eq.6.6
Eb = V (pstat − pv) (6.5)
Tb = 2 · 0.915 ·Rmax
√
ρ
pstat − pv (6.6)
where Tb represents the bubble oscillation period(Thornton et al., 2013)(Franc
and Michel, 2004), i.e. time elapsed between bubble formation and the collapse
so if the bubble evolution is symmetric in time, bubble oscillation period is dou-
ble Rayleigh collapse time derived in Sect.3.2.2, eq.3.23. Eq.6.6 was derived for
spherical bubbles but has already been proven to be applicable for elongated bub-
bles(Thornton et al., 2013) as is the case here. Rmax for elongated bubbles cor-
responds to the radius of a sphere which has the same volume as the elongated
bubble.(Sasaki et al., 2009) In this way the energy stored in the bubble was calcu-
lated to be 10.8 mJ. Three bubble rebounds were detected after the initial collapse,
which lasted up to 900 µs after the laser pulse. During the successive rebounds the
bubble had less regular shape and it was flattened on the target.
From the results presented in this Section, it is evident that the large part of the
input laser energy is spent on the bubble formation.
6.2.4 Probe techniques for investigation of the bubble’s dy-
namic
The imaging techniques with the iCCD camera, despite its numerous advantages,
especially in the temporal and spatial resolution, suffer also from some drawbacks.
The main is that the temporal evolution of the bubble (or plasma or shock wave)
can not be recorded from the one laser shot. Laser probe techniques have been
used for a long time to obtain closer insight into processes of laser ablation (Koren,
1987) (Sell et al., 1989)(Diaci and Mozˇina, 1995) since they enable one to follow
70
6.2 Bubble dynamics
-200 0 200 400 600 800 1000
0.0
0.1
0.2
0.3
Transmission  Scattering
I
n
t
e
n
s
i
t
y
 
(
a
.
u
.
)
Time ( s)
 
Figure 6.16: Probe beam techniques on alumina sample, blue circles transmission,
magenta circles scattering
the temporal evolution of the processes from one laser pulse. Having this in mind,
besides imaging the bubble, two additional probe beam techniques were applied for
studying the bubble, and example of the result is shown in the Fig.6.7.
The technique (transmission) was recently applied in (Matsumoto et al., 2016)
to study the difference in bubble oscillations between laser ablation with long and
short nanosecond laser pulses, and it was called the laser-beam transmission probe
(LBTP). The same technique under the name beam deflection was used in the (Nath
and Khare, 2008) for studying the velocities of the charged particles and the bubble
wall. Slightly different setup called again beam deflection technique was used in
(Chen et al., 2004) for investigation of the shock wave and cavitation bubble os-
cillation. Very good agreement between the various probe and imaging techniques
was proved in several publications (Petkovsˇek and Gregorcˇicˇ, 2007)(Gregorcˇicˇ et al.,
2007) (Gregorcˇicˇ et al., 2008)(Nath and Khare, 2011)(Gregorcˇicˇ et al., 2016)(Chen
et al., 2017) giving the higher credibility to the probe beam techniques.Recently,
dependence of the transmission signal on the bubble radius was derived in the case
when the transmission signal was additionally blocked by an aperture in front of the
fast photodiode used for the signal detection(Koch et al., 2012)
The first thing to notice is that the two probe beam techniques applied here
show very good agreement in times of the bubble expansion and shrinking phases.
The transmission technique has slight advantage in signal strength and hence higher
71
6.2 Bubble dynamics
order bubble oscillations could be captured with it, as illustrated on the example of
the pure aluminium target 6.17, while with the scattering technique whose signal is
significantly lower, only the first bubble rebound could be seen in pure Al6.18. In
the case of alumina where successive rebound bubbles have very short duration and
small sizes, no successive rebounds were detected with the scattering method6.16.
Higher sensitivity of the transmission can also be seen in the interval when complete
period of bubble oscillation is over. Namely, as reported in (Matsumoto et al., 2016),
the transmission signal level after the bubble oscillation has finished doesn’t recover
completely to its initial value. The reason for this lowering of the ”undisturbed”
transmission signal intensity can be the presence of the material that was trapped
inside the bubble. After the bubble oscillation period ends, material is expelled
in the liquid and stays preferentially in the proximity of the target blocking the
part of the probe beam and thus causing the intensity of the transmitted signal
to be lower, see Fig.6.17. This effect is less pronounced for ceramic alumina target
and it is related to material’s characteristics which will be addressed in the chap.6.5.
0 500 1000 1500 2000 2500 3000 3500 4000
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
 
T
r
a
n
s
m
i
s
s
i
o
n
 
s
i
g
n
a
l
 
Time ( s)
Initial bubble
1st bubble rebound
2nd bubble rebound
3rd bubble rebound
4th bubble rebound
Expelled material 
 
 
Figure 6.17: Complete bubble evolution from one laser pulse recorded with trans-
mission technique - four bubble rebounds detected. Decrease in the signal after the
extinction of the bubble is attributed to the material expelled in the liquid.
Since the transmission technique is quite simple and gives complete evolution of
the bubble in time, the question remains whether this changing signal intensity can
72
6.2 Bubble dynamics
-250 0 250 500 750 1000 1250 1500
-0.1
0.0
0.1
0.2
0.3
0.4
0.5
Sc
a
tt
er
in
g 
sig
n
a
l
Time (m s)
 Raw scattering  
 Reference (probe beam off)
 Scattering minus reference 
 
 
Figure 6.18: Complete bubble evolution from one laser pulse recorded with scatter-
ing technique-only the first bubble rebound detected.
somehow be linked the to time evolution of the bubble radius. The procedure applied
for that cause is similar to the one described in (Chen et al., 2017) ,(Chen et al.,
2004) and (Lazic et al., 2013b) and it is based on the change of the distance between
the probe beam and the target. Besides changing the distance, the probe beam was
additionally restricted with the rectangular slit to 100 µm width, in order to probe
the different heights above the target. The concept of the technique is to determine
the distance from the target, at which the characteristic dip in the transmission
signal due to the interception of the probe beam by the growing vapour bubble,
disappears. The typical result is shown in the Fig.6.19 for the case of the ceramic
alumina target. None of the transmission signals detected with the slit showed
clear bubble rebounds, possibly due to the significantly lower intensity of the probe
beam, whose size is reduced from the ∼5 mm to only 100 µm. The relative change
in transmission signal intensity is not the same for all investigated distances.
Namely, for the closest examined distance above the target d=0.2 mm, the de-
tected decrease in the signal intensity due to the appearance of the bubble is rela-
tively small, light green curve in the main Fig.6.19. As the distance of the probe
beam above the target increases, the relative change of the signal intensity enlarges
and reaches its maximum change (minimum in the transmission signal) for the dis-
73
6.2 Bubble dynamics
-200 0 200 400 600 800 1000 1200 1400 1600 1800 2000
-1.0
-0.8
-0.6
-0.4
-0.2
0.0
T
r
a
n
sm
is
si
o
n
 
a
t 
v
a
r
io
u
s 
di
st
a
n
c
e
s 
a
bo
v
e
 
th
e
 
a
lu
m
in
a
Time ( m s)
 0.2 mm   0.5 mm   0.8 mm    1 mm   1.15 mm   1.30 mm   1.50 mm        
 
 
-200 0 200 400 600 800 1000
-0.016
-0.012
-0.008
-0.004
0.000
 1.6 mm   1.8 mm
 1.9 mm   2 mm
 2.1 mm
 
 zoomed in
Figure 6.19: Transmission signal on alumina target for different positions above the
target surface. Inset figure shows signal for distances from the target which are close
to the bubble radius value
tance d=0.8 mm from the target surface, blue curve in the main Fig.6.19. Further
increase of the distance results in shallower dips, with decreasing durations. The
longest duration of the signal perturbation is detected at the distances closest to the
target, which is explained through the joined effects of the bubble and ejected ma-
terial in proximity of the target surface and on the probe beam transmission. This,
at the first glance unusual, increase of the transmission dip with departure from the
target might be explained by analysing the propagation of the characteristic rays
through the expanded bubble, with the refractive index ∼1. Three representative
beam positions are depicted in the Fig.6.20. The rays traveling very close to the
target surface encounter the bubble surface under very small angles θ1, thus expe-
riencing the least deviation from their original path. With increase in distance, the
angle of interaction increases, leading to the complete deviation of the incoming
beam which cannot reach the detector. Further increase, on the contrary, enhances
optical effects of the bubble, already mentioned in the Sect.6.2.2 and causes more
rays to be able to reach the detector after multiple interaction with the bubble wall.
In the inset of the Fig.6.19 the signals at the furthest distances of the probe beam
74
6.2 Bubble dynamics
from the target are zoomed in. The last signal with observable dip corresponds to
the separation of the probe beam and the target of ∼2.2 mm, leading to the conclu-
sion that the bubble diameter is at least 2.2 mm. This value is just the lower limit
for the maximum bubble radius determined using transmission, since the beam at
distances from the target comparable to the bubble size, is partially transmitted
directly to the detector without interaction with the bubble, masking the weak in-
tensity transmission dip . From the dimensionally calibrated images of the bubble
on alumina in the same experimental conditions, maximum bubble radius is deter-
mined to be 2.3 mm, which means that this form of the transmission technique can
be used for bubble sizing with uncertainty better than 20%.
d1
d3
d2
TARGET DETECTOR
θ1
θ2
θ3
Figure 6.20: Explanation of the transmission with the slit
The second probe beam technique, scattering is less sensitive and slightly more
difficult for alignment compared to the previously described transmission technique.
It was previously shown how to measure maximum bubble radius with the transmis-
sion technique. Scattering signal with subtracted background, on the other hand,
can give insight in complete evolution of the bubble size if independent calibration
is provided (Lentz et al., 1995). Typical shape of the scattering signal is shown in
the Fig6.7 together with the corresponding transmission signal on alumina target,
while in the Fig.6.18 the measured scattered signal for the pure Al target is shown.
The focusing conditions and chamber dimensions were different for the two targets,
but since there will be no comparison between the two targets, this fact is for now
irrelevant.
In the Fig.6.18 raw scattered signal exhibits and additional peak corresponding
to the plasma emission. Although the scattered radiation was recorded with the
interference filter, the plasma radiation still has very high intensity, in some cases
even stronger than the useful scattered signal. To observe just the pure scattered
signal, the signal from the plasma radiation in the absence of the probe beam was
75
6.2 Bubble dynamics
0 2 5 0 5 0 0 7 5 0 1 0 0 0
0 . 0
0 . 5
1 . 0
1 . 5
2 . 0
2 . 5
3 . 0
3 . 5  C a l c u l a t e d  f r o m  s c a t t e r i n g
  M e a s u r e d  f r o m  s h a d o w g r a p h y
 Ra
diu
s (m
m)
D e l a y  ( m s )
p u r e  A l   
 
 
0 2 5 0 5 0 0 7 5 0 1 0 0 0
0 . 0
0 . 5
1 . 0
1 . 5
2 . 0
2 . 5  C a l c u l a t e d  f r o m  s c a t t e r i n g
 M e a s u r e d  f r o m  s h a d o w g r a p h y
 Ra
diu
s (m
m)
D e l a y  ( m s )
A l u m i n a
 
 
Figure 6.21: Comparison of the calibrated scattered signal based on the maximum
bubble radius determined from the shadowgraphy measurements and bubble radius
evolution determined from shadowgraphy
76
6.2 Bubble dynamics
recorded (red curve in the Fig. 6.18) and than subtracted from the recorded scattered
signal. This procedure was applied for every target and all experimental conditions.
As discussed in sec.3.2.4, the underlying physics of scattering technique is Mie’s
scattering. But, in our experimental conditions, the bubble has dimension in the
mm range and the condition r/λ1 is fulfilled, where r is the bubble radius and λ
is the wavelength of the probe beam light. In such cases, much simpler geometrical
optics approximation can be used to describe the scattering from the bubble. Both
Mie and geometrical optics approximation correlate the scattered light intensity with
the squared value of the bubble radius, irrespective of the observation angle. In this
way, recorded signal, which is actually photomultiplier voltage, can be expressed as
a function of the bubble radius:
R(t) ∝ T (t) = [V (R = 0)− V (t)](1/2) (6.7)
where V(t) is the PMT voltage and V(R=0) is the background scatter detected
when there is no bubble present(Barber and Putterman, 1992). The only missing
thing is the constant of proportionality, which can be obtained through some sort
of the independent calibration for (at least) one point. The independent calibration
applied here was the fast shadowgraphy technique. The choice was to calibrate at
the point of the maximum scattered signal (highest signal to noise ratio, SNR) and
to relate that value to the maximum bubble radius obtained from the shadowgraphy.
In this way calibration constant is determined, and used to retrieve the complete
evolution of the bubble radius from the scattered signal intensity. The results of this
procedure on the example of the pure aluminium and alumina target are shown in
the Fig.6.21. The agreement is quite good for the first bubble, while higher order
bubble oscillations couldn’t be reproduced. The other possibility for an independent
calibration is the maximum bubble size determined from the transmission measure-
ments with the slit.
To conclude, although the accuracy of the bubble sizing with the probe beam
techniques is lower compared to the fast shadowgraphy, this combination also offers
several advantages: i) There is no need to use an iCCD camera, which significantly
reduces the cost of the equipment ii) the alignment procedure is much simpler, espe-
cially for the transmission iii) complete evolution can be obtained in just several laser
shots (one shot to record the scattered signal and a few to scan the different heights
above the target and determine the maximum bubble size with transmission).
77
6.3 Shock wave propagation
6.3 Shock wave propagation
Besides the bubble, other important mechanical effect in LA underwater is the
shock wave (SW) which emerges, as a result of the disturbance induced by the
pressure developed during plasma formation(Delale (Ed.), 2013), Fig.6.22 Based on
the measured position of the outer shock wave front, on the pure Al target, initial
velocity of the shock wave is estimated to be 3 km/s at 100 ns delay but drops down
very quickly to 1.6 km/s after 1 µs. Based on the determined shock wave velocity,
the peak pressure as a function of a shock wave velocity can be obtained using the
equation 3.6 in Chap.3 (Delale (Ed.), 2013)(Kennedy et al., 1997).
Figure 6.22: Images of the shock wave on pure Al obtained using the Schlieren
technique, at different delays D from the laser pulse; W is the acquisition gate width.
Derived values of pressure are 2.29 GPa at the earlier delays and 86 MPa after the
1 µs, similar to those reported in the literature (T.Tsuji et al., 2007).The velocity
of the second shock wave could not be determined with employed measurement
procedure due to the slight fluctuations in the time of the bubble collapse and
subsequent release of the second shock wave (±2 µs). Second collapse did not
produce observable shock wave. Determination of the energy consumed by the
shock wave was not possible since the largest part of the shock weave energy is
deposited in the vicinity of the target where and in very early stages of the LIB in
which the measurement of the shock wave front propagation was not possible due
to very bright plasma emission.
Based on the calculated velocities and pressures and having in mind the work
of (Vogel et al., 1999) it can be stated that a large part of the input laser energy is
spent on the shock wave.
6.4 Optical emission spectroscopy results
Optical emission spectra recorded from the laser induced plasma inside liquid envi-
ronment has several particular characteristics, which make it very different from the
78
6.4 Optical emission spectroscopy results
one in the gaseous surrounding. Since in the employed experimental setup plasma
evolves in the two phases, primary and secondary, described in the Sect.6.1 spec-
trally resolved emission also differs significantly between the two phases. The typical
behaviour of the emission signal in the both phases will be described on the example
of the metallic aluminium and ceramic alumina target.
6.4.1 Primary plasma phase
Spectral characteristics of the plasma immediately after the laser pulse impacts the
target are very similar to the ones usually reported from underwater plasmas. The
one that makes the plasma diagnostic from this spectra extremely difficult and is
the most pronounced is the intense continuum emission. Number of papers exist
where the continuum radiation was the only spectral feature that could be recorded
(Suzuki et al., 2002)(Ushida et al., 2007). Spectral line emission superimposed on
strong continuum emission is illustrated on the example of the spectra from alumina
target, recorded at a delay of 500 ns shown in the Fig. 6.23.
Figure 6.23: Typical spectra acquired after LIB on alumina. Blue curve represents
continuum contribution which has been subtracted from experimental spectrum (black
curve) and result (magenta curve) is marked as corrected spectrum.
The raw recorded spectra is displayed as a black line, while the spectra after the
subtraction of the continuum component (the blue curve) is shown as a magenta
line. For the case of the pure Al target, the continuum contribution is even more
79
6.4 Optical emission spectroscopy results
pronounced, especially at the early delays of the 0.1 µs which is the earliest delay at
which the spectroscopic measurements were performed. The temporal evolution of
the spectra on the pure Al target is shown in the Fig6.24, demonstrating difficulties
that one encounters when performing spectroscopic measurements at delay times ¡
1 µs, as usually done in the literature. It can be seen in both cases that emission
bellow 350 nm is very low, partially due to the strong absorption by water in that
wavelength range.
300 400 500 600 700
0
2
4
6
8
10
12
14
In
te
n
sit
y 
(a.
u
.
)
Wavelength (nm)
 D 0.1 m s    D 0.5 m s   D 1 m s     D 2 m s          
 
 
Figure 6.24: Strong continuum radiation in the spectrum of the pure Al target
recorded at delays of 0.1,0.5,1 and 2 µs after the laser pulse
Procedure for evaluation and subtraction of the background from the recorded
spectra was written in the Matlab R© according to the procedure described in the
(Rekhi et al., 2003). Graphical user interface of the program for the continuum
component evaluation and subtraction is shown in the Fig.6.25. Besides the need
to remove the continuum component that might interfere with the line emission,
the continuum itself is recognized as valuable tool for diagnostics of laser induced
plasma(De Giacomo et al., 2010). In underwater laser plasmas the value of informa-
tions extracted from the continuum increases due to the lack of ionic emission lines
and a few atomic emission lines, as can be seen in the Figs.6.236.24.
The program illustrated in the Fig.6.25 works as follows Pressing the buttons
’Import data’ and ’Import CF’ opens the dialog where the user can navigate to the
requested file. If the ’show’ check boxes are on the inputed spectra is displayed in
the main window and calibration curve in the separate figure. Then the user inputs
80
6.4 Optical emission spectroscopy results
in the field ’temperature’ estimated value of the temperature for which the contin-
uum component should be created. The next three buttons ’Planck’,’Emissivity’
and ’Int.Corr’ create the Planckian curve for the inputed temperature value in the
spectral range determined by the experimental spectrum, correct it for the emissiv-
ity of the ”grey” body and for the spectral response of the optical system used for
the recording of the spectra. All of the steps can be displayed in the main window
of the GUI, as desired. Since all the curves and the experimental spectrum are nor-
malized to unity, it is necessary to scale the resultant ’Int.Cor.’ curve to compare
it with the spectrum, which is achieved by entering the scaling factor in the box
’scale’. This curve is shown in the example of the GUI in 6.25. The process can be
repeated until the satisfactory agreement is obtained. Based on the wavelength for
which the generated curve reaches its maximum value the temperature is evaluated
using the Vien’s displacement law, eq.6.8
λ =
2.8977682 · 106[nmK]
T
(6.8)
Figure 6.25: Graphical user interface of the program for determination and sub-
traction of the background component from the recorded spectra and temperature
evaluation based on the Planck’s law
When the correct background curve is determined it can be subtracted from
the original spectra by pressing the button ’Background subtracted’ and corrected
for the calibration function of the system (’Intensity CorrSpectra’). ’Save data’
option saves altogether the experimental data, continuum curve, corrected spectra
and determined value of the temperature from the Vien’s law.
The usual values of the electron temperature obtained by fitting the continuum
81
6.4 Optical emission spectroscopy results
component with the Planck’s law are in the range 6000-7000K. This temperature
represent just an estimate averaged over the gate width of 0.5µ. Still, at the earliest
delays after the laser pulse for which the spectra were recorded, 0.1 µs, this is the only
temperature information that could be obtained, since no well resolved line profile
was recorded at that delay for any of the investigated materials. The example of
the line emission quality at early delays is shown in Fig. 6.26.
390 395 400 405
0
2
4
6
8
10
12
14
In
te
n
sit
y 
(a.
u
.
)
Wavelength (nm)
 Delay 0.1 m s Delay 0.3 m s Delay 0.5 m s
 
 
Figure 6.26: Absence of line emission in primary plasma phase on aluminium and
slow emerging of the line at delay of 0.5 µs when the continuum intensity drops
significantly
The line emission that was present only in the primary plasma phase is emission
from the hydrogen Hα line on both the pure Al and ceramic alumina, and oxy-
gen triplet 2s22p3(4S0)3s-2s22p3(4S0)3p (5S0-5P) on alumina. Profiles of the Hα and
oxygen triplet were reported in the electrical discharge machining plasmas (Descoeu-
dres et al., 2008) (Descoeudres et al., 2005)(Descoeudres, 2006) and laser induced
plasmas at a water/gas interface(Adamson et al., 2007)(Lo and Cheung, 2002) but
these line emissions are seldom observed in the plasmas produced by the laser in the
liquid environment. Few examples to the contrary are (Escarguel, Ferhat, Lesage
and Richou, 2000) for laser induced plasma in the pure liquid and (Kumar and
Thareja, 2010) for the laser induced breakdown on the submerged aluminium tar-
get. Although, at the first glance, that might seem odd, few explanations for this
rare appearance of hydrogen and oxygen emissions exist. Namely, temperatures of
the laser produced underwater plasmas are relatively low (several thousands of K
(Kennedy et al., 1997)) due to the higher thermal dissipation capacities of liquids
82
6.4 Optical emission spectroscopy results
compared to the gases(Lam, 2015). As a consequence, ionic lines and lines with the
higher excited state energies are not frequently observed. Large energy spent in the
shock wave emission and Bremsstrahlung emission additionally cools the plasma,
which increases the rate of the electron-ion recombination (De Giacomo et al., 2007)
making the emission from the higher excited states less possible.
7 7 1 7 7 4 7 7 7 7 8 0 7 8 3 7 8 6
1 . 0
1 . 5
2 . 0
2 . 5
3 . 0
W a v e l e n g t h  ( n m )
 O  I D e l a y  2 0 0  n s
W i d t h  1 0 0  n s
 
 
6 0 0 6 2 0 6 4 0 6 6 0 6 8 0 7 0 0
0
4
8
1 2
1 6
Inte
nsi
ty (
a.u
.)
W a v e l e n g t h  ( n m )
H aD e l a y  2 0 0  n sW i d t h  1 0 0  n s
 
 
Figure 6.27: Profiles of the Hα and oxygen I line recorded in the first plasma phase
on the alumina target
Examples of the hydrogen and oxygen line profiles at delay 0.2 µs for ceramic
alumina are shown in Fig6.27. On the metallic aluminium target the Hα line was
recorded only at delay of 0.1 µs, while for the ceramic alumina it was detected up to
0.8 µs. Oxygen emission is extinguished rapidly (visible up to 0.4 µs) not only due
to the high excitation level of the transition, but also due to an intense oxidation of
the target species, as visible from AlO emission appearing already 100 ns after the
laser pulse.
Hα line was used for the evaluation of the electron number density, since the
higher member of the series, which would be more reliable, were not detected in the
spectrum. This method of the electron number density determination is convenient
since it does note require knowledge of the plasma temperature. The width of the
line is extremely large, going up to 26 nm for the pure Al target at a delay of 0.1
µs. Using the equation for electron number density diagnostics from the (Konjevic´
et al., 2012)
Ne[m−3] = 1023 ∗ (wSA[nm]
1.098
)
1.47135
(6.9)
83
6.4 Optical emission spectroscopy results
electron number densities in the range (0.6-7.7)*10 18cm−3 for ceramic alumina
were obtained The spectral profiles were fitted with the Voigt function, and Lorentz
width was used in the eq.6.9. Gaussian part was kept fixed and it was set to
determined value of instrumental profile of 0.105 nm, which was negligible compared
to extremely large widths of the profile. Typical fit of the Hα line and derived
temporal evolution of the electron number density on the plasma from alumina is
shown in the Fig.6.28
6 0 0 6 1 0 6 2 0 6 3 0 6 4 0 6 5 0 6 6 0 6 7 0 6 8 0 6 9 0 7 0 0 7 1 0 7 2 0
0
1 0 0 0 0
2 0 0 0 0
3 0 0 0 0  E x p .  d a t a V o i g t  f i t  
Inte
nsit
y (a
.u)
W a v e l e n g t h  ( n m )
D e l a y  0 . 2  m sG a t e  w i d t h  0 . 1  m s
N e = 5 . 7 2  *  1 0 1 8  c m - 3
 
 
0 1 0 0 2 0 0 3 0 0 4 0 0 5 0 0 6 0 0 7 0 0 8 0 0 9 0 00
2
4
6
8
1 0 N e  [ m - 3 ]  =  1 0 2 3 * ( w s  [ n m ]  /  1 . 0 9 8 ) 1 . 4 7 1 3 5  
N e(1
*10
18 cm
-3 )
D e l a y  ( n s )
T i m e  e v o l u t i o n  o f  N e  d e t e r m i n e d  f r o m  H a  p r o f i l e  
 
 
Figure 6.28: Experimental profile and Voigt fit of the Hα line at delay of 0.2µs(upper
figure) Temporal evolution of the Ne determined from the Stark widths of the Hα using
the eq.6.9(Konjevic´ et al., 2012)
One has to keep in mind that possible influence of self-absorption on the line
width was not checked, which increase significantly the uncertainty in electron num-
ber density determination In addition, the underlying physical assumptions (Kon-
84
6.4 Optical emission spectroscopy results
jevic´ et al., 2012) used in deriving the eq.6.9 are suited to use in low electron number
diagnostics, so the uncertainty of this method of Ne determination in high density
plasma can not be precisely determined. Having this in mind, the error bars in the
Fig.6.28, especially at the earliest delays are set to the high values (∼50%).
Another approach used for the electron number density determination from the Hα
profile is based on the whole profile fitting using the tabulated line profiles from
(Gigosos et al., 2003). The tables cover very wide range of the electron densities,
but for the range of densities of the interest here ( 1018 cm−3), assumed plasma
temperatures need to be significantly higher so as to be able to neglect the effects of
the plasma non-ideality and coupling effects. If neglecting the fact that the actual
temperatures are supposed to be lower, determined densities are in the same range
(±20%), while the overall agreement of the generated and experimental profiles is
slightly worse than with the Voigt fit. The absence of the other lines of the hydrogen
can also be used as a method for the estimation of electron number density using
the Inglis-Teller relation
log n = 23.26− 7.5 · log nmax (6.10)
which connects the value of the electron number density in cm−3 with the quan-
tum number of the upper energy level of the last observable emission in the spectrum.
(Descoeudres, 2006). Since the last observable line is the Hα (transition from the
level 3 to 2) and there was no Hβ line in the spectrum (transition from the level 4
to 2) it follows
3 ≤ nmax ≤ 4 (6.11)
which substituted in eq.6.10 gives the lower limit of electron number density of
6*1018 cm−3. This crude estimate agrees with the values obtained from the Hα line
width at the earliest delays. One more attempt was made to obtain information
about electron number density in the primary plasma phase, by matching the ex-
perimentally recorded profile of oxygen I triplet at 777 nm with the generated one
using the atomic data form the (R.Griem, 1974) and the line parameters from the
NIST database (Kramida et al., 2015). The synthetic profile in the Fig. 6.29 is
generated for the T = 8000 K. Since the exact temperature value at this early delay
can not be determined, and the temperature measurements from the continuum,
integrated over the time interval 100-500 ns, gives the temperature values around
85
6.4 Optical emission spectroscopy results
6000 K, it is believed that 8000 K represent reasonable assumption. For this electron
temperature and density combination the screening parameter (or Debye shielding
parameter)(Konjevic´, 1999) given by the equation
R = 8.99× 10−2N1/6e T−1/2e (6.12)
has the value of 1.31. Original Hooper’s distribution functions for a low fre-
quency electric micro field arising from singly charged ions which interact through
the shielded Coulomb potential (Hooper Jr, 1968), used for the generation of the
neutral line profile as in (Cvejic´ et al., 2013), are given only for the values of R<
0.8. However, in (D’yachkov, 1998) analytical approximation for the cumulative
Holtsmark and Hooper micro field distribution is presented for the values of R up
to the 1.4. Although there is no physical justification for the extrapolation of R> 1,
since the Debye radius becomes less than the ion-sphere and loses the meaning of
the screening radius, formal extrapolation is possible up to the R∼=1.5. In addition,
this analytical approximation was compared against the Model Micro field Method
in (Stehle et al., 2000) and good agreement was found. It is concluded that the
extrapolation is valid up to the R< 1.4. These results justify the use of the oxygen
I triplet for the electron density measurement, and again confirm the high electron
density present in the fist phase of the LIP. Still, one has to keep in mind that the
plasma parameters are out of the range of commonly used theories of spectral line
shapes so great care is needed when applying analytical formulas and interpreting
the obtained results.
6.4.2 Secondary plasma phase
If the plasma emission extinguishes within the first few hundreds of nanosecond, as
often reported in the literature, than little information can be obtained from the
spectra, since the continuum component is very strong in that period. But if the
secondary plasma is formed, good quality line profiles can be recorded, since the
spectral characteristics of the secondary plasma are much better due its expansion
inside the bubble. In the spectrum of the secondary plasma no ionic emission is
present since the recombination occurres fast in the primary plasma phase, and due
to the low temperatures, molecular bands dominate the spectra. When working both
on metallic aluminium and ceramic alumina target, blue-green AlO emission was
very intensive from the 500 ns on, and it was used for the vibrational temperature
86
6.4 Optical emission spectroscopy results
770 772 774 776 778 780 782 784 786
0
5000
10000
15000
20000
In
te
n
sit
y(a
.
u
.
)
Wavelength (nm)
 Experimental profile
 Generated profile
Delay 0.2 m s
Width 0.1 m s
N
e  =  5 *  1018cm-3
 
 
Figure 6.29: Experimental and generated profile of oxygen I triplet at delay of 0.2µs.
Atomic data for the generated neutral line profile are taken from the (R.Griem, 1974)
determination.
Figure 6.30: Example of recorded and synthetic spectra of AlO vibrational band.
In the inset figure typical Boltzmann plot is shown. Temperatures determined from
Boltzmann plot are used for synthetic spectra generation using program from (Parigger
et al., 2015)
87
6.4 Optical emission spectroscopy results
In the Fig.6.30, the vibrational temperature determination for the alumina target
at 0.5 µs delay is shown. Temperature was determined in two different ways. The
first one is use of Boltzmann plot technique, described in (Ornstein and Brinkman,
1931), while the relevant values of the transition probabilities are taken from the
ref (Sriramachandran et al., 2013). Applying the summation rule of the molecular
spectra, band intensities with the same initial vibrational state were summed. These
sums were considered for each n’ as a function of the vibrational energy. The sum
of the band intensities divided by ν4 in each n” progression is proportional to the
Boltzmann factor for the vibrational levels. Plotting these sums against the energy
of vibrational levels enables temperature determination, see the inset in the Fig.6.30.
The temperature determined from the Boltzmann plot was crosschecked with the
synthetic spectra, generated using the program from (Parigger et al., 2015). The
earliest delay for which the temperature could be measured is 0.5 µs, and for the
pure Al T =5200 K.
Figure 6.31: Recorded and calculated AlO band head at delay of 10 µs from the
laser pulse
These temperature values indicate very special properties of the underwater
plasma, where high electron number density evident from the continuum emission
and very broad profiles of the few lines is at the same time obtained at relatively
low temperature ( 0.5 eV). As already mentioned in the Sect. 6.10, temperature
determination with the use of molecular AlO band was possible up to delay of 10
µs. Example of the recorded and generated band head spectrum on the pure Al
target is shown in the Fig. 6.31. The complete evolution of the temperatures de-
termined from the vibrational AlO band on the pure Al target is given in the Fig.
88
6.4 Optical emission spectroscopy results
6.32. It can be concluded from there that the temperature decrease is rather slow
after initial fast decay and that the bubble indeed holds the temperatures above
1000 K throughout whole period of its first oscillation as already predicted by mod-
eling(Casavola et al., 2005). This also justifies the use of relatively large gate widths
of 500 ns used for the spectra recording. Thus, the main contribution to the mea-
surement uncertainty is expected to come from any spatial inhomogeneity (spatially
averaged measurements) and possible self–absorption, especially in the band head.
Since the spectrally resolved measurements doesn’t give any clue about the origin
of the late emission detected from within the bubble on the surface of the pure Al
target, reported in the Fig. 6.8, information about the temperature values can help
in narrow the search for the emission mechanisms responsible for the late emission.
-2 0 2 4 6 8 10 12
2500
3000
3500
4000
4500
5000
5500
6000  Measured 
 Polynomial fit
T
em
pe
ra
tu
re
 
(K
)
Delay ( m s)
 
 
Figure 6.32: Temporal evolution of the vibrational temperatures determined from
the AlO band head on the pure aluminium target
Although the contribution of blackbody radiation for the temperature of 1000K
below 850 nm (detection limit of the employed camera) is just few percents, and also
the quantum efficiency of the photo cathode in the iCCD is steeply decreasing in this
wavelength region, the contribution of the blackbody emission cannot be completely
excluded. Moreover, since during the bubble evolution temperatures inside bubble
can be somewhat higher than 1000K, the contribution from the blackbody emission
would increase and move towards shorter wavelengths for which the quantum effi-
ciency is higher. Other possible explanations include the weak emission form Na
impurity, which has low temperature of evaporation (∼1156.090 K) and low lying
excited level (16973, 16956 cm−1) and is the longest observable line in the spectrum,
up to 20 µs. For deeper understanding of the nature of these glowing centres, further
studies are required.
Besides the molecular AlO emission, in the secondary plasma phase on the pure
89
6.4 Optical emission spectroscopy results
Figure 6.33: Spectra obtained with SP LIBS on aluminium target underwater with
160 accumulations: a) delay 0.5 µs gate width 0.5 µs b) delay 2 µs gate width 0.5 µs
c) delay 5 µs gate width 100 µs.
90
6.4 Optical emission spectroscopy results
Al target, lines from the main sample constituent Al I 3s23p–2p63p (2P0–2S) and
impurity sodium doublet 2p63s–2p63p (2S–2P0) were detected. Their appearance at
different delays after the laser pulse is shown in the Fig. 6.33. It can be seen that
continuum emission is almost negligible for these later delays. Another interesting
point is that by increasing the acquisition delay to 2µs increase in the analytical
signal is observed. Comparing the spectra obtained at delay of 0.5 µs and 2 µs, in
the second case there is a significant improvement in signal strengths and quality of
the lines. By setting the gate width to 100 µs at delay of 5 µs clearly resolved spectra
with further improvement in signal to noise ratio are obtained 6.33c. The longest
lasting emission is detected from Na atoms, probably due to the lowest energy of
excited level. Weak Na emission was detected even at delays of 20 µs. This result is
quite comparable with emission duration in the gas surrounding(Musazzi and Perini,
2014)(Cremers and Radziemski, 2006)(Noll, 2012)
Figure 6.34: Intensity and SNR evolution of different excited species during SP
LIBS on pure Al target underwater: a and b) gate width 0.5 µs; c and d) gate width
100 µs.
Evolution of the peak line intensities after the background subtraction, and the
corresponding signal-to-noise ratios (SNR) are depicted in Fig. 6.34. The delay
times before 0.5 µs are omitted here due to a preponderant continuum contribution.
Although the peak emission intensity occurs at different delays for different lines,
for all of them the most intense emission is delayed in respect to the laser pulse for
91
6.4 Optical emission spectroscopy results
1–2 µs. Late increase in the signal intensity is particularly noticeable for molecular
AlO species. This is the consequence of the known fact that the optimal emission of
AlO species is in the temperature range of 3500–4000 K.(Boumans, 1995) At higher
temperatures AlO radicals dissociate into atoms, and at lower temperatures they
form polyatomic aggregates.
Figure 6.35: SP LIBS spectra from impurities in submerged alumina target, obtained
with 160 accumulations: (a) delay: 0.5 µs, gate width: 0.5 µs; (b)delay 2 µs, gate
width 0.5 µs; (c) delay 5 µs, gate width 100 µs.
The signal quality in terms of SNR importantly increases for longer delays, see
Fig.6.34b. Peak emission intensity for molecular bands is at 2µs delay, while the
SNR of atomic lines increases up to delay of 4 µs. When the acquisition gate width is
92
6.4 Optical emission spectroscopy results
further increased to 100 µs for secondary plasma recordings, significantly higher SNR
are obtained compared to the detection of the early plasma with short acquisition
gates often used in the SP LIBS experiments(De Giacomo et al., 2005)(Matsumoto
et al., 2013a)(Suzuki et al., 2002). Similar situation is also obtained for the emission
from the impurities in ceramic alumina target, see Fig.6.35. In both cases, the lines
are well resolved, intense and almost free of the continuum component. Judging by
the FWHM (full width at half maximum) of the lines in the secondary plasma, elec-
tron number density (Ne) values are much lower. Determination of the Ne from these
lines was not possible since the instrumental width of the employed spectrometer-
iCCD combination was too large (∼0.195nm). Secondary plasma phase is without
any abrupt changes in the pressure, temperature, shape and intensive material ex-
pulsion. Unique property of the ablation in liquids, cavitation bubble, provides
plasma-friendly environment which extends the plasma emission duration, similar
as in DP LIBS experiments (Cristoforetti et al., 2012)(Lazic et al., 2013a)
Figure 6.36: Examples of long lasting emission from low lying states of alkali ele-
ments on alumina target: black line delay 10 µs, blue line delay 50 µs.
At the beginning of its oscillation cycle, bubble dimensions are small and it ef-
ficiently prevents the exchange of the heat with the surrounding liquid. Plasma is
also spatially restricted inside the bubble and remains well localized above the tar-
get, which makes possible detection of the LIBS signal during tens of microseconds.
When the bubble growth becomes significant plasma no longer holds high temper-
ature and pressure and the emission from the excited states disappears. In the case
of pure Al, see Fig.6.8, position of the plasma inside the bubble is changed after tens
of microseconds, leading to less efficient focusing to the spectrometer slit and hence
loss of the signal. For the ceramic alumina target, however, the emission from the
Na I doublet was visible even at delay of 50 µs, see Fig. 6.36. Further increase of
93
6.5 Influence of the target material on the laser induced breakdown in liquid
medium
the optical emission duration is material related property and it will be discussed in
the next section.
6.5 Influence of the target material on the laser
induced breakdown in liquid medium
Appearance of the plasma at different delays after the laser pulse is shown in Fig.
6.37 for the case of metallic pure Al and ceramic alumina target. The size of the
plasmas are approximately the same, but with very different shapes. In the initial
phase, the plasma from the alumina target seems more homogeneous and with a
more regular shape in comparison to aluminium. This can be attributed to a smaller
quantity of ablated material in the case of the ceramic target, which has a very
high hardness and a higher melting point compared to the metal (see Table 2 in
(Gavrilovic´ et al., 2017)). Backward motion is observed on both targets at 0.1 µs
after initial expansion and detachment from the target at delay of 0.06 µs The
surface of the target is still in a high temperature, melted state, and is further
heated by the backward plasma motion, which lead to the evaporation and new
plasma phase. Due to the large thermal conductivity of metallic Al, deposited energy
spreads effectively from the interaction area, producing relatively large volume of
the secondary plasma. Plasma starts to obtain flat shape, with dimension along
the target significantly larger than dimension normal to the target. In contrast, on
alumina the new evaporation phase initially affects only a narrow region around the
laser spot while the position of the maximum intensity of the plasma region remains
well above the target surface.
Right after the laser impacts the target, plasma on the alumina has higher inten-
sity which decays more slowly compared to the plasma on pure aluminium. But on
aluminium, intensity enhancement occurs at ∼0.5 µs, see Fig.6.5 due to efficient and
extended secondary plasma formation, connected to the material’s high conductivity,
low melting/evaporation temperature (933/2753 K(Totten and MacKenzie, 2003))
and to certain extent higher ablation rate. On alumina, on the other hand intensity
enhancement is very small due to the narrower thermally affected region and higher
melting/evaporation temperatures (2327/3250 K(Garcia-Giron et al., 2016)(Samant
and Dahotre, 2009)). The influence of these difference in the plasma intensity are
visible in the OES of the primary plasma also, see Fig. 6.38.
94
6.5 Influence of the target material on the laser induced breakdown in liquid
medium
Figure 6.37: Plasma evolution on different target materials placed on the bottom,
left column – pure aluminium , right column –alumina target
95
6.5 Influence of the target material on the laser induced breakdown in liquid
medium
Figure 6.38: LIBS emission from Al I lines recorded at short acquisition delays on
the two target materials
Although the formation of the secondary plasma on alumina is less efficient than
on the pure Al, on a larger timescale, the overall emission intensity above alumina
target drops to an almost constant value after ∼30µs and than persist up to 650 µs,
longer than the emission on the pure Al which has slower decay but diminishes with
the first bubble collapse around 475 µs. Emission on alumina survives the bubble
collapse, and even gain in the intensity in the moments that correspond to the bub-
ble collapse, which on the alumina happens at 350 µs. This intensity enhancement
is visible at delay 380 µs in the Fig 6.9 The reasons for this extremely persistent
emission on alumina can be various and some of them are listed below:
– Low thermal conductivity of alumina in comparison to aluminium, previously
mentioned in the context of the less efficient secondary plasma formation, might
have beneficial effect in this phase, since it helps maintaining the high temperature
of the vapour inside the bubble, where heat exchange with the surrounding water
(thermal conductivity ∼ 0.6 W m-1K−1 only) is very slow;
– The temperature of the melted alumina layers created by the laser pulse and to
a lesser extent by the backward propagating plasma decays slowly: a high tem-
perature state for the melted alumina can persist for several seconds (Zhang and
Modest, 1998)(Bityukov et al., 2008) (Bityukov and Petrov, 2013)(Vadim A Petrov,
2007) accompanied by optical emission in the visible spectral range.(Bityukov et
al., 2008) This sort of temperature plateau during the cooling of alumina has also
been observed for other materials,(Sola et al., 1997) with a duration proportional
to the depth of the modified layer. If melted layer several microns thick is assumed
and the scaling of the solidification time is performed according to the results of
(Bityukov et al., 2008)(Geiger et al., 1996), the expected time for duration of the
high temperature state on the alumina surface matches to the timescales of the
plasma persistence
96
6.5 Influence of the target material on the laser induced breakdown in liquid
medium
– Alkali impurities present in alumina(Na, Li and K) have low energies of excited
states, thus contribution to the overall signal.
Figure 6.39: Images of the shock wave on alumina obtained using the Schlieren
technique, at different delays D from the laser pulse
Differences of the optical emitting regions at late delays can be seen by comparing
Fig.6.8 and Fig. 6.9 On the alumina target up to delay of 20µs plasma occupies
almost entire bubble volume, while at later delays the strongest emission is coming
from the thin layer in the contact with the target. The luminous central spot can
be attributed both to the optical effects of the bubble described in Sect.6.2.2 and
to the position of the laser induced crater, where the temperature is the highest.
In contrast, on aluminium target, optical emission is spread throughout the bubble
volume and many luminous centres are visible. In the fig. 6.39 similar time evolution
of the shock wave obtained for the ceramic α-alumina target is shown. Initial speed
is just slightly lower, 2.9km/s, while the release of the second shock wave is occurring
much earlier, about 350 µs after the laser pulse.
Figure 6.40: Bubble evolution on different target materials: top row – pure alu-
minium, bottom row – alumina target
Comparison of the bubble evolution on the two targets is given in the Figs.6.406.41.
The radius of the bubble is larger on aluminium (3.2 mm) than on alumina (2.3 mm),
97
6.5 Influence of the target material on the laser induced breakdown in liquid
medium
where the bubble’s collapse time is also longer. Using the equations 6.6 and 6.5 as in
ref.(Gavrilovic´ et al., 2016) ,(Cristoforetti et al., 2012)(Sasaki et al., 2009)(Thornton
et al., 2013)(Franc and Michel, 2004) forthe energy stored inside the vapour cavity
after the ablation of alumina is estimated to be 3.1 mJ, about three times lower
than on aluminium under the same experimental conditions.
0 2 0 0 4 0 0 6 0 0 8 0 0
0
1
2
3
4  p u r e  A l
 a l u m i n a
Size
 (m
m)
D e l a y   ( m s )
Figure 6.41: Bubble size on different materials determined from fast shadowgraphy
measurements
In Fig. 6.42 the Al I, AlO and Na I lines produced by laser ablation of both
targets, detected at delays of 0.5 µs, 2 µs and 5 µs, are shown comparatively. The
significant difference between the Na I doublet intensities from the two targets is due
to the larger quantity of material impurities in alumina compared to the aluminium
sample. Due to a relatively weak continuum component in the secondary plasma
on alumina the considered transitions are of a good quality already at a delay of
0.5 µs, while for the pure Al the lines are well resolved at delays of 2 µs or longer.
On the other hand, since the secondary plasma formation is more efficient on the
metallic target than on the ceramic one, at longer delays the Al I lines and the AlO
band become more intense on aluminium than on alumina. Using a gate width of
0.5 µs’see Fig. 6.42b,c, on aluminium the maximum Al I intensity and SNR were
obtained at a delay of 2 µs. In contrast, on the alumina target the Al I intensity and
SNR are the highest at a delay of 0.5 µs. For both targets, the emission from Na I
decays slowly starting from 2-3 µs of delay and reaches its maximum SNR later than
the other transitions. Delayed appearance of the emission from the alkali elements
could also be due to the time needed for the plasma to fully expand and atoms of
alkali elements to be immersed inside the plasma(De Giacomo et al., 2004).
Important conclusion is that by increasing the acquisition gate width up to 100
µs and detecting the LIBS spectra only from the secondary plasma, both the line
98
6.5 Influence of the target material on the laser induced breakdown in liquid
medium
intensities and their SNR values show a large increase irrespective of the target
material.
Figure 6.42: SP LIBS spectra from α-alumina (blue) and pure Al target (black)
underwater obtained with 160 accumulations: (a) delay 0.5 µs, gate width 0.5 µs, (b)
delay 2 µs, gate width: 0.5 µs and (c) delay: 5 µs, width 100 µs
99
Chapter 7
Conclusion
The mutual influences of the cavitation bubble and plasma emission, created by
the ablation with single pulse from the commercial nanosecond laser source, are
studied here. The presented study is primarily focused on different mechanical and
optical phenomena of SP laser induced breakdown (LIB) on the metallic aluminium
and ceramic alumina target in water, with special attention on the late stage of
the plasma evolution. Several experimental techniques including fast photography,
Schlieren, shadowgraphy, probe beam techniques and optical emission spectroscopy
are used to give insight in the plasma formation and evolution, shock wave propa-
gation, cavitation bubble dynamics, and plasma spectral emission.
At first, differences between the plasmas formed in the liquid and the gaseous sur-
rounding were investigated experimentally using plasma photography to track the
spatial and temporal evolution of the plasma formed on the ceramic α–alumina tar-
get. From the series of the photographs, strong influence of the confining liquid
media on the size and the propagation speed of the plasma is clearly observable.
In contrast to the commonly accepted opinion about the short lived plasmas inside
liquid, when comparative studies of the plasmas in air and liquid were made, the
emission from the plasma in liquid, although much weaker, appeared also up to de-
lay of 10µs. Since this was an unusual and rarely reported result, the investigation
continued in this direction, and very long lasting optical emissions comparable to
bubble collapse time ( few hundreds of µs) were detected both for the pure, metallic
aluminium and ceramic alumina target in underwater SP laser ablation.
Besides long duration, plasma obtained in this study shows another particular fea-
ture. Namely, plasma evolves in the two characteristic phases. After the initial
plasma formation by the laser pulse, followed by its detachment from the target and
100
collapse to a small volume, the plasma grows back to the target, and approximately
after the 500 ns, the formation of the secondary plasma is observed. On the ther-
mally conductive aluminium, the backward propagating plasma finds an enlarged
heated area, which together with the low vaporization temperature of the material
leads to efficient secondary target evaporation. The shape of the secondary plasma
is initially flattened on the sample surface and its overall emission increases during
the first 2 µs. Low thermal conductivity of alumina on the other hand, keeps the
backward plasma-target interaction in proximity of the laser induced crater, where
a high temperature in the surface layers is present, and also prevents important
heat losses of the plume through the target. Simultaneously, a high vaporization
temperature of the material limits the emission of the sample material into the
vapour bubble leading to a less efficient secondary plasma formation compared to
the metallic target. The backward interacting plume on alumina is slightly elliptical
and inside the expanded vapour bubble it is reduced to a glowing point close to
the hot crater. The transition between the primary and secondary plasma phases
is evidenced by observing the photographs, and even more clearly, after performing
some basic manipulation over the image matrix, for which the programing routines
were developed.
Further, the mutual relations of the plasma and the bubble were investigated on the
example of the two chosen target materials. It was shown that duration of the opti-
cal emission on the pure Al target coincides with the collapse of the laser produced
bubble, and it is concluded that the unique temperature and pressure conditions
inside the bubble favour this long lasting optical emission. In this late stage, plasma
on aluminium is no more compact. and lot of glowing centres were detected from the
whole bubble volume. Even though the process of the secondary plasma formation
is less efficient on ceramic alumina target, the total duration of emission exceeds the
lifetime of the bubble, and even certain increase in emission intensity is detected
after the bubble collapse. The long lasting optical emission remains well confined
inside the bubble, but without visible particle clustering detected in the case of
metalic aluminium. Dedicated comparison between all the studied phenomena on
the two target materials is given in the separate section, and is shown graphically
below, Fig. 7.1
In addition to the studies of bubble influence on the duration of plasma emis-
sion, bubble influence on the signal collection efficiency is also investigated. Due to
the lowering of its index of refraction with the expansion, bubble enhances Snell’s
101
Figure 7.1: Graphical comparison of the timescale of LIB processes on two different
target materials
reflections at its surface which alters significantly the propagation of light rays. This
influence in some cases can even be positive, since it enables detection of the outer
part of the plasma emission otherwise lost for the detection, by forming a bright
central spot. Multiple reflections and refractions inside the bubble are used here to
derive the expression for the calculation of the refractive index of the bubble, which
was found to change during the bubble evolution. It is proven that bubble alters
significantly spatial distribution of the plasma emission and this important factor
needs to be taken into account when optimizing the optics collection system. These
effects are also very important for DP LIBS and other applications where the laser
beam propagates through the laser formed bubble, like the probe beam techniques
used in this study to resolve the dynamics of the cavitation bubble. Probe beam
techniques have proven to be very convenient and cheaper comparative method,
which under the right experimental configuration can provide information about
the whole bubble dynamics from the one laser shot. Bubble sizing was performed
from the scattering and transmission measurements, and also from the fast shad-
owgraphy images with the use of the routine for extracting the bubble dimensions
from the dimensionally calibrated images. Information about the bubble size and
the bubble collapse time obtained from the second shock wave recorded with the
schlieren technique, served to calculate the energy contained in the bubble, while
the shock front propagation was used for the evaluation of the pressures developed
in the interaction area.
Optical emission spectroscopy analysis is also divided in the two sections which cor-
respond to the two plasma phases. In the first stage, immediately after the laser
pulse and with duration of less than 500 ns, the plasma rapidly expands and decays,
where the optical emission is characterized by an intense continuum component and
very large widths of the detected lines. Here, the LIBS signal has a poor quality
102
in terms of the spectral resolution and signal-to-noise ratio. The program for the
continuum component determination and subtraction from the experimental spec-
tra was made, which calculates the temperature from the Vien’s displacement law.
Electron temperature deduced from the continuum emission was found to be in the
range 6000–7000 K. Electron number density is determined in the primary plasma
phase on alumina from the Hα line, oxygen triplet and by using the Inglis-Teller
limit, obtaining the values above 6·1018 cm−3. The plasma conditions suggest the
non-ideal plasma case, and the implications for the electron number density de-
termination are discussed in short. The LIBS signal detected from the secondary
plasma on aluminium is very intense at the beginning, but later suppressed also by
the particle formation and the heat dissipation through the target. The secondary
plasma phase starts from the backward reheated target. Its emission has a stable
position close to the target and still growing intensity over tens of microseconds
from the pulse, which enables the use of a large number of accumulations to obtain
the better quality of lines.
Figure 7.2: Sketch of the processes occurring after LIB on the Al target
The secondary plasma has significantly lower electron densities with respect to
the first plasma stage, hence it is almost free from the continuum component. Tem-
peratures in the secondary plasma phase are around 5000K, as determined from
the vibrational AlO bands on both target materials. The LIBS spectra from the
secondary plasma contain the lines from low excited levels and have a good quality,
with very narrow transitions almost free of the continuum component. The inten-
sities of the emission lines decay below the detection limit at delays in order of 10
µs, whereas Na I doublet remains visible even after 50 µs from the laser pulse on
alumina target. Sketch of the processes induced after LIB on a metallic aluminium
target 7.2, where the bubble, plasma emission and particles inside the bubble are
all present at the same time, serves to illustrate the complexity of the phenomena
103
that needs to be studied in order to understand the SP LA in the liquid.
In summary, a good quality LIBS spectra after underwater laser ablation with ns
pulses from a commercial laser source can be obtained by detecting only the de-
layed, long lasting plasma and avoiding the initial explosive phase, characterized by
high electron density, intense continuum and a very few, strongly broadened emis-
sion lines. With the proposed experimental approach, LIBS detection of underwater
plasma is feasible both on metallic and ceramic samples, also by using a not gated
and a relatively cheap CCD or camera. In the process of the sustaining and detect-
ing the plasma emission the role of the cavitation bubble proved to be extremely
important.
104
Bibliography
Abraham, E., Minoshima, K. and Matsumoto, H. (2000), Femtosecond laser-induced
breakdown in water: time-resolved shadow imaging and two-color interferometric
imaging, Optics Communications 176, 441–452. 20
Adamson, M., Padmanabhan, A., Godfrey, G. and Rehse, S. (2007), Laser-induced
breakdown spectroscopy at a water/gas interface: A study of bath gas-dependent
molecular species, Spectrochim. Acta Part B 62 62, 1348–1360. 82
Akhatov, I., Lindau, O., Topolnikov, A., Mettin, R., Vakhitova, N. and Lauterborn,
W. (2001), Collapse and rebound of a laser-induced cavitation bubble, Phys. Flu-
ids 13, 2805–2819. 21
Al-Mamun, S. A., Nakajima, R. and Ishigaki, T. (2013), Tuning the size of alu-
minum oxide nanoparticles synthesized by laser ablation in water using physical
and chemical approaches, J. Coll. Int. Sci. 392, 172–182. 14
Alexiou, S. (2005), Stark broadening of hydrogen lines in dense plasmas: Analysis
of recent experiments, Phys. Rev. E 71, 066403, 11pp. 43
Alexiou, S. and Leboucher-Dalimier, E. (1999), Hydrogen balmer-α broadening in
dense plasmas, Phys. Rev. E 60, 3436–3438. 43
Alloncle, A., Dufresne, D. and Autric, M. (1994), Pressure waves induced in liquid
under high laser irradiation, J. de Physique IV 4, 131–134. 19, 20
Amendola, V. and Meneghetti, M. (2013), What controls the composition and the
structure of nanomaterials generated by laser ablation in liquid solution?, Phys.
Chem. Chem. Phys. 15, 3027–3046. 5, 13
Bai, X., Ma, Q., Perrier, M., Motto-Ros, V., Sabourdy, D., Nguyen, L., Jalocha, A.
and Yu, J. (2013), Experimental study of laser-induced plasma: Influence of laser
fluence and pulse duration, Spectrochim. Acta Part B 87, 27–35. 3
105
BIBLIOGRAPHY
Bang, J. H. and Suslick, K. S. (2010), Applications of ultrasound to the synthesis
of nanostructured materials, Adv. Mater. 22, 1039–1059. 3
Barber, B. P. and Putterman, S. J. (1992), Light scattering measurements of the
repetitive supersonic implosion of a sonoluminescing bubble, Phys. Rev. Lett.
69, 3839–3842. 29, 31, 77
Barcikowski, S. and Compagnini, G. (2013), Advanced nanoparticle generation and
excitation by lasers in liquids, Phys.Chem. Chem. Phys. 15, 3022–3026. 13
Barcikowski, S., Devesa, F. and Moldenhaueri, K. (2009), Impact and structure of
literature on nanoparticle generation by laser ablation in liquids, J. Nanopart.
Res. 11, 1883–1893. 13
Barnes, P. A. (1969), Studies of laser induced breakdown phenomena in liquid water,
Simon Fraser University,thesis. 32
Beddows, D., Samek, O., Liˇska, M. and Telle, H. (2002), Single-pulse laser-induced
breakdown spectroscopy of samples submerged in water using a single-fibre light
delivery system, Spectrochim. Acta Part B 57, 1461–1471. 15
Benjamin, T. B. and Ellis, A. T. (1966), The collapse of cavitation bubbles and
the pressures thereby produced against solid boundaries, Philos. Trans. R. Soc.
London, Ser. A 260, 221–240. 28
Bityukov, V. K. and Petrov, V. A. (2013), Absorption coefficient of molten aluminum
oxide in semitransparent spectral range, Appl. Phys. Research 5, 51–71. 96
Bityukov, V. K., Vorobev, A. Y., Petrov, V. A. and Titov, V. E. (2008), The radi-
ation of alumina melt in the visible and near ir spectra upon solidification in the
process of cooling in ambient air, High Temperature 46, 782–794. 96
Bo¨ddeker, S., Gu¨nter, S., Ko¨nies, A., Hitzschke, L. and Kunze, H.-J. (1993), Shift
and width of the hα line of hydrogen in dense plasmas, Phys. Rev. E 47, 2785–
2791. 43
Bongiovanni, C., Chevaillier, J. P. and Fabre, J. (1997), Sizing of bubbles by inco-
herent imaging: defocus bias, Exp. in Fluids 23, 209–216. 65, 68
Bongiovanni, C., Dominguez, A. and Chevaillier, J.-P. (2000), Understanding images
of bubbles, Eur. J. Phys. 21, 561–570. 65
106
BIBLIOGRAPHY
Boumans, P. W. J. M. (1995), Theory of Spectrochemical Excitation, Springer US.
92
Brennen, C. E. (1995), Cavitation and Bubble Dynamics, Oxford University
Press,Inc. 20, 22
Brenner, M. P., Hilgenfeldt, S. and Lohse, D. (2002), Single-bubble sonolumines-
cence, Rev. Mod. Phys. 74, 425–484. 3, 21, 26
Brujan, E.-A., Nahen, K., Schmidt, P. and Vogel, A. (2001), Dynamics of laser-
induced cavitation bubbles near an elastic boundary, J. Fluid Mech. 433, 251–281.
28
Bu¨scher, S., Wrubel, T., Ferri, S. and Kunze, H.-J. (2002), The stark width and
shift of the hydrogen hα line, J. Phys. B: At. Mol. Opt. Phys. 35, 2889–2897. 43
Calisti, A., Bureyeva, L., Lisitsa, V., Shuvaev, D. and Talin, B. (2007), Coupling
and ionization effects on hydrogen spectral line shapes in dense plasmas, Eur.
Phys. J. D 42, 387–392. 43
Casavola, A., De Giacomo, A., DellAglio, M., Taccogna, F., Colonna, G., De Pascale,
O. and Longo, S. (2005), Experimental investigation and modelling of double
pulse laser induced plasma spectroscopy under water, Spectrochim. Acta Part B
60, 975–985. 26, 63, 89
Chen, J., Li, X., Gua, Y., Wang, H., Song, X. and Zeng, H. (2017), Probing meso-
scopic process of laser ablation in liquid by integrated method of optical beam
deflection and time-resolved shadowgraphy, J. Coll. Int. Sci. 489, 38–46. 13, 21,
38, 71, 73
Chen, X., Xu, R.-Q., Chen, J.-P., Shen, Z.-H., Jian, L. and Ni, X.-W. (2004), Shock-
wave propagation and cavitation bubble oscillation by nd:yag laser ablation of a
metal in water, Appl. Opt. 43, 3251–3257. 71, 73
Chu, H.-C., Vo, S. and Williams, G. A. (2009), Precursor luminescence near the
collapse of laser-induced bubbles in alkali-salt solutions, PRL 102, 204301, 4pp.
3
Cirisan, M., Cvejic´, M., Gavrilovic´, M., Jovic´evic´, S., Konjevic´, N. and J.Hermann
(2014), Stark broadening measurement of al ii lines in a laser-induced plasma, J.
Quant. Spectrosc. Radiat. Transfer. 133, 652–662. 4
107
BIBLIOGRAPHY
Cremers, D. A. and Radziemski, L. J. (2006), Handbook of Laser-Induced Breakdown
Spectroscopy, John Wiley & Sons Ltd. 4, 40, 91
Cristoforetti, G., Tibberi, M., Simonelli, A., Marsili, P. and Giammanco, F. (2012),
Toward the optimization of double-pulse libs underwater: effects of experimen-
tal parameters on the reproducibility and dynamics of laser-induced cavitation
bubble, Appl. Opt. 51, B30–B41. 5, 21, 26, 27, 45, 70, 93, 98
Crum, L. A., J., T., Reisse, M. J. L. and Suslick, K. S. (1999), Sonochemistry and
Sonoluminescence, Springer Science+Business Media Dordrecht. 29
Cvejic´, M., Gavrilovic´, M., Jovic´evic´, S. and Konjevic´, N. (2013), Stark broadening
of mg i and mg ii spectral lines and debye shielding effect in laser induced plasma,
Spectrochim. Acta Part B 85, 20–33. 4, 86
Cvejic´, M., Stambulchik, E., Gavrilovic´, M., Jovic´evic´, S. and Konjevic´, N. (2014),
Neutral lithium spectral line 460.28 nm with forbidden component for low tem-
perature plasma diagnostics of laser-induced plasma, Spectrochim. Acta Part B
100, 86–97. 4
De Giacomo, A., De Bonis, A., DellAglio, M., De Pascale, O., Gaudiuso, R., Or-
lando, S., Santagata, A., Senesi, G. S., Taccogna, F. and Teghil, R. (2011), Laser
ablation of graphite in water in a range of pressure from 1 to 146 atm using single
and double pulse techniques for the production of carbon nanostructures, J. Phys.
Chem. C 115, 5123–5130. 41
De Giacomo, A., DellAglio, M., Casavola, A., Colonna, G., De Pascale, O. and
Capitelli, M. (2006), Elemental chemical analysis of submerged targets by double-
pulse laser-induced breakdown spectroscopy, Anal. Bioanal. Chem. 385, 303–311.
15
De Giacomo, A., DellAglio, M., Colao, F. and Fantoni, R. (2004), Double pulse
laser produced plasma on metallic target in seawater: basic aspects and analytical
approach, Spectrochim. Acta Part B 59, 1431–1438. 21, 98
De Giacomo, A., DellAglio, M., Colao, F., Fantoni, R. and Lazic, V. (2005),
Double-pulse libs in bulk water and on submerged bronze samples, Appl. Surf.
Sci 247, 157–162. 15, 41, 93
108
BIBLIOGRAPHY
De Giacomo, A., Dell’Aglio, M., Gaudiuso, R., Amoruso, S. and Pascale, O. D.
(2012), Effects of the background environment on formation, evolution and emis-
sion spectra of laser-induced plasmas, Spectrochim. Acta, Part B 78, 1–19. 3, 39,
41
De Giacomo, A., Dell’Aglio, M., O.De Pascale and Capitelli, M. (2007), From single
pulse to double pulse ns-laser induced breakdown spectroscopy under water: El-
emental analysis of aqueous solutions and submerged solid samples, Spectrochim.
Acta, Part B 62, 721–738. 5, 6, 27, 38, 41, 44, 55, 83
De Giacomo, A., DellAglio, M., Santagata, A., Gaudiuso, R., De Pascale, O., Wa-
gener, P., Messina, G. C., Compagnini, G. and Barcikowski, S. (2013), Cavita-
tion dynamics of laser ablation of bulk and wire-shaped metals in water during
nanoparticles production, Phys.Chem.Chem.Phys. 13, 3083–3092. 14, 27, 28, 60,
61
De Giacomo, A., Gaudiuso, R., Dell’Aglio, M. and Santagata, A. (2010), The role
of continuum radiation in laser induced plasma spectroscopy, Spectrochim. Acta
Part B 65, 385–394. 41, 80
Dean, C. E. and Marston, P. L. (1991), Critical angle light scattering from bubbles:
an asymptotic series approximation, Appl. Opt. 30, 4764–4776. 31
Dehaeck, S. and van Beeck, J. P. A. J. (2007), Simultaneous determination of bubble
diameter and relative refractive index using glare circles, Appl. Opt. 46, 5957–
5963. 65, 66
Delale (Ed.), C. (2013), Bubble Dynamics & Shock Waves, SHOCKWAVES 8,
Springer-Verlag Berlin Heidelberg. 3, 19, 78
DellAglio, M., Gaudiuso, R., De Pascale, O. and De Giacomo, A. (2015), Mecha-
nisms and processes of pulsed laser ablation in liquids duringnanoparticle produc-
tion, Applied Surface Science 348, 4–9. 5, 55
Descoeudres, A. (2006), characterization of electrical discharge machining plasmas,
E´COLE POLYTECHNIQUE FE´DE´RALE DE LAUSANNE. 82, 85
Descoeudres, A., Hollenstein, C., Wa¨lder, G., Demellayer, R. and Perez, R. (2008),
Time- and spatially-resolved characterization of electrical discharge machining
plasma, Plasma Sources Sci. Technol. 17, 024008, 10pp. 82
109
BIBLIOGRAPHY
Descoeudres, A., Hollenstein, C., Wa¨lder, G. and Perez, R. (2005), Time-resolved
imaging and spatially-resolved spectroscopy of electrical discharge machining
plasma, J. Phys. D: Appl. Phys. 38, 4066–4073. 82
Detalle, V., Sabsabi, M., St-Onge, L., Hamel, A. and He´on, R. (2003), Influence
of er:yag and nd:yag wavelengths on laser-induced breakdown spectroscopy mea-
surements under air or helium atmosphere, Appl. Opt. 42, 5971–5977. 3
Diaci, J. and Mozˇina, J. (1995), Multiple-pass laser beam deflection probe for detec-
tion of acoustic and weak shock waves in fluids, Rev. Sci. Instrum. 66, 4644–4648.
70
Docchio, F., Regondi, P., Capon, M. R. C. and Mellerio, J. (1988), Study of the
temporal and spatial dynamics of plasmas induced in liquids by nanosecond nd:yag
laser pulses. 1: Analysis of the plasma starting times, Appl. Opt. 27, 3661–3668.
32
Docchio, F. and Sacchi, C. A. (1988), Shielding properties of laser-induced plasmas
in ocular media irradiated by single nd:yag pulses of different durations, Inves-
tigative Ophthalmology & Visual Science 29, 437–443. 12
Docchio, F., Sacchi, C. A. and Marshall, J. (1986), Experimental investigation of
optical breakdown thresholds in ocular media under single pulse irradiation with
different pulse durations, Lasers Ophthalmol. 1, 83–93. 36
Duocastella, M., Ferna´ndez-Pradas, J. M., Morenza, J. L. and Serra, P. (2009),
Time-resolved imaging of the laser forward transfer of liquids, J. Appl. Phys.
106, 1084907, 7pp. 15
D’yachkov, L. (1998), Approximation for the probabilities of the realization of atomic
bound states in a plasma, J. Quant, Spectrosc. Radiat. Transfer 59, 65–69. 86
Ellis, A. T. (1953), Observation on cavitation bubble collapse, Thesis, California
Institute of Technology. 28
Escarguel, A., Ferhat, B., Lesage, A. and Richou, J. (2000), A single laser spark in
aqueous medium, J. Quant. Spectrosc. Radiat. Transfer. 64, 353–361. 42, 43, 82
Escarguel, A., Oks, E., Richou, J. and Volodko, D. (2000), Highly nonlinear, sign-
varying shift of hydrogen spectral lines in dense plasmas, Phys. Rev. E 62, 2667–
2671. 43
110
BIBLIOGRAPHY
Feng, Q., Moloney, J. V., Newell, A. C., Wright, E. M., Cook, K., Kennedy, P. K.,
Hammer, D. X., Rockwell, B. and Thompson, C. R. (1997), Theory and simulation
on the threshold of water breakdown induced by focused ultrashort laser pulses,
IEEE J. Quantum Electron 33, 127–137. 34
Fortes, F. J., Guirado, S., Metzinger, A. and Laserna, J. J. (2015), A study of
underwater stand-off laser-induced breakdown spectroscopy for chemical analysis
of objects in the deep ocean, J. Anal. At. Spectrom. 30, 1050–1056. 15
Fortes, F. J., Moros, J., Lucena, P., Cabal´ın, L. M., and Laserna, J. J. (2013),
Laser-induced breakdown spectroscopy, Appl. Spectroc. 85, 640–669. 4, 15
Fortov, V. E. and Iakubov, I. T. (2000), The Physics of Non-Ideal Plasma, World
Scientific Publishing Co. Pte. Ltd. 42
Franc, J.-P. and Michel, J.-M. (2004), Fundamentals of Cavitation, Kluwer Academic
Publishers. 20, 22, 70, 98
Gaitan, D. F. and Holt, R. G. (1999), Experimental observations of bubble response
and light intensity near the threshold for single bubble sonoluminescence in an
air-water system, Phys. Rev. E 59, 5495–5502. 3
Garcia-Giron, A., Sola, D. and Pen˜a, J. (2016), Liquid-assisted laser ablation of
advanced ceramics and glass-ceramic materials, Appl. Surf. Sci. 363, 548–554.
14, 94
Gaudiuso, R., DellAglio, M., De Pascale, O., Senesi, G. S. and De Giacomo, A.
(2010), Laser induced breakdown spectroscopy for elemental analysis in environ-
mental, cultural heritage and space applications: A review of methods and results,
Sensors 10, 7434–7468. 4
Gavrilovic´, M. R., Cvejic´, M., Lazic, V. and Jovic´evic´, S. (2016), Secondary plasma
formation after single pulse laser ablation underwater and its advantages for laser
induced breakdown spectroscopy (LIBS), Phys. Chem. Chem. Phys. 18, 14629–
14637. 40, 62, 98
Gavrilovic´, M. R., Lazic, V. and Jovic´evic´, S. (2017), Influence of the target mate-
rial on secondary plasma formation underwater and its laser induced breakdown
spectroscopy (LIBS) signal, J. Anal. At. Spectrom. 32, 345–353. 40, 62, 94
111
BIBLIOGRAPHY
Geiger, M., Becker, W., Rebhan, T., Hutfless, J. and Lutz, N. (1996), Increase of
efficiency for the xecl laser ablation of ceramics, Appl. Surf. Sci. 96–98, 309–315.
96
Gigosos, M. A., Gonza´lez, M. A. and Carden˜oso, V. (2003), Computer simulated
balmer-alpha, -beta and -gamma stark line profiles for non-equilibrium plasmas
diagnostics, Spectrochimica Acta Part B 58, 1489–1504. 85
Gladush, G. G. and Smurov, I. (2011), Physics of Laser Materials Processing,
Springer-Verlag Berlin Heidelberg. 14
Gregorcˇicˇ, P., Jezersˇek, M. and Mozˇina, J. (2012), Optodynamic energy-conversion
efficiency during an er:yag-laser-pulse delivery into a liquid through different fiber-
tip geometries, J. Biomed. Opt. 17, 075006, 9pp. 21
Gregorcˇicˇ, P., Lukacˇ, N., Mozˇina, J. and Jezersˇek, M. (2016), Synchronized delivery
of er:yag-laser pulses into water studied by a laser beam transmission probe for
enhanced endodontic treatment, Appl. Phys. A 122, 459, 7pp. 71
Gregorcˇicˇ, P., Petkovsˇek, R. and Mozˇina, J. (2007), Investigation of a cavitation
bubble between a rigid boundary and a free surface, J. Appl. Phys. 102, 094904,
8pp. 21, 71
Gregorcˇicˇ, P., Petkovsˇek, R., Mozˇina, J. and Mocˇnik, G. (2008), Measurements
of cavitation bubble dynamics based on a beam-deflection probe, Appl. Phys. A
93, 901–905. 71
Griem, H. (2000), Stark broadening of the hydrogen balmer-a line in low and high
density plasmas, Contrib. Plasma Phys. 40, 46–56. 43
Griem, H. R. (2001), Comment on highly nonlinear, sign-varying shift of hydrogen
spectral lines in dense plasmas, Phys. Rev. E 64, 058401, 2pp. 43
Griem, H. R., Halenka, J. and Olchawa, W. (2005), Comparison of hydrogen balmer-
alpha stark profiles measured at high electron densities with theoretical results,
J. Phys. B: At. Mol. Opt. Phys. 38, 975–1000. 43
Guirado, S., Fortes, F., Lazic, V. and Laserna, J. (2012), Chemical analysis of
archeological materials in submarine environments using laser-induced breakdown
spectroscopy. on-site trials in the mediterranean sea, Spectrochim. Acta Part B
74–75, 137–143. 15, 40
112
BIBLIOGRAPHY
Hahn, D. W. and Omenetto, N. (2010), Laser-induced breakdown spectroscopy
(libs), part i: Review of basic diagnostics and plasmaparticle interactions:still-
challenging issues within the analytical plasma community, Appl. Spectroc.
64, 335A–366A. 4
Hahn, D. W. and Omenetto, N. (2012), Laser-induced breakdown spectroscopy
(libs), part ii: Review of instrumental and methodological approaches to ma-
terial analysis and applications to different fields, Appl. Spectroc. 66, 347–419.
4
Halenka, J. (2004), Comment on highly nonlinear, sign-varying shift of hydrogen
spectral lines in dense plasmas, Phys. Rev. E 69, 028401, 3pp. 43
Hammer, D. X., Jansen, E. D., Frenz, M., Noojin, G. D., Thomas, R. J., Noack,
J., Vogel, A., Rockwell, B. A. and Welch, A. J. (1997), Shielding properties of
laser-induced breakdown in water for pulse durations from 5 ns to 125 fs, Appl.
Opt. 36, 5630–5640. 12
Harilal, S. S., Bindhu, C. V., Tillack, M. S., Najmabadi, F., and Gaeris, A. C.
(2003), Internal structure and expansion dynamics of laser ablation plumes into
ambient gases, J. Appl. Phys. 93, 2380–2388. 4
Harilal, S. S., Bindhu, C. V., Tillack, M. S., Najmabadi, F. and Gaeris, A. C. (2002),
Plume splitting and sharpening in laser-produced aluminium plasma, J. Phys. D:
Appl. Phys. 35, 2935–2938. 4
Harilal, S. S., OShay, B., Tao, Y. and Tillack, M. S. (2006), Ambient gas effects on
the dynamics of laser-produced tin plume expansion, J. Appl. Phys. 99, 083303,
10 pp. 3
He, H., Li, W., Zhang, X., Xia, M. and Yang, K. (2012), Light scattering by a
spheroidal bubble with geometrical optics approximation, J. Quant. Spectrosc.
Radiat. Transfer. 113, 1467–1475. 31
Heisterkamp, A., Ripken, T., Mamom, T., Drommer, W., Welling, H., Ertmer, W.
and Lubatschowski, H. (2002), Nonlinear side effects of fs pulses inside corneal
tissue during photodisruption, Appl. Phys. B 74, 419–425. 12
Holt, R. G. and Crum, L. A. (1990), Mie scattering used to determine spherical
bubble oscillations, Appl. Opt. 29, 4182–4191. 30
113
BIBLIOGRAPHY
Hooper Jr, C. (1968), Low-frequency component electric microfield distribution in
plasmas, Phys. Rev. 165, 215–222. 86
Hopp, B., Smausz, T., T.Wittmann and Igna´cz, F. (2000), Comparative time-
resolved study of solid-state and liquid ablation of polyethylene-glycol 1000: tem-
perature, viscosity and surface tension dependence, Appl. Phys. A 71, 315–318.
14
Huber, P., Jo¨chle, K. and Debus, J. (1998), Influence of shock wave pressure ampli-
tude and pulse repetition frequency on the lifespan, size and number of transient
cavities in the field of an electromagnetic lithotripter, Phys. Med. Biol. 43, 3113–
3128. 26, 30
Hutson, M. S. and Ma, X. (2007), Plasma and cavitation dynamics during pulsed
laser microsurgery in vivo, Phys. Rev. Lett. 99, 158104, 4pp. 13
Hwang, Z.-W., Teng, Y.-Y., Li, K.-P. and Sneddon, J. (1991), Interaction of a
laser beam with metals. part i: Quantitative studies of plasma emission, Appl.
Spectrosc. 45, 435–441. 3
Ibrahimkutty, S., Wagener, P., dos Santos Rolo, T., Karpov, D., Menzel, A., Baum-
bach, T., Barcikowski, S. and Plech, A. (2015), A hierarchical view on mate-
rial formation during pulsed-laser synthesis of nanoparticles in liquid, Sci. Rep.
5, 16313, 11pp. 5, 13, 27, 69
Ibrahimkutty, S., Wagener, P., Menzel, A., Plech, A. and Barcikowski, S. (2012),
Nanoparticle formation in a cavitation bubble after pulsed laser ablation in liquid
studied with high time resolution small angle x-ray scattering, Appl. Phys. Lett.
101, 103104, 4pp. 21, 27
Isselin, J.-C., Alloncle, A.-P., and Autric, M. (1998), On laser induced single bubble
near a solid boundary: Contribution to the understanding of erosion phenomena,
J. Appl. Phys. 84, 5766–5771. 2
Jomni, F., Denat, A. and Aitken, F. (2009), The dynamics of microscopic bubbles
in viscous insulating liquids, J. Appl. Phys. 105, 053301, 9pp. 3
Kannatey-Asibu, E. (2009), Principles of laser materials processing, John Wiley &
Sons, Inc. 14
114
BIBLIOGRAPHY
Kennedy, P., Hammer, D. and Rockwell, B. (1997), Laser induced breakdown in
aqueous media, Prog. Quant. Electr. 21, 155–248. 12, 30, 32, 33, 44, 61, 78, 82
Kennedy, P. K. (1995), A first-order model for computation of laser-induced break-
down thresholds in ocular and aqueous media: Part i-theory, IEEE J. Quantum
Electron 31, 2241–2249. 1, 34, 35
Kennedy, P. K., Boppart, S. A., Hammer, D. X., Rockwell, B. A., Noojin, G. D. and
Roach, W. P. (1995), A first-order model for computation of laser-induced break-
down thresholds in ocular and aqueous media: Part ii-comparison to experiments,
IEEE J. Quantum Electron 31, 2250–2257. 1
Kielkopf, J. F. (1995), Spectroscopic study of laser-produced plasmas in hydrogen,
Phys. Rev. E 52, 2013–2024. 43
Kielkopf, J. F. (2000), Laser-produced plasma bubble, Phys. Rev. E 63, 016411,6pp.
5
Kielkopf, J. F. and Allard, N. F. (2014), Shift and width of the balmer series ha line
at high electron density in a laser-produced plasma, J. Phys. B: At. Mol. Opt.
Phys. 747, 155701, 6pp. 43
Kim, K. K., Roy, M., Kwon, H., Song, J. K. and Park, S. M. (2015), Laser ablation
dynamics in liquid phase: The effects of magnetic field and electrolyte, J. Appl.
Phys. 117, 074302, 7pp. 55
Koch, S. (2012), Qualitative and quantitative trace analysis on liquids and solid
samples by laser-induced breakdown spectroscopy (LIBS), Von der Fakulta¨t fr
Mathematik und Naturwissenschaften der Carl von Ossietzky Universita¨t Old-
enburg,Dissertation. 41
Koch, S., Garen, W., Hegedu¨s, F., Neu, W., Reuter, R. and Teubner, U. (2012),
Time-resolved measurements of shock-induced cavitation bubbles in liquids, Appl.
Phys. B 108, 345–351. 71
Koch, S., Garen, W., M’´uler, M. and Neu, W. (2004), Detection of chromium in
liquids by laser induced breakdown spectroscopy (libs), Appl. Phys. A 79, 1071–
1073. 15
115
BIBLIOGRAPHY
Kohsakowski, S., Go¨kce, B., Tanabe, R., Wagener, P., Plech, A., Ito, Y. and Bar-
cikowski, S. (2016), Target geometry and rigidity determines laser-induced cavi-
tation bubble transport and nanoparticle productivity a high-speed videography
study, Phys.Chem.Chem.Phys. 18, 16585–16593. 14
Konjevic´, N. (1999), Plasma broadening and shifting of non-hydrogenic spectral
lines: present status and applications, Phys. Rep. 316, 339–401. 86
Konjevic´, N., Ivkovic´, M. and Jovic´evic´, S. (2010), Spectroscopic diagnostics of
laser-induced plasmas, Spectrochim. Acta Part B 65, 593–602. 4
Konjevic´, N., Ivkovic´, M. and Sakan, N. (2012), Hydrogen balmer lines for low
electron number density plasma diagnostics, Spectrochim. Acta Part B 62 76, 16–
26. 83, 84
Koren, G. (1987), Observation of shock waves and cooling waves in the laser ablation
of kapton films in air, Appl. Phys. Lett. 51, 569–571. 70
Kou, L., Labrie, D. and Chylek, P. (1993), Refractive indices of water and ice in the
0.65– to 2.5–µm spectral range, Appl. Opt. 32, 3531–3540. 46
Kovalchuk, T., Toker, G., Bulatov, V. and Schechter, I. (2010), Laser breakdown in
alcohols and water induced by λ= 1064 nm nanosecond pulses, Chem. Phys. Lett.
500, 242–250. 3, 36
Kramida, A., Yu. Ralchenko and and NIST ASD Team, J. R. (2015), NIST Atomic
Spectra Database (ver. 5.3), [Online]. Available: http://physics.nist.gov/asd
[2017, February 28]. National Institute of Standards and Technology, Gaithers-
burg, MD. 85
Krieger, J. R. and Chahine, G. L. (2005), Acousitc signals of underwater explosions
near surfaces, J. Acoust. Soc. Am. 118, 2961–2974. 3
Krstulovic´, N., Shannon, S., Stefanuik, R. and Fanara, C. (2013), Underwater-laser
drilling of aluminum, Int. J. Adv. Manuf. Technol. 69, 1765–1773. 15
Kruusing, A. (2004a), Underwater and water-assisted laser processing: Part 1general
features, steam cleaning and shock processing, Optics and Lasers in Engineering
41, 307–327. 14
116
BIBLIOGRAPHY
Kruusing, A. (2004b), Underwater and water-assisted laser processing: Part 2etch-
ing, cutting and rarely used methods, Optics and Lasers in Engineering 41, 329–
352. 14
Kruusing, A. (2008), Handbook of Liquids-Assisted Laser Processing, Elsevier Ltd.
14
Kumar, B. and Thareja, R. K. (2010), Synthesis of nanoparticles in laser ablation
of aluminum in liquid, J. Appl. Phys. 108, 064906,6pp. 42, 82
Lam, J. (2015), Pulsed laser ablation in liquid: towards the comprehension of the
growth processes, These de l’Universite de Lyon. 27, 83
Lam, J., Amans, D., Chaput, F., Diouf, M., Ledoux, G., Mary, N., Masenelli-Varlot,
K., Motto-Ros, V. and Dujardin, C. (2014), γ-al2o3 nanoparticles synthesised by
pulsed laser ablation in liquids: a plasma analysis, Phys. Chem. Chem. Phys.
16, 963–973. 3, 42, 56, 58
Lam, J., Lombard, J., Dujardin, C., Ledoux, G., Merabia, S. and Amans, D. (2016),
Dynamical study of bubble expansion following laser ablation in liquids, AP-
PLIED PHYSICS LETTERS 108, 074104, 5pp. 13, 42
Langley, D. S. and Marston, P. L. (1991), Forward glory scattering from bubbles,
Appl. Opt. 630, 3452–3458. 29, 65
Lazic, V., Colao, F., Fantoni, R. and Spizzicchino, V. (2005), Recognition of arche-
ological materials underwater by laser induced breakdown spectroscopy, Spec-
trochim. Acta Part B 60, 1014–1024. 15
Lazic, V., Colao, F., Fantoni, R., Spizzicchino, V. and Jovic´evic´, S. (2007), Under-
water sediment analyses by laser induced breakdown spectroscopy and calibration
procedure for fluctuating plasma parameters, Spectrochim. Acta Part B 62, 30–39.
15
Lazic, V. and Jovic´evic´, S. (2014), Laser induced breakdown spectroscopy inside
liquids: Processes and analytical aspects, Spectrochim. Acta, Part B 104, 288–
311. 10, 11, 29, 40, 42, 46, 58
Lazic, V., Jovicevic, S. and Carpanese, M. (2012), Laser induced bubbles inside
liquids: Transient optical properties and effects on a beam propagation, Appl.
Phys. Lett. 101, 054101, 4pp. 29, 65, 66, 67
117
BIBLIOGRAPHY
Lazic, V., Jovievic, S., Fantoni, R. and Colao, F. (2007), Efficient plasma and bubble
generation underwater by an optimized laser excitation and its application for
liquid analyses by laser-induced breakdown spectroscopy, Spectrochim. Acta Part
B 62, 1433–1442. 26, 31
Lazic, V., Laserna, J. and Jovicevic, S. (2013a), Insights in the laser-induced break-
down spectroscopy signal generation underwater using dual pulse excitation part
i: Vapor bubble, shockwaves and plasma, Spectrochim.Acta Part B 82, 42–49. 5,
27, 40, 56, 58, 65, 93
Lazic, V., Laserna, J. and Jovicevic, S. (2013b), Insights in the laser induced break-
down spectroscopy signal generation underwater using dual pulse excitation part
ii: Plasma emission intensity as a function of interpulse delay, Spectrochim.Acta
Part B 82, 50–59. 27, 73
Lentz, W. J., Atchley, A. A. and Gaitan, D. F. (1995), Mie scattering from a sono-
luminescing air bubble in water, Appl. Opt. 34, 2648–2654. 30, 75
Lo, K. M. and Cheung, N. H. (2002), Arf laser-induced plasma spectroscopy for part-
per-billion analysis of metal ions in aqueous solutions, Appl. Spect. 56, 682–688.
82
Margulis, M. A. and Margulis, I. M. (2006), Luminescence mechanism of acoustic
and laser-induced cavitation, Acoust. Phys. 52, 283–292. 3, 21
Mart´ı-Lo´pez, L., Ocan˜a, R., Pin˜eiro, E. and Asensio, A. (2011), Laser peening
induced shock waves and cavitation bubbles in water studied by optical schlieren
visualization, Physics Procedia 12, 442–451. 15, 19
Matsumoto, A., Tamura, A., Fukami, K., Ogata, Y. H. and Sakka, T. (2013a),
Single-pulse underwater laser-induced breakdown spectroscopy with nongated de-
tection scheme, Anal. Chem. 85, 3807–3811. 16, 62, 93
Matsumoto, A., Tamura, A., Fukami, K., Ogata, Y. H. and Sakka, T. (2013b),
Two-dimensional space-resolved emission spectroscopy of laser ablation plasma in
water, J. Appl. Phys. 113, 053302, 7pp. 62
Matsumoto, A., Tamura, A., Kawasaki, A., Honda, T., Gregorcˇicˇ, P., ichi Amano,
N. N. K., Fukami, K. and Sakka, T. (2016), Comparison of the overall temporal
118
BIBLIOGRAPHY
nehavior of the bubbles produced by short- and long-pulse nanosecond laser ab-
lations in water using a laser-beam-transmission probe, Appl. Phys. A 122, 234,
6pp. 71, 72
Matula, T. J., Hilmo, P. R., Storey, B. D. and Szeri, A. J. (2002), Radial response
of individual bubbles subjected to shock wave lithotripsy pulses in vitro, Phys.
Fluids 14, 913–921. 21, 31
Ma¨tzler, C. (2002), MATLAB Functions for Mie Scattering and Absorption, IAP
Res. Rep. No. 02-08,. 31
McK. Ellison, J. and Peetz, C. V. (1959), The forward scattering of light by spheres
according to geometrical optics, Proc. Phys. Soc. 74, 105–123. 31
Mene´ndez-Manjo´n, A., Wagener, P. and Barcikowski, S. (2011), Transfer-matrix
method for efficient ablation by pulsed laser ablation and nanoparticle generation
in liquids, J. Phys. Chem. C 115, 5108–5114. 13
Michel, A. P. M. and Chave, A. D. (2008), Double pulse laser-induced breakdown
spectroscopy of bulk aqueous solutions at oceanic pressures: interrelationship of
gate delay, pulse energies, interpulse delay, and pressure, Appl. Opt. 47, G131–
G143. 5, 15
Mie, G. (1908), Beitra¨ge zur optik tru¨ber medien, speziell kolloidaler metallo¨sungen,
Annalen der physik 330, 377–445. 30
Miziolek, A., Palleschi, V. and Schechter, I. (2006), Laser-Induced Breakdown Spec-
troscopy (LIBS), Cambridge University Press. 4
Musazzi, S. and Perini, U. (2014), Laser-Induced Breakdown Spectroscopy - Theory
and Applications, Springer-Verlag Berlin Heidelberg. 4, 10, 38, 91
Mun˜oz, M. P., Luengo, M. C. L., Llorente, J. M. S., Sa´nchez, M. P., Garc´ıa, A. A. . A.
and Pedraz, P. M. (2010), Morphological alterations in dentine after mechanical
treatment and ultrashort pulse laser irradiation, Lasers Med. Sci. 27, 53–58. 12
Nath, A. and Khare, A. (2008), Measurement of charged particles and cavitation
bubble expansion velocities in laser induced breakdown in water, Laser Part.
Beams 26, 425–432. 71
119
BIBLIOGRAPHY
Nath, A. and Khare, A. (2011), Transient evolution of multiple bubbles in laser
induced breakdown in water, Laser Part. Beams 29, 1–9. 21, 71
Nguyen, T. T. P., Tanabe, R. and Ito, Y. (2013), Influences of focusing condi-
tions on dynamics of laser ablation at a solid-liquid interface, Appl. Phys. Express
6, 122701(4pp). 38, 45
Niemz, M. H. (2004,2007), Laser-Tissue Interactions,Fundamentals and Applica-
tions, Springer-Verlag Berlin Heidelberg. 1, 2, 7, 8, 9, 11, 12, 13
Noack, J., Hammer, D. X., Noojin, G. D., Rockwell, B. A. and Vogel, A. (1998),
Influence of pulse duration on mechanical effects after laser-induced breakdown
in water, J. Appl. Phys. 83, 7488–7495. 20
Noll, R. (2012), Laser-Induced Breakdown Spectroscopy - Fundamentals and appli-
cations, Springer-Verlag Berlin Heidelberg. 4, 37, 39, 91
Nuzzo, V., Savoldelli, M., Legeais, J.-M. and Plamann, K. (2010), Self-focusing and
spherical aberrations in corneal tissue during photodisruption by femtosecond
laser, Journal of Biomedical Optics 15, 038003, 8pp. 12
Nyga, R. and Neu, W. (1993), Double-pulse technique for optical emission spec-
troscopy of ablation plasmas of samples in liquids, Opt. Lett. 18, 747–749. 5
Oguchi, H., Sakka, T. and Ogata, Y. H. (2007), Effects of pulse duration upon the
plume formation by the laser ablation of cu in water, J. Appl. Phys. 102, 023306,
6pp. 56, 58, 62
Ohata, M., Iwasaki, Y., Furuta, N. and Brenner, I. B. (2002), Studies on laser
defocusing effects on laser ablation inductively coupled plasma-atomic emission
spectrometry using emission signals from a laser-induced plasma, Spectrochim.
Acta Part B 57, 1713–1725. 3
Ohl, C.-D., Kurz, T., Geisler, R., Lindau, O. and Lauterborn, W. (1999), Bub-
ble dynamics, shock waves and sonoluminescence, Phil. Trans. R. Soc. Lond. A
357, 269–294. 21, 28
Ornstein, L. S. and Brinkman, H. (1931), Temperature determination from band-
spectra. i. vibrational energy distribution and vibrational transition probabilities
in the cyanogen 2
∑−−2∑ bandsystem., Communication from the Physical In-
stitute of the University of Utrecht pp. –. 88
120
BIBLIOGRAPHY
Parigger, C., Woods, A., Surmick, D., Gautam, G., Witte, M. and Hornkohl, J.
(2015), Computation of diatomic molecular spectra for selected transitions of alu-
minum monoxide, cyanide, diatomic carbon, and titanium monoxide, Spectrochim.
Acta, Part B 107, 132–138. 87, 88
Peel, C. S., Fang, X. and Ahmad, S. R. (2011), Dynamics of laser-induced cavitation
in liquid, Appl Phys A 103, 1131–1138. 21
Peter Schaaf, e. (2010), Laser Processing of Materials Fundamentals, Applications
and Developments, Springer-Verlag Berlin Heidelberg. 14
Petkovsˇek, R. and Gregorcˇicˇ, P. (2007), A laser probe measurement of cavitation
bubble dynamics improved by shock wave detection and compared to shadow
photography, J. Appl. Phys. 102, 044909, 9pp. 71
Petkovsˇek, R., Gregorcˇicˇ, P. and Mozˇina, J. (2007), A beam-deflection probe as a
method for optodynamic measurements of cavitation bubble oscillations, Meas.
Sci. Technol. 18, 2972–2978. 21
Peyre, P., Berthe, L., Scherpereel, X. and Fabbro, R. (1998), Laser-shock processing
of aluminium-coated 55c1 steel in water-confinement regime, characterization and
application to high-cycle fatigue behaviour, J. Mater. Sci. 33, 1421–1429. 15
Pichahchy, A., Cremers, D. and Ferris, M. (1997), Elemental analysis of metals
under water using laser-induced breakdown spectroscopy, Spectrochim. Acta Part
B 52, 25–39. 15
Plesset, M. (1949), The dynamics of cavitation bubbles, J. Appl. Mech. 16, 277–282,.
21
Plesset, M. S. and Prosperetti, A. (1977), Bubble dynamics and cavitation, Ann.Rev.
Fluid. Mech. 9, 145–185,. 21
Povarnitsyn, M. E., Itina, T. E., Levashov, P. R. and Khishchenko, K. V. (2013),
Mechanisms of nanoparticle formation by ultra-short laser ablation of metals in
liquid environment, Phys.Chem. Chem. Phys. 15, 3108–3114. 13
Pu, X. Y. and Cheung, N. H. (2003), Arf laser induced plasma spectroscopy of lead
ions in aqueous solutions: Plume reheating with a second nd : Yag laser pulse,
Appl. Spectrosc. 57, 588–590. 5
121
BIBLIOGRAPHY
Rai, N. K. and Rai, A. (2008), Libsan efficient approach for the determination of cr
in industrial wastewater, J. Hazard. Mater. 150, 835–838. 15
Rayleigh, L. (1917), The pressure developed in a liquid during the collapse of a
spherical cavity, Philos. Mag. 34, 94–98,. 2, 21
Reich, S., Scho¨nfeld, P., Wagener, P., Letzel, A., Ibrahimkutty, S., Go¨kce, B., Bar-
cikowski, S., Menzel, A., dos Santos Rolo, T. and Plech, A. (2017), Pulsed laser
ablation in liquids: Impact of the bubble dynamics on particle formation, J. Coll.
Int. Sci. 489, 106–113. 13
Rekhi, S., Tempere, J. and Silvera, I. F. (2003), Temperature determination for
nanosecond pulsed laser heating, Rev. Sci. Instrum. 74, 3820–3825. 80
R.Griem, H. (1974), Spectral line broadening by plasmas, Academic Press, inc. (Lon-
don) ltd. 42, 85, 87
Saito, K., Takatani, K., Sakka, T. and Ogata, Y. H. (2002), Observation of the light
emitting region produced by pulsed laser irradiation to a solidliquid interface,
Appl. Surf. Sci. 197–198, 56–60. 55
Sakka, T., Oguchi, H., Masai, S., Hirata, K., Ogata, Y. H., Saeki, M. and Ohba, H.
(2006), Use of a long-duration ns pulse for efficient emission of spectral lines from
the laser ablation plume in water, Appl. Phys. Lett. 88, 061120, 3pp. 16
Sakka, T., Saito, K. and Ogata, Y. H. (2005), Confinement effect of laser ablation
plume in liquids probed by self-absorption of c2 swan band emission, J. Appl.
Phys. 97, 014902,4pp. 42
Sakka, T., Takatani, K., Ogata, Y. H. and Mabuchi, M. (2002), Laser ablation at
the solidliquid interface: transient absorption of continuous spectral emission by
ablated aluminium atoms, J. Phys. D: Appl. Phys. 35, 65–73. 55
Sakka, T., Tamura, A., Matsumoto, A., Fukami, K., Nishi, N. and Thornton, B.
(2014), Effects of pulse width on nascent laser-induced bubbles for underwater
laser-induced breakdown spectroscopy, Spectrochim. Acta Part B 97 (2014) 9498
97, 94–98. 21
Sakka, T., Tamura, A., Nakajima, T., Fukami, K. and Ogata, Y. H. (2012), Syn-
ergetic effects of double laser pulses for the formation of mild plasma in water:
122
BIBLIOGRAPHY
Toward non-gated underwater laser-induced breakdown spectroscopy, J. Chem.
Phys. 136, 174201, 5pp. 27
Salminen, T. (2013), Production of Nanomaterials by Pulsed Laser Ablation, Thesis,
Tampere University of Technology, Publication 1121. 13
Samant, A. N. and Dahotre, N. B. (2009), Laser machining of structural ceramics –
a review, J. Eur. Cer. Soc. 29, 969993. 94
Sasaki, K., Nakano, T., Soliman, W. and Takada, N. (2009), Effect of pressurization
on the dynamics of a cavitation bubble induced by liquid-phase laser ablation,
Appl. Phys. Express 2, 046501, 3pp. 70, 98
Sasaki, K. and Takada, N. (2010), Liquid-phase laser ablation, Pure Appl. Chem.
82, 1317–1327. 5, 13
Sell, J. A., Heffelfinger, D. M., Ventzek, P. and Gilgenbach, R. M. (1989), Laser beam
deflection as a probe of laser ablation of materials, Appl. Phys. Lett. 55, 2435–
2437. 70
Settles, G. (2006), Schlieren and Shadowgraph techniques: Visualizing Phenomena
in transparent media, Springer-Verlag Berlin Heidelberg. 47, 48
Shangguan, H., Casperson, L. W., Paisley, D. L. and Prahl, S. A. (1998), Pho-
tographic studies of laser-induced bubble formation in absorbing liquids and on
submerged targets: implications for drug delivery with microsecond laser pulses,
Opt. Eng. 37, 2217–2226,. 13
Shangguan, H., Casperson, L. W., Shearin, A., Paisley, D. L. and Prahl, S. A.
(1997), Effects of material properties on laser-induced bubble formation in ab-
sorbing liquids and on submerged targets, Proc. SPIE 2869, 783–791. 1
Shen, Y. R. (1984), The Principles of Nonlinear Optics, John Wiley and Sons, New
York. 35
Shockley, W. (1961), Problems related to p–n junction, Solid-State Electron. 2, 35–
67. 33
Singh, J. P. and Thakur, S. N. (2007), Laser-Induced Breakdown Spectroscopy, El-
sevier B.V. 4
123
BIBLIOGRAPHY
Sola, X., Ignatiev, M., Flamant, G., Smurov, I. and Konstantinov, S. (1997), Mon-
itoring of melting and solid state phase transformation induced by pulsed laser
action on pure zirconium, Appl. Surf. Sci. 109–110, 106–112. 96
Soliman, W., Takada, N. and Sasaki, K. (2010), Growth processes of nanoparticles
in liquid-phase laser ablation studied by laser-light scattering, Appl. Phys. Express
3, 035201, 3pp. 13, 63, 70
Sriramachandran, P., Viswanathan, B. and Shanmugavel, R. (2013), Occurrence of
alo molecular lines in sunspot umbral spectra, Solar Phys. 286, 315–326. 88
Stambulchik, E., Fisher, D., Maron, Y., Griem, H. and Alexiou, S. (2007), Corre-
lation effects and their influence on line broadening in plasmas: Application to
hα, HighEnergyDensityPhys.3, 272−−277.42, 43
Stehle, C., Caillol, J., Escarguel, A. and Gilles, D. (2000), Hydrogen lines in corre-
lated plasmas, J. Phys. IV France 10, 501–504. 43, 86
Suzuki, H., Nishikawa, H. and Sandy Lee, I.-Y. (2002), Laser-induced breakdown
spectroscopy at metalwater interfaces, Phys. Chem .Comm. 5, 88–90. 41, 79, 93
Tagawa, Y., Yamamoto, S., Hayasaka, K. and Kameda, M. (2016), On pressure
impulse of a laser-induced underwater shock wave, J. Fluid Mech. 808, 5–18. 19
Takada, N., Nakano, T. and Sasaki, K. (2010), Formation of cavitation-induced pits
on target surface in liquid-phase laser ablation, Appl. Phys. A 101, 255–258. 2,
29, 40
Tamura, A., Matsumoto, A., Fukami, K., Nishi, N. and Sakka, T. (2015), Simul-
taneous observation of nascent plasma and bubble induced by laser ablation in
water with various pulse durations, J. Appl. Phys. 117, 173304, 8pp. 6
Tao, S., Zhou, Y., Wu, B. and Gao, Y. (2012), Infrared long nanosecond laser
pulse ablation of silicon: Integrated two-dimensional modeling and time-resolved
experimental study, Appl. Surf. Sci. 258, 7766–7773. 58, 61
Thornton, B., Sakka, T., Takahashi, T., Tamura, A., Masamura, T. and Matsumoto,
A. (2013), Spectroscopic measurements of solids immersed in water at high pres-
sure using a long-duration nanosecond laser pulse, Appl. Phys. Express 6, 082401,
4pp. 70, 98
124
BIBLIOGRAPHY
Thornton, B., Takahashi, T., Sato, T., Sakka, T., Tamura, A., Matsumoto, A.,
Nozaki, T., Ohki, T. and Ohki, K. (2015), Development of a deep-sea laser-
induced breakdown spectrometer for in situ multi-element chemical analysis, Deep
Sea Res. Part I 95, 20–36. 16
Thornton, B. and Ura, T. (2011), Effects of pressure on the optical emissions ob-
served from solids immersed in water using a single pulse laser, Appl. Phys. Express
4, 022702, 3pp. 56
Tian, Y., Xue, B., Jiaojian Song, Y. L. and Zheng, R. (2015), Non-gated laser-
induced breakdown spectroscopy in bulk water by position-selective detection,
Appl. Phys. Lett. 107, 111107, 5pp. 37
Tian, Y., Xue, B., Song, J., Lu, Y. and Zheng, R. (2016), Stabilization of laser-
induced plasma in bulk water using large focusing angle, Appl. Phys. Lett.
109, 061104, 5 pp. 37
Tiwari, D., Bellouard, Y., Dietzela, A., Ren, M., Rubingh, E. and Meinders, E.
(2010), Dynamical observation of femtosecond-laser-induced bubbles in water us-
ing a single laser source for probing and sensing, Appl. Phys. Express 3, 127101,
3pp. 65
Toker, G., Bulatov, V., Kovalchuk, T. and Schechter, I. (2009), Micro-dynamics
of optical breakdown in water induced by nanosecond laser pulses of 1064 nm
wavelength, Chem. Phys. Lett. 471, 244–248. 36, 37
Tomita, Y. and Kodama, T. (2003), Interaction of laser-induced cavitation bubbles
with composite surfaces, J. Appl. Phys. 94, 2809–2816. 21
Totten, G. E. and MacKenzie, D. S. (2003), Handbook of Aluminum, Volume 1,
Physical Metallurgy and Processes, Marcel Dekker, Inc. 94
Tsai, C.-H. and Li, C.-C. (2009), Investigation of underwater laser drilling for brittle
substrates, Journal of materials processing technology 209, 2838–2846. 15
T.Tsuji, Okazaki, Y., Tsuboi, Y. and Tsuji, M. (2007), Nanosecond time-resolved
observations of laser ablation of silver in water, Jpn. J. Appl. Phys. 46, 1533–1535.
19, 20, 55, 63, 78
125
BIBLIOGRAPHY
Ushida, H., Takada, N. and Sasaki, K. (2007), Diagnostics of liquid-phase laser
ablation plasmas by spectroscopic methods, J. Phys.: Conf. Ser. 59, 563–566. 41,
79
Vadim A Petrov, A. Y. V. (2007), Spectral emissivity and radiance temperature
plateau of self-supporting al2o3 melt at rapid solidification, High Temperatures-
High Pressures 35/36, 321–329. 96
Vasil’ev, L. (1971), Schlieren methods, IIsrael Program for Scientific translation. 48
Venugopal Rao, S., Podagatlapalli, G. K. and Hamad, S. (2014), Ultrafast laser
ablation in liquids for nanomaterials and applications, Journal of Nanoscience
and Nanotechnology 14, 1–25. 13
Vitel, Y., Bezzari, M. E., Gavrilova, T., Skowronek, M.,
D’yachkov, L. and Kurilenkov, Y. (1996), Shift and width of the
hαlineofhydrogenindenseplasmas, PhysicaB228, 161−−165.43
Vogel, A. (1997), Nonlinear absorption: intraocular microsurgery and laser
lithotripsy, Phys. Med. Biol. 42, 895–912. 12
Vogel, A. (1998), Optical Breakdown in Water and Ocular Media, and its Use
for Intraocular Photodisruption, Habilitation dissertation, Medical University of
Lu¨beck, Germany. 1, 17, 19
Vogel, A., Busch, S. and Parlitz, U. (1996), Shock wave emission and cavitation
bubble generation by picosecond and nanosecond optical breakdown in water, J.
Acoust. Soc. Am. 100, 148–165. 12, 19
Vogel, A. and Lauterborn, W. (1988), Acoustic transient generation by laser-
produced cavitation bubbles near solid boundaries, J. Acoust. Soc. Am. 84, 719–
731. 19
Vogel, A., Lauterborn, W. and Timm, R. (1989), Optical and acoustic investigations
of the dynamics of laser-produced cavitation bubbles near a solid boundary, J.
Fluid Meh. 206, 299–338. 2, 21, 26, 28
Vogel, A., Noack, J., Nahen, K., Theisen, D., Busch, S., Parlitz, U., Hammer, D.,
Noojin, G., Rockwell, B. and R.Birngruber (1999), Energy balance of optical
breakdown in water at nanosecond to femtosecond time scales, Appl. Phys. B
68, 271–280. 78
126
BIBLIOGRAPHY
Vogel, A., Schweiger, P., Frieser, A., Asiyo, M. N. and Birngruber, R. (1990), In-
traocular nd:yag laser surgery:light-tissue interaction, damage range, and reduc-
tion of collateral effects, IEEE J. Quant. Electron. 26, 2240–2260. 12
Vogel, A. and Venugopalan, V. (2003), Mechanisms of pulsed laser ablation of bio-
logical tissues, Chem. Rev. 103, 577–644. 1, 12, 28
Wagener, P., Ibrahimkutty, S., Menzel, A., Plech, A. and Barcikowski, S. (2013),
Dynamics of silver nanoparticle formation and agglomeration inside the cavitation
bubble after pulsed laser ablation in liquid, Phys. Chem. Chem. Phys. 15, 3068–
3074. 13, 63
Wallace, J. M. and Hobbs, P. V. (2006), Atmospheric Science, An Introductory
Survey, Elsevier Inc. 30
Wen, S.-B., Mao, X., Greif, R. and Russo, R. E. (2007a), Expansion of the laser ab-
lation vapor plume into a background gas.i. analysis, J. Appl. Phys. 101, 023114,
13pp. 4
Wen, S.-B., Mao, X., Greif, R. and Russo, R. E. (2007b), Laser ablation induced
vapor plume expansion into a background gas. ii. experimental analysis, J. Appl.
Phys. 101, 023115, 14pp. 4
Williams, F., Varma, S. P. and Hillenius, S. (1976), Liquid water as a lone-pair
amorphous semiconductor, J. Chem. Phys. 64, 1549–1554. 34
Wiscombe, W. J. (1980), Improved mie scattering algorithms, Appl. Opt. 19, 1505–
1509. 31
Wook Kang—, H., Lee, H. and Welch, A. J. (2008), Laser ablation in a liquid-
confined environment using a nanosecond laser pulse, J. Appl. Phys. 103, 083101,
6pp. 14, 41
Yang, G. (2012), Laser ablation in liquids, Principles and applications in the prepa-
ration of the nanomaterials, Pan Stanford Publishing. 5, 13
Yu, H., Shen, J. and Wei, Y. (2008), Geometrical optics approximation of light
scattering by large air bubbles, Particuology 6, 340–346. 30, 31
127
BIBLIOGRAPHY
Yueh, F.-Y., Sharma, R. C., Singh, J. P. and Zhang, H. (2002), Evaluation of laser-
induced breakdown spectroscopy as technetium monitor, J. Air Waste Manag.
Assoc. 52, 1307–1315. 15
Zel’dovich, Y. B., Raizer, Y. P., Hayes, W. D. and Probstein, R. F. (1966), Physics
of shock waves and high-temperature phenomena, Academic press Inc. 17
Zeng, H., Du, X.-W., Singh, S. C., Kulinich, S., Yang, S., He, J. and Cai, W.
(2012), Nanomaterials via laser ablation/irradiation in liquid: A review, Adv.
Funct. Mater. 22, 1333–1353. 13
Zhang, Z. and Modest, M. F. (1998), Temperature-dependent absorptances of ceram-
ics for nd:yag and co2 laser processing applications, J. Heat Transfer 120, 322–327.
96
Zhong, P., Zhou, Y. and Zhu, S. (2001), Dynamics of bubble oscillation in con-
strained media and mechanisms of vessel rupture in swl, Ultrasound Med. Biol.
27, 119–134. 2
Zhou, Y., Tao, S. and Wu, B. (2011), Backward growth of plasma induced by long
nanosecond laser pulse ablation, Appl. Phys. Lett. 99, 051106, 3pp. 58, 61
Zhu, S., Lu, Y. F., Hong, M. H. and Chen, X. Y. (2001), Laser ablation of solid
substrates in water and ambient air, J. Appl. Phys. 89, 2400–2403. 14
128
Biography
Marijana Gavrilovic´ was born 12.06.1986 in Kragujevac. After graduating from
the I Gymnasium in Kragujevac in 2005, she enrolled in the School of Electrical
Engineering at the University of Belgrade. She graduated from the Department
of Physical Electronics in 2009, with the GPA of 8.36 (out of 10.00). In 2011
she received her master degree from the Department of Physical Electronics of the
School of Electrical Engineering. She entered the Nanoelectronics and Photonics
Ph.D. program at the same university in 2011. She is currently at the third year od
the Ph.D. program and she passed all the exams successfully
M. R. Gavrilovic´ has been working as a research assistant in the Laboratory for
the Plasma Spectroscopy and Lasers in Photonics Center at the Institute of Physics
Belgrade since 2011, participating in the national research project “Spectroscopic
diagnostics of low temperature plasma and gas discharges: spectral line shapes and
interaction with surfaces”(OI171014).
She participated in bilateral project with France Pavle Savic´ (680-00-132/2012-
09/03, (2012-2014)) and Slovakia (451-03-545/2015-09/12, 2015-2016).
She was one of te chosen candidates for the participation in conference and school
at the International Center for Theoretical Physics (ICTP) in Trieste, organized
jointly by the ICTP and IAEA (International Atomic Energy Egency) under the
title ”Modern methods in plasma spectroscopy”
As a guest researcher, she visited the laboratories of the Photonics Department
at the Jagiellonian University, where she was involved in the experimental studies
of laser induced plasma by Thomson scattering and optical emission spectroscopy.
The most important research results of the Ph.D. candidate Marijana R. Gavrilovic´
are summarized in seven articles published in international journals. ——————
——————————
129
List of publications
1. M. R. Gavrilovic´ and M. Cvejic´ and V. Lazic and S. Jovic´evic´, Secondary
plasma formation after single pulse laser ablation underwater and its advan-
tages for laser induced breakdown spectroscopy (LIBS), Phys. Chem. Chem.
Phys. 18 14629–14637 (2016) ( ISSN: 1463-9076 doi: 10.1039/C6CP01515H
Impact factor: 4.449)
2. M. R. Gavrilovic´ and V. Lazic and S. Jovic´evic´, Influence of the target material
on secondary plasma formation underwater and its laser induced breakdown
spectroscopy (LIBS) signal, J. Anal. At. Spectrom. 32 345–353 (2017) (ISSN
0267-9477 DOI: 10.1039/c6ja00300a, Impact factor: 3.379)
3. M. Cvejic´ and M.R. Gavrilovic´ and S. Jovic´evic´ and N. Konjevic´, Stark broad-
ening of Mg I and Mg II spectral lines and Debye shielding effect in laser
induced plasma, Spectrochim. Acta Part B 85 20–33 (2013) (ISSN: 0584-8547
DOI:10.1016/j.sab.2013.03.011 Impact factor: 3.194)
4. M. Cirisan and M. Cvejic´ and M.R. Gavrilovic´ and S. Jovic´evic´ and N. Konjevic´
and J.Hermann, Stark broadening measurement of Al II lines in a laser-induced
plasma, J. Quant. Spectrosc. Radiat. Transfer. 133 652–662 (2014) (ISSN:
0022-4073 DOI: 10.1016/j.jqsrt.2013.10.002 Impact factor: 2.768)
5. M. Cvejic´ and E. Stambulchik and M.R. Gavrilovic´ and S. Jovic´evic´ and N.
Konjevic´, Neutral lithium spectral line 460.28 nm with forbidden component
for low temperature plasma diagnostics of laser-induced plasma, Spectrochim.
Acta Part B 100 86–97 (2014) (ISSN: 0584-8547 DOI: doi:10.1016/j.sab.2014.08.007
Impact factor: 3.176)
6. Irene L. Epstein and Marijana Gavrilovic´ and Sonja Jovic´evic´ and Nikola Kon-
jevic´ and Yuri A. Lebedev and Alexey V.Tatarinov, The study of a homoge-
neous column of argon plasma at a pressure of 0.5 torr, generated by means of
130
the Beenakkers cavity, Eur. Phys. J. D 68 334–343 (2014) (ISSN: 1434-6060
DOI:10.1140/epjd/e2014-50182-7 Impact factor: 1.24 )
7. Olivera Ciraj-Bjelac, Marijana Gavrilovic, Danijela Arandjic, Milan Vujovic,
Predrag Bozovic, Radiation exposure during x-ray examinations in a large
paediatric hospital in Serbia, Radiation Protection Dosimetry 165 220–225
(2015) (ISSN: 0144-8420 DOI: 10.1093/rpd/ncv084 Impact factor: 0.916)
131