«by YUAN LIU B.S. Information Engineering, Zhejiang University, 2006 M.S. Optics, University of Central Florida, 2010 A dissertation submitted in ...»
DEVELOPMENT OF LASER SPECTROSCOPY FOR ELEMENTAL AND MOLECULAR
B.S. Information Engineering, Zhejiang University, 2006
M.S. Optics, University of Central Florida, 2010
A dissertation submitted in partial fulfillment of the requirements
for the degree of Doctor of Philosophy
in the College of Optics and Photonics at the University of Central Florida Orlando, Florida Spring Term 2013 Major Professor: Martin C. Richardson © 2013 Yuan Liu ii
ABSTRACTLaser-Induced Breakdown Spectroscopy (LIBS) and Raman spectroscopy are still growing analytical and sensing spectroscopic techniques. They significantly reduce the time and labor cost in analysis with simplified instrumentation, and lead to minimal or no sample damage.
In this dissertation, fundamental studies to improve LIBS analytical performance were performed and its fusion with Raman into one single sensor was explored.
On the fundamental side, Thomson scattering was reported for the first time to simultaneously measure the electron density and temperature of laser plasmas from a solid aluminum target at atmospheric pressure. Comparison between electron and excitation temperatures brought insights into the verification of local thermodynamic equilibrium condition in laser plasmas.
To enhance LIBS emission, Microwave-Assisted LIBS (MA-LIBS) was developed and characterized. In MA-LIBS, a microwave field extends the emission lifetime of the plasma and stronger time integrated signal is obtained. Experimental results showed sensitivity improvement (more than 20-fold) and extension of the analytical range (down to a few tens of ppm) for the detection of copper traces in soil samples.
Finally, laser spectroscopy systems that can perform both LIBS and Raman analysis were developed. Such systems provide two types of complimentary information – elemental composition from LIBS and structural information from Raman. Two novel approaches were reported for the first time for LIBS-Raman sensor fusion: (i) an Ultra-Violet system which combines Resonant Raman signal enhancement and high ablation efficiency from UV radiation, and (ii) a Ti:Sapphire laser based NIR system which reduces the fluorescence interference in Raman and takes advantage of femtosecond ablation for LIBS.
iii To my family iv
ACKNOWLEDGMENTSI would like to express my sincere gratitude and appreciation to my advisor Dr. Martin Richardson for the guidance and support during myPh.D. study. I would also like to thank my committee members Dr. Eric Van Stryland, Dr. Michael Bass, and Dr. Michael Sigman for their time and advice throughout my candidacy, proposal and defense.
Dr. Matthieu Baudelet and his great passion in spectroscopy played irreplaceable roles during my Ph.D. study. We went through all the projects described in this dissertation together.
Through numerous discussions, presentation rehearsals, and paper revisions with him, I gradually gained the skills and experiences needed to become a young scientist. Many thanks for all your time and inputs during the last five years.
Certainly Dr. Bruno Bousquet from the University of Bordeaux in France is a must-have name in this list. He is a great collaborator and mentor in LIBS and physics. His critical thinking and his way to plan and organize research projects impress me a lot. Without his support, a large portion of this dissertation could not be completed. It was my great pleasure to work with him during his visit.
I wish to thank Dr. Santiago Palanco, who brought me into the Laser Plasma Laboratory and introduced LIBS to me. I also want to thank Dr. Christopher Brown, and Dr. Matthew Weidman, from whom I learned the basic laboratory skills and fundamentals about LIBS.
Past and present members in the Laser Plasma Laboratory kindly offered me consistent and generous help. I would like to thank Tony and Nick for their help on all sorts of electronic problems, thank John for the assistance in setting up the laser channel used in the Thomson
system and Zygo interferometer, thank Ji-Yeon and Pankaj for showing me how to operate the Cary spectrometer, and thank Khan, Nathan, Benn and many others (sorry we are a large group and can’t put all the names who helped me) for lending me optical and mechanical components and also the conversations and happy time with them.
Last but not least, I would like to thank many friends I met at CREOL and UCF. These people added diversity and color during this important period of my life.
TABLE OF CONTENTS
LIST OF FIGURES
LIST OF TABLES
LIST OF ABBREVIATIONS
CHAPTER 1 INTRODUCTION
1.2 Pros and Cons of Current methods
1.3 LIBS and Raman: key anchors in analysis
1.4 Overview of the thesis
CHAPTER 2 LASER-INDUCED BREAKDOWN SPECTROSCOPY
2.1 LIBS Introduction
2.2 Fundamentals of LIBS Plasma
2.2.2 Equilibrium state of laser-induced plasma
2.3 Diagnostics for LIBS Plasmas
2.3.1 Excitation Temperature
2.3.2 Electron Density
2.3.3 Thomson Scattering for Te and ne Diagnostics
2.4 LIBS Instrumentation
2.4.4 Improvement of LIBS
CHAPTER 3 LASER PLASMA DIAGNOSTICS BY THOMSON SCATTERING..... 36
3.3 Air plasma
3.3.1 Collective Thomson scattering spectra
3.3.3 Time resolved Te and ne measurements for air plasma
3.4 Aluminum plasma
3.4.1 Plasma imaging and probe position
3.4.2 Plasma heating
3.4.3 Thomson scattering spectrum
3.4.4 Time resolved Te and ne measurements
3.4.5 McWhirter criterion and Local Thermodynamic Equilibrium
4.2.1 Microwave system
4.2.2 Ablation system
4.2.3 Detection system
4.3 MA-LIBS Plasma Characterization
4.3.1 Plasma Emission
ix 4.3.2 Plasma temperature estimation
4.3.3 Electron density
4.4 Performance of MA-LIBS
4.4.1 Signal enhancement
4.4.2 Improvement of sensitivity
4.4.3 Selective enhancement of lines
4.4.4 Enhancement of molecular emission
4.4.5 Enhancement dependence on laser irradiance
4.4.6 Influence of atmosphere condition
CHAPTER 5 FUNDAMENTALS OF RAMAN SPECTROSCOPY
5.1 Raman Scattering and Molecule Vibration
5.1.1 Spontaneous Raman scattering
5.1.2 Raman active modes
5.1.3 Raman intensity
5.2 Raman spectroscopy
5.2.1 Spontaneous Raman Spectroscopy
5.2.3 Comparison with IR spectroscopy
5.2.4 Fluorescence in Raman spectroscopy
CHAPTER 6 MOLECULAR INFORMATION FROM LIBS SPECTRA AND LIBS-RAMANSENSOR FUSION
6.2 Correlation between LIBS signal and moisture content
6.2.3Results and discussions
6.3 LIBS-Raman Sensor Fusion
6.3.2 532 nm LIBS-Raman System
6.3.3 266 nm LIBS-Raman System
6.3.4 785 nm LIBS-Raman System
6.3.5 Summary of LIBS-Raman sensor fusion
CHAPTER 7 CONCLUSIONS AND FUTURE WORK
7.2 Future work
Figure 2.1: AES procedure and comparison with LIBS.
Figure 2.2:Plasma creation steps
Figure 2.3: Ionization with (a) multiphoton ionization, and (b) cascade ionization.
In the figure, e- represents electrons; hν represents photons; IB is inverse Bremsstrahlung absorption........... 11 Figure 2.4: Illustration of the regimes where Saha equation and Boltzmann distribution are applied.
Parameters in this figure can be found in the Saha equation and Boltzmann distribution.
Figure 2.5: The evolution of plasma emission over time.
(Courtesy of Matthew Weidman)...... 16 Figure 2.6: A LIBS spectrum from an yttrium aluminum garnet (YAG) crystal.
Figure 2.7: Calibration curve for LIBS measurements of Cu in soil using the 324.
75 nm line... 19 Figure 2.8: Boltzmann plots of Cr I in plasma from a Cr:LiCAF crystal. Estimated plasma temperature is about 7980 K.
Figure 2.9: Direction diagram for Thomson scattering.
Figure 2.10: Simulated normalized Thomson scattering spectra with Te = 8000 K and ne ranges from 1017 to 1013 cm-3.
The α values are 3.08, 2.18, 0.97, 0.69, 0.31, 0.22, 0.10, 0.07 and 0.03 respectively.
Figure 2.11: Fitting a real TS spectrum.
An air plasma was created and probed by a 532 nm laser.
The Te is estimated to be 66811 ± 360 K, and ne is 3.936 ± 0.016 × 1017 cm-3.
Figure 2.12: A typical LIBS system.
(DG: delay generator)
Figure 2.13: Schematic diagram of a Czerny-Turner spectrometer (a), and light diffraction from a reflection grating (b).
Figure 2.16: Common configurations of DP-LIBS.
(a) Collinear configuration, (b) orthogonal configuration with a reheating pulse, (c) orthogonal configuration with a pre-ablation pulse..... 34 Figure 2.17: MA-LIBS experimental setup 
Figure 3.1: Schematics of the experimental setup.
In the figure, WP = wave plate, PBS = polarization beam splitter, L1-6 = Lens 1-6.
Figure 3.2: Timing sequence for synchronization.
(a) Flash lamp and Q-switch; (b) Synchronization of the two lasers, the shutter, and the camera.
Figure 3.3: Synchronization for Thomson scattering experiment.
FL: flash lamp, QS: Q-switch, Tgr: External trigger input.
Figure 3.4: Typical collective Thomson scattering signal: ICCD image of the spectrum (left) and spectrum after binning (right).
Figure 3.5: Fitting (red) of the spectrum (blue) in Figure 3.
4. Te was found to be 58500 K and ne was 4.39×1017 cm-3.
Figure 3.6: Thomson scattering spectrum image taken at 2 µs after plasma creation.
................. 47 Figure 3.7: The difference between two normalized Rayleigh scattering spectra from air (red) and the difference between the normalized TS spectrum from an air plasma 4 µs after the plasma creation and the normalized Rayleigh scattering spectrum (blue).
Figure 3.8: Gaussian fit of the Thomson scattering spectrum from air plasma at 4 µs after plasma creation.
Figure 3.9: Te and ne of air plasma as a function of the delay time.
Figure 3.10: Probe laser and plasma images at different delay time.
2.2, and 3 μs after the plasma formation.
Figure 3.12: a) Raw Thomson scattering spectrum and b) Fitting of the blue-shifted satellite peak of the spectrum for the Al plasma 800 ns after plasma creation at 1 atm pressure.
Figure 3.13: Time evolution of the Te and ne of the Al plasma at 1 atm pressure.
Figure 3.14: The maximum transition energy that satisfies the McWhirter criterion as a function of delay time between the plasma ignition and detection.
The black squares are the calculated ΔE values from the measured Te and ne pairs at their specific delay time, when the McWhirter criterion is just satisfied. The red line is a linear fit of the energy as a function of the delay time.
Figure 3.15: Aluminum spectrum 600 ns after plasma creation for the Boltzmann plot.
............. 60 Figure 3.16: Excitation temperatures measured based on lines from atoms (black squares) and singly ionized ions (red circles).
Figure 4.1: Schematic illustration of the microwave cavity (Courtesy of Guangming Tao).
....... 66 Figure 4.2: Experimental setup for MA-LIBS.
Figure 4.3: Comparison between LIBS (black) and MA-LIBS (red) spectrum of Pb samples.
... 69 Figure 4.4: Comparison between the plasma emission at 405.8 nm with (red dot) and without (black square) the microwave. Plasma emission was normalized to characterize the emission per 50 ns.
Figure 4.5: Spectra of Pb plasma with (red) and without (black) microwave 18 ms after the laser pulse with gate width 2 ms.
Comparing with Figure 4.3, the high intensity in this spectrum is mainly due to much higher gain from the ICCD camera (from 1 to 100).
Figure 4.7: Electron density estimate of MA-LIBS plasma using non-collective Thomson scattering method, and a comparison with normal LIBS plasma.
Figure 4.8: Spectra for (a) Mg; (b) Al; (c) Ba; (d) Na; (e) K and (f) Ca.
(1: MA-LIBS, 2: LIBS)
Figure 4.9: Spectra of Montana soil NIST standard reference material using (a) traditional LIBS and (b) MA-LIBS.
Figure 4.10: Spectra of the soil sample S1 containing 1232 mg kg-1 of Cu with MA-LIBS (left) and LIBS (right).
Figure 4.11: Calibration curves of Cu for MA-LIBS (black squares) and the traditional LIBS (red circles).