Download Laser Induced Breakdown Spectroscopy - Lecture Notes | PHYS 598 and more Exams Physics in PDF only on Docsity!
Laser Induced Breakdown Spectroscopy
Molly Rossow
Abstract
In laser induced breakdown spectroscopy (LIBS) a high power laser pulse is used to ablate a small volume of a sample and excite it to a plasma state. As the ionized particles recombine, they emit light that can be analyzed to determine the elemental composition of the material. LIBS can be applied to a variety of applications in medicine, industry, and research. Samples need very little preparation and the surrounding area is not damaged. The accuracy of LIBS is dependent on selecting appropriate laser parameters for the application. There is no single LIBS system that will work in all situations. Much of the work done with LIBS is qualitative; it is used to test for the presence of a particular compound with out concern for the concentration. The complexity of the plasma plume makes it difficult to get accurate qualitative LIBS results but it is possible under certain conditions. Laser micromachining and art restoration are two areas of where LIBS is currently being applied.
Introduction
Laser induced breakdown spectroscopy (LIBS) is a technique for determining the elemental composition of a material. In LIBS, a high power laser pulse is focused onto the surface of the sample. Enough energy is delivered to a small volume to not only vaporize the material but to break all chemical bonds and ionize the elements present creating a small plasma plume. As the species in the plasma relax they emit at a characteristic wavelength. The spectrum evolves over time, becoming more distinct after several micro seconds. From this emission spectrum the constituent elements of the sample can be determined.
LIBS has many advantages over other spectroscopy techniques. It can be applied to a wide variety of samples. A sample of almost any size can be used. Liquids, solids, and gases can be analyzed. Very little material, from .1 micro p to 1mg, is destroyed leaving the sample essentially intact (Andor). This makes LIBS well suited to in situ measurements and delicate samples. Once instrumentation has been developed and calibrated for a particular application, LIBS measurements are straightforward because the sample does not need to be prepared beforehand. Instrumentation can be designed for a variety of applications. Fiber optics can be used to improve flexibility. Compact, portable systems are commercially available. Results are available almost instantaneously. It is even possible to use LIBS at a distance from the sample. This is useful for high temperature or toxic materials. LIBS can provide detailed information about the sample. The small sample volume gives it high spatial resolution (Evans). LIBS can be a quick analysis method. Many elements can be detected simultaneously (Andor). Results are available in real time (Hahn).
Despite these advantages the potential applications for LIBS are limited (Evans). Information about the sample is lost in the ablation process and only the elemental composition can be determined. Nothing is known about the molecular structure. Despite the potential for LIBS to be applied to many different situations there is no single method that will work for all applications. Techniques for each application need to be developed and optimized. This can be time consuming. Additionally, for most applications LIBS is less accurate than other available techniques (Tong). LIBS is only useful when the benefits outlined above outweigh this drawback or similar measurements will need to be taken repeatedly.
The history of LIBS dates back to the invention of the pulsed laser (Hahn). It was discovered that pulsed lasers could be used to turn small amounts of material into a plasma. Atomic emission spectroscopy using electrodes to generate the plasma was a developing field at the time and the potential to use lasers in the same was apparent (Radziemski). Leon Radziemski and David Cremers first used LIBS to determine the chemical composition of a substance in the early 1980s (Radziemski). Improvements in laser and detector technology increased the usefulness of LIBS through out the 1980s. High resolution spectrometers developed in the late 1990s increased sensitivity and allowed multiple elements to be detected at once (Hahn, Fichet). Ultrafast lasers have the potential to make LIBS even more powerful. Today LIBS instrumentation is commercially available for laboratory and industrial applications.
LIBS has application in fields ranging from medicine (Kumar) to steel manufacturing (Michaud). Medical applications include distinguishing between malignant and benign tumors based on elemental composition (Kumar). LIBS is useful in steel manufacturing because it can analyze high temperature molten steel from a distance. Equipment does not need to withstand high temperatures and technicians can stay at a safe distance. Forensics is a natural application since very little material is needed. LIBS has been used to measure levels of heavy metals in hair (Corsi) and to detect trace amounts of gun powder (Dockery). Infectious agents and toxins can be detected using LIBS both on surfaces and in the air (Morrel, Stepputat) making it useful for environmental research and in the rapidly advancing field of biosecurity. The real-time detection feature has been exploited in art restoration (Greorgio) and micromachining (Tong). In both these applications the goal is use the laser to remove a thin, controlled layer of material: dirt in the case of art restoration and one layer of a previously fabricated multilayer substrate in micromachining. A laser is used to ablate the surface and the plasma plume is monitored to determine when the composition of the material changes.
Methods and Analysis
The basic requirements for making LIBS measurements are a laser and a spectrometer. There is no universal LIBS system for all applications. All parameters must be optimized for each application. The laser wavelength, repetition rate, power, and spot size must be determined. The optimal detection range for the elements present must be used. The detected spectrum needs to be gated to eliminate background. The spectrum needs to be
One algorithm for spectra comparison is the largest peaks method. The largest intensity peaks are determined. Each peak is compared to a normalized reference spectrum at the same wavelength. The number of peaks selected is determined by the number of elements the experiment is designed to detect. The degree of comparison is ranked from 0 to 1 with 1 being identical to the reference. This technique is fast and easy to implement but its resolving capability is limited since it does not take the entire spectrum into account.
Linear correlation is a more complex, more accurate method. The linear correlation coefficient to compare the intensity of the data, Idata, to a reference intensity Ireference, at each wavelength, λ, is defined as
(1)
, ,
2 1/ 2 2 1/ 2 ( (^) , ) ( ,
data reference
data reference j j
I I
r I I
λ λ λ
λ λ
=
â
â â )
A correlation value close to 1 indicates a close match. This method can give inaccurate results when spectral lines are close together and overlap.
The characteristic lines correlation method combines the best qualities of these two methods. In this method the preprocessing step is different. Instead of discarding everything below a threshold value, local maxima are found in the normalized data and all other values are set to 0. This processed data is then correlated with the reference spectrum. This method is the slowest of the three but provides the most accurate results (Tong).
Quantitative Spectroscopy
In some applications it is necessary to determine the absolute concentration of the elements present. This is not one of LIBSâ strengths since the complexity of the plasma plume makes the relationship between measured intensity and elemental concentration indirect. The first step in quantitative spectroscopy is to determine the temperature of the plasma plume. This can be accomplished by comparing the comparing the ratio of two emission lines for the same atomic species and using the following equation.
(2)
1 0 1 (^1 2 ) /
2 0 2
( )
( )
ji j E (^) j E (^) j kT
ji j
I f A g e I f A g
â â
Where fo is the center frequency of the emission line. Aij is the Einstein coefficient of radiative transition probability and gj is the degeneracy of the upper energy level. E1j and E2j are the jth energy levels for each species measured with respect to ground ad k is the Boltzmann constant. The spectral lines chosen for this calibration must correspond to elements for which these values, or ratios, are known or can be measured. Once temperature is determined the following equation can be used to find the ratio of two elements. (3)
0 (^ ) /
0
a ji^ j^ a b a E aj^ Ebj^ kT
b ji j b a b
I f A g Q N
e
I f A g Q N
â â
Were Na and Nb are the number densities of the elemental species of interest. Qa and Qb are the partition functions for the ionized species. The ratio (4)
ji j a (^) b
ji j b a
f A g Q
f A g Q
can be determined by calibrating the system using a plasma of known composition. A correction for the temperature needs to be made using the definition of the partition function. (5) Es / kT s s
Q g e
â
Theses calculations only work if the plasma is optically thin. A plasma is considered optically thin if the product of the absorption coefficient and the depth of the plasma is less than 1. It is also assumed that the plasma has the same elemental composition as the sample which is dependent on choosing the proper laser parameters.
Examples
LIBS in Femto Second Laser Micromachining
Micromachining control is an application of LIBS that takes advantage of the on-line analysis capability. Laser micromachining makes micro meter scale cuts, grooves, or
In Tong's experiment LIBS feedback is only used to control the pulse rate and hence the cutting depth. The positioning of the sample is independent and preprogrammed. That is, the translation stage moves the sample in the laser beam in the pattern that is to be cut. The default state of the laser is 1kHz pulses, fast enough to ablate the top layer of the substrate. The LIBS emission spectrum is monitored continuously throughout the cutting. When a spectrum is detected that indicates the second layer of the substrate has been reached, the repetition rate of the laser is reduced to 100Hz. This rate has been determined to be suitable for continuing to monitor the composition of the material without doing much damage. When the LIBS spectrum indicates that the first layer is again in the focus of the laser the repeat rate returns to 1kHz.
LIBS and Art Restoration
LIBSâ capacity to analyze a sample using very small amounts of material and with minimal damage to surrounding areas is exploited in the field of art restoration. Painted artworks accumulate dirt and pollutants on their surface over time. They need to be cleaned to restore their original appearance and to prolong their life. Some time dirt is not the only material that needs to removed. Sometimes a second painting has been painted on top of the original. Art restorationists want to remove the top layer of paint without damaging the painting below. Traditional restoration methods involve chemical solvents or abrasives. These techniques are difficult to control and without an experienced practitioner they may damage the artwork. LIBS has the potential to provide an automated system for efficient art restoration that is less dependant on individual skill.
The surface of a painting is similar to the substrate used in micromachining. Instead of layers of metals the painting has layers of canvas, pigment, varnish, glaze and contaminates. Unlike the materials used in micromachining, these layers and not uniform. The paint may be intentionally uneven to give the picture texture. The varnish and glaze may crack over time and the resulting gaps fill up with dirt. LIBS based cleaning methods need to selectively ablate only the contaminates. Aged varnish is also removed. However, the laser can do damage to the pigment itself so a thin layer of varnish is left intact.
S. Gregiou et al. describe a technique used to restore an 18th century oil painting that had accumulated dirt and pollutants on its surface. The painting is mounted on an positioning stage that moves in three dimensions. An excimer laser is focused onto the painting. The optimal laser parameters for this particular painting are determined by tested on a representative area. A wavelength of 248nm has been found to work well for oil painting. Varnish absorbs highly at this wavelength so penetrations is minimal. Higher wavelength heat the surrounding pigment and can cause damage. A fluence of 200- mJ/cm^2 is used. With these parameters optimized there appears to be no significant harm done to the painting.
Sketch of a Proposal
The LIBS controlled ultrafast laser micromachining could be further improved by controlling the stage position as well as the pulse rate. In the system designed by Tong it is possible for the stage to move the sample in such a way that some of the top layer would be inadvertently left intact. Tongâs test of the system only looked for problems with cutting too deep or too far, not problems with removing to little material. Positioning the stage based on the LIBS would lessen the potential for damage to be done by the monitoring pulse as well.
This LIBS micromachining system would respond to a change in the composition of the material being ablated in two ways. First, the monitoring laser would be turned off completely. Second, the translation stage would move the sample a distance corresponding to the beam diameter. The laser would then b turned back on at the monitoring pulse level. If the spectrum indicated that the top level was intact this location the laser would switch to the cutting rate. If the top layer had already been ablated the stage would move again. The pattern to be cut would be preprogramed into the system but the rate of the cutting would be determined as it went along. Potentially, this system could be more accurate but would take longer.
References
M. Coris, G. Cristoforetti, M. Hidalgo, S. Legnaioli, P. Vincenzo, A. Salvetti, E. Tongoni, and C. Vallebona. (2003) Applied Optics. 42, 6133 (2003)
C. R. Dockery and S. R. Goode. âLaser-Induced Breakdown Spectroscopy for the Detection of GunshotResidues of the Hands of a Shooter,â Applied Optics. 42, 6153 (2003)
Y. Kim. âFundamentals of Analysis of Solids by Laser-Produced Plasams,â 327 In L. Radziemski and D. Cremers (ed.) Applications of laser-induced plasmas. Marcel Dekker, (New York, 1989)
P. Fichet, D. Menut, R. Brennetot, E. Vors, and A. Rivoallan. âAnalysis by Laser- Induced Breakdown Spectroscopy of Complex Solids, Liquids, and Powders with an Echelle Spectrometer,â Applied Optics. 42, 602 (2003)
S. Georgio, V. Zafiropulos, D. Anglos, C. Balas, V. Tornari and C. Fotakis. âExcimer Laser Restoration of Painted artworks: Procedures Mechanisms and Effects,â Applied Surface Science. 127, 738(1998)
D. Michaud, E. Proulx, J. Chartrand, and L. Barrette. âShooting Slurries with Laser- Induced Breakdown Spectroscopy: Sampling is the Name of the Game,â Applied Optics. 42, 6179 (2003)
