Analysis of the State of the Art: LIBS

In honor of Spectroscopy's celebration of 30 years covering the latest developments in materials analysis, we asked a panel of experts to assess the current state of the art of laser-induced breakdown spectroscopy (LIBS), and to try to predict how technology will develop in the future.

Laser-induced breakdown spectroscopy (LIBS) is a type of atomic emission spectroscopy that uses a highly energetic laser pulse as the excitation source (1). Although the spectral analysis of laser-produced spark emission dates back to the invention of the pulsed laser in 1960, the pursuit of this phenomenon for analytical chemistry applications and the coining of the term "LIBS" did not happen until the early 1980s (2). We asked a panel of experts to share their views about the current state of LIBS, including recent and expected future developments, ongoing challenges, and the applications where it can be used. This article is part of a special group of six articles covering the state of the art of key techniques, also including inductively coupled plasma–mass spectrometry (ICP-MS), X-ray fluorescence (XRF) spectroscopy, infrared (IR) spectroscopy, near-infrared (NIR) spectroscopy, and Raman spectroscopy.

Recent Advances

We started by asking our panelists what they considered the most important advance in LIBS or its instrumentation in the last 5–10 years. Vassilia Zorba, a staff scientist at Lawrence Berkeley National Laboratory, said it's hard to pin down only one important advance in LIBS during the past decade. "Significant progress has been made in instrument components, data processing algorithms, and extending its capabilities to isotopic analysis," she said.

Other panelists agreed that quite a few developments had come together in recent years. "The combination of the improvement of LIBS hardware, including the three major components of LIBS setup, the deeper understanding to the mechanism of laser-induced plasma and its emission, as well as the development of quantitative modeling, are the most important advancements," said Zhe Wang, an associate professor at Tsinghua University

"Commercially available LIBS instruments for many market applications are now available as a result of a perfect storm," said Rick Russo, a senior scientist at Lawrence Berkeley National Laboratory and president and CEO of Applied Spectra. "This storm includes advances in fundamental science, lasers, detectors, data analysis algorithms, and the success of ChemCam."

ChemCam, the LIBS instrument currently exploring Mars on the Curiosity rover, was top of mind for several other panelists, as well. "Integrating a LIBS system into the Mars Rover would have to rank on the list of milestones in this field," said Ishan Barman, an assistant professor in the Departments of Mechanical Engineering and Oncology at Johns Hopkins University.

Vincent Motto-Ros, an associate professor at the Light and Matter Institute at Lyon University, said ChemCam exemplifies the advances that have been made in LIBS instrumentation, particularly to develop systems adapted for specific applications. "A lot of effort has been made in the last few years in the development of compact (handheld) designs, LIBS platforms for laboratory analysis, and LIBS systems dedicated to operate in harsh conditions," said Motto-Ros.

Barman was also impressed by the development of handheld devices. "A major advance, which has happened gradually, has been the translation of a large, unwieldy laboratory bench setup into a portable, briefcase-sized device that can be deployed in a variety of field applications," said Barman.

Several panelists remarked on the importance of data analysis and computerized chemometric techniques for rapid sample identification. "The integration of that chemometric software with the analytic hardware has pushed the technique to true real-time analysis," said Steven J. Rehse, an associate professor in the Department of Physics at the University of Windsor."We can now objectively and rapidly identify, sort, classify, or discriminate a wide variety of targets ranging from hazardous materials to gemstones, geological samples, alloys, and pathogens on the basis of a LIBS spectrum that is acquired in under 1 second."

Zorba agreed. "Chemometric analysis has gone from limited to routine use in LIBS in the last decade, and will continue to be a significant step toward data interpretation and classification," she said. "Recent developments in calibration-free LIBS have also been an encouraging step toward eliminating the need for calibration standards."
Other panelists said that the most important advances were in lasers. Zorba explained that femtosecond lasers have been increasingly used in LIBS during the past 10 years, thanks to the development of more stable and user-friendly systems.


Matthieu Baudelet, a research assistant professor at the University of Central Florida, said that the compactness of the new diode pumped sources allow for high repetition rate analysis (100+ Hz) for both laboratory instrumentation and handheld systems with the very compact low energy kilohertz systems. "Fiber lasers are showing a great stability and ruggedness that allows for robust implementation in harsh environments, such as conveyor belts in mining and sorting facilities," he said.


Zorba added that the development of laser ablation molecular isotopic spectrometry (LAMIS) stands out as a significant new   technology for extending LIBS from a largely elemental analysis tool to a real-time isotopic one.

Key Challenges and How to Solve Them

Next, we asked our panelists about some of the key challenges or questions that spectroscopists developing or using LIBS currently face, in terms of current limitations of LIBS, limits to the understanding of how it really works, or difficulties using it (such as for specific types of application). We also asked which of those challenges are likely to be addressed in the coming years and how progress might be made.

Challenges

Motto-Ros said that LIBS is subject to comparison with other analytical techniques such as X-ray fluorescence (XRF) or inductively coupled plasma (ICP)-based technologies and that such comparisons tend to slow down adoption. "An observation that I find very disturbing is that despite more than 50 years of existence, it is still rare to find a LIBS instrument used for routine analyses in analytical laboratories," he said. "We have to change that."

Richard R. Hark, the H. George Foster professor in the Department of Chemistry at Juniata College and Chief Scientist at Synergos Global Security, LLC, cited the need to improve our understanding of the factors that affect the physics of ablation and the chemistry that occurs in the plasma that is subsequently formed. "Such an understanding will enhance the utility of the LIBS method for qualitative and quantitative analysis," he said.

Rehse agreed. "As a physicist, the inability to predict and calculate from first principles what the plasma emission spectrum will be from an arbitrary analyte atom in an arbitrary medium (commonly called the matrix) remains a problem," he said. Motto-Ros echoed the point. "Efforts must continue on fundamental investigations of the laser-induced plasma," he said.

Zorba agreed, noting that several aspects of the laser-material interaction remain unknown. "Particularly in the femtosecond-time regime, nonlinear processes (such as multiphoton absorption) and the distinction between thermal and nonthermal processes hamper an a priori prediction of optical and thermal diffusion lengths," she said. "These parameters are critical toward defining laser sampling parameters (ablated material mass), plasma conditions, and resulting LIBS emission."

A closely related problem is the so-called "matrix effect"-the fact that the signal from a specific analyte atom can depend on the sample matrix. "Spectroscopists who use LIBS know that parameter optimization for specific analytes in a particular matrix is both challenging and essential," added Hark.

"The matrix effect makes the preparation of high-quality standards important," said Rehse. "It also makes it more difficult to sample highly heterogeneous samples like complex minerals or complicated powders like dirt or soil."

Wang noted that for practical quantitative analysis, the most important issue is to provide consistent analytical results. "Because of uncertain laser-sample interaction, unstable plasma and surrounding gas interaction, the inhomogeneous spatial distribution, and uncontrolled temporal revolution, it is very difficult for LIBS to provide such a stable spectral signal even using multi-pulse analysis and averaging a large number of spectra to remove the random noise," he said. "Reducing the variation between groups from the same sample is the overwhelming important issue that must be solved for LIBS to realize quantitative analysis and large-scale commercialization."

Hark also noted that reproducibility between instruments is another potential challenge. "Unlike other analytical techniques like Fourier transform infrared (FT-IR) or UV–visible spectroscopy, LIBS spectra obtained on different instruments using the same experimental parameters are not necessarily identical," said Hark. "Further work has to be done to make LIBS spectra interchangeable."

Rehse cited another aspect of variability in LIBS measurements. "The pulse-to-pulse variation in plasma properties caused by laser shot repeatability issues is another challenge," he said.

Russo feels the concerns about matrix effects and lack of reproducibility are overblown. "Like all analytical methodologies, LIBS requires standards for quantitative analysis," he said, adding that so-called "standardless" LIBS calibration methods eventually will be developed to meet specific applications. "And similar to any atomic spectroscopic approach, knowledge of spectral interferences and matrix effects is important for getting the best performance."

He acknowledged that getting good analytical performance requires knowledge of the plasma temporal and spatial properties, as well as of the sample, and that a set of parameters has to be established for a particular sample material to give optimum performance. "Once the optimum method is established, precision and accuracy are repeatable and reliable," he said.

Barman suggested that rather than fretting over the reproducibility problem, we should not solely focus on trying to use LIBS for quantitative studies. "An alternate, and better, route would be to focus on the development of LIBS primarily as a robust screening tool," he said. "As a screening tool, the relative variations-due to laser fluctuations, environmental and matrix effects, and self-absorption-would cease to have as much of an impact."

Motto-Ros also feels that the sensitivity and accuracy of LIBS should be improved. Russo, in turn, believes there is a misconception regarding sensitivity and precision in LIBS. It's true that limits of detection (LODs) and precision for LIBS are generally considered poor compared to ICP-OES, he said, but on an absolute level, LIBS is very sensitive. "One of the primary reasons why relative LODs are not at the same level as ICP-OES, for example, is that LIBS only analyzes sub-microgram quantities of the sample," Russo stated. "However, the analyst can easily increase the amount of mass analyzed by raster scanning the laser over the sample. There are very few cases where sample quantity is limited." Thus, when comparing LIBS LOD and precision to ICP-OES, it is important to consider the actual mass being analyzed, he concluded.

In terms of instrumentation, a key challenge is to develop better compact, high-pulse-energy laser sources that are suitable for LIBS use outside the laboratory, said Hark. "Small, relatively low-cost lasers with high beam quality capable of delivering multiple pulses of suitable energy are ideal for portable LIBS operation," he said. "As the components of LIBS systems are further miniaturized and ruggedized, the range of applications will consequently expand," Hark said.

Zorba also considered how to expand the applicability of the technique. "Chemical imaging with extreme resolution is required for a variety of new applications, especially those related to energy research (generation, conversion, storage)," she noted. "LIBS is largely a macro- to meso-scale tool, and detecting LIBS emission signals from limited amounts of ablated mass becomes challenging as we scale down toward nanoscale LIBS."

From a market standpoint, Russo commented that it is important to get LIBS knowledge out to the public, so that people are aware that LIBS can provide a good solution for them. "This message is nowhere near where it needs to be," he said. "We have only scratched the surface."

Solutions

According to Motto-Ros, finding answers to these challenges will not be easy, but several research teams are working on them. "For example, Alessandro De Giacomo's group in Bari, Italy, has recently proposed a simple method (called 'nanoparticle-enhanced LIBS,' or NELIBS) to significantly improve sensitivity," he said. Motto-Ros also mentioned recent developments at his own institute that clearly show that measurement repeatability and reproducibility can be greatly improved by accurately controlling experimental parameters. Also, several groups are working on advanced algorithms that would greatly facilitate LIBS data analysis, he said. "But at some point, all of these research developments must be regrouped and integrated in a system that will then need to be validated and certified through a quality program," Motto-Ros said. "This will certainly take time."

Barman agreed that progress has already been made on a number of the technical challenges outlined above. He mentioned research by his and other groups to explore the feasibility of nongated LIBS to gain significant cost and size benefits-despite a modest reduction in overall performance. "I believe that some of the progress will also have to be application-specific," he said. "In other words, LIBS instrumentation for biomarker diagnostics in 10 years may look vastly different from a handheld system used for crime-scene investigations."

Hark is optimistic that progress will be made to solve all of these problems in the next decade. "A significant amount of work has already been done with mathematical modeling to comprehend the complex processes that occur during the formation and analysis of plasma," he said. Hark also pointed to work that has been done with calibration-free LIBS quantitative analysis, which is based on the measurement of spectral line intensities and certain plasma properties. "As the connection between instrumental characteristics, experimental parameters, and spectral output is better understood, the calibration-free LIBS approach can be refined to deal with many of the problems in LIBS," Hark stated.

Rehse also sees work progressing on both of these areas. "The theory behind the plasma emission continues to develop," he said. "And the advent of calibration-free LIBS suggests the possibility of obtaining high-accuracy analyte measurements from even the most complicated matrices without requiring any sort of standard by using self-consistent calculations based on known atomic emission parameters."

Lasers are also always getting better, Rehse added. "It is already well-known that the femtosecond lasers produce much higher quality plasma due to a more controlled ablation process," he said. "Unfortunately, the current cost of femtosecond lasers, and their lack of portability and ruggedness, probably precludes their widespread adoption in the field or in industry." However, newer lasers, with a more tightly controlled spatial mode and shorter pulse durations can overcome this problem, Rehse explained.

Zorba also anticipates advances that will broaden the scope of LIBS use. "Progress toward extending the range of new applications, including next-generation energy research, is likely to be significant in the next 5 years," she said. "This will be aided by improvements in spatial and axial resolution, chemical imaging, and visualization capabilities."

Russo agreed. "The drive to make things smaller, cheaper, and faster will lead to a new development that will change the field," he said-just as occurred with other analytical spectroscopy instruments like ICP-OES and graphite furnace atomic absorption systems.

Finally, Wang believes that the further development of LIBS is largely dependent on the success of its large-scale commercialization, and that this can be accomplished in next 5 years. "To realize large-scale commercialization, LIBS still needs to prove to customers the technical feasibility of using LIBS to realize consistent analyses," he said. "The combination of the improvement of hardware and data processing methods will make this happen," Wang added. "Once one or two commercial applications for LIBS are successful in realizing on-line or at-line measurement, the cost of LIBS will decrease and LIBS will be deployed more broadly."

New Developments

Our panelists also foresee a number of other new developments in LIBS in the coming years. These include handheld units, improved spectrometers and charge-coupled device (CCD) detectors, multisensor modalities for fast LIBS sensing, increased undergraduate teaching using LIBS, and application-specific uses such as biologics.

Rehse believes handheld LIBS units will become more and more powerful and broadly applied in the coming years. He mentioned that two handheld LIBS units were presented at Pittcon 2015 that offered very attractive features like long battery lifetimes, on-board high-resolution imaging, custom-developed chemometric packages for specific applications, and easy-to-understand user-interfaces. "I would expect that these units will improve rapidly and dramatically as future generations are developed," he said.

Similarly, Baudelet said he hopes that the miniaturized spectrometers, especially those based on static Fourier transform, will find their way into the LIBS community, allowing for easily adaptable spectral ranges, resolutions, and sensitivities. Russo agreed that further improvement in spectrometers and CCD detectors is foreseeable, and will result in overall enhancement of LIBS instrumentation.

Russo also believes that using LIBS in combination with ICP and LAMIS will advance LIBS capabilities. "By using LIBS with ICP-MS, for example, the dynamic range is extended for all elements," he said. "Major and light elements (such as H, N, C, and F) can be measured simultaneously with the heavy trace elements and isotopes by the ICP-MS system." LAMIS, he explained, extends the power of LIBS by allowing light isotope measurements at the same time as the full periodic table of elements. "With the LAMIS capability, isotope ratios already can be measured with per-mil precision," he said.

Motto-Ros, in turn, believes that fast LIBS elemental imaging is going to be a strong development in the coming years. "Even though high-speed LIBS mapping was demonstrated more than 10 years ago, an increasing number of research groups and companies are working on its development, with the field of applications spreading from industrial analysis to biological tissue imaging," he said.

Hark is also pleased to see that the LIBS technique is beginning to be introduced in the undergraduate science curriculum. "The use of LIBS for pedagogical purposes will be a significant development in the near future," he said. "The introduction of LIBS instruments specifically targeted for use in the undergraduate college or university laboratory will not only provide educators with a new and powerful tool for teaching, but will also introduce this technology to a worldwide audience."

Zorba expects to see various instrumentation improvements, including new spectrometer designs (such as spatial heterodyne schemes), low energy–high repetition rate scanning sampling methods for LIBS bulk analysis, and improvements in resolution that will transform LIBS into a true three-dimensional chemical imaging instrument. "These advances will be important for introducing LIBS into new and developing fields of advanced material applications," she said.

Application Areas and Challenges

Continuing with our discussion on the application areas for LIBS, we asked our panel what the most important areas of application are now, what new areas are emerging, and how well LIBS competes with other methods in those new areas.

Hark noted that LIBS has found broad application as a tool for materials analysis in a variety of industries ranging from recycling of scrap metals and plastics to analysis of coal, foundry metals, and electronic components. As for emerging areas in which LIBS will likely have an impact, he sees promise in biomedical testing, forensics analysis, identification of hazardous materials, and examination of geological and environmental samples. "LIBS analysis will continue to expand in laboratory and industrial settings, but as field analysis applications mature, LIBS will be a superior single instrument, and uniquely capable when fused with other sensor modalities," he predicted.

Russo agreed that LIBS is gaining acceptance in many industrial applications and continues to be a strong research emphasis around the world. "Geology, ecology, forensics, pharmaceuticals, semiconductors, consumer electronics, metallurgy, mining, photovoltaics, and battery applications are seeing good returns using this technology," he said. He added that new areas exist for standoff measurements and real-time continuous process monitoring in industry, such as in raw-material or product screening for impurities and contamination.

Rehse is a big proponent of using LIBS in medical applications, which is the primary focus of his research. "I believe that the trends in 21st century medicine to perform faster, cheaper, and less-invasive procedures perfectly complement the advantages that LIBS brings to the table," he said. Just as industry has seen that this technique can be used for real-time quality assurance and process monitoring, he envisions LIBS assays being used in a host of medical and biomedical procedures for rapid tests performed at the "point of contact" -in a doctor's office or clinic. "In many of the areas that I am describing there are no other methods being used at all, and that is where LIBS could make an immediate impact," he said.

Rehse mentioned two examples of such applications. The first was related to how lasers are already being used to "drill" cavities in teeth prior to filling. It has been shown that the LIBS emission from such a tooth can be used to unequivocally differentiate healthy tooth tissue from the cavity (the carious tissue), Rehse explained. "Yet as far as I know, this technology has not been adopted, and so instead a dentist uses no technology and just goes by training and experience to decide how much tissue to remove," he said.

The second example is the use of LIBS emission during transcranial surgery to monitor how deeply into the skull the surgeons are cutting (3,4). "They have been to leave an extremely thin (basically transparent) but intact 'window' of skull bone over the brain and perform optical procedures in the brain through that window without ever exposing it to the environment," he said.

In Rehse's own field of microbiological identification, he said, the speed of LIBS tops all the competing techniques that are based on genetic identification, serological (antibody-based) identification, or other principles. "LIBS will probably never perform as well as the gold-standard culturing techniques, and it may not be as sensitive as a genetic 'real-time' polymerase chain reaction (PCR) test, but it is fast enough to begin directing treatment immediately, suggest other tests that could be run, or even just confirm whether or not bacteria are present-in a simple 'yes or no' test that does not require true identification," he said.

Motto-Ros agreed that biology and medicine are emerging application areas for LIBS. "We have been working for several years in my institute on the development of LIBS for elemental imaging of biological tissue," he said. "With its all-optical design and its parts-per-million scale of sensitivity, LIBS imaging is unique among methods of elemental detection because of its full compatibility with standard optical microscopy." Motto-Ros explained that the biomedical community is very enthusiastic about the imaging capability of LIBS because current methods for the observation of trace elements in tissues at microscopic spatial resolution often require complex equipment with restricted access.

Also in the medical application area, Barman said that an intriguing route to pursue would be the engineering of LIBS tags for biomarker characterization. "Next-generation cancer diagnostics, particularly in the form of 'liquid biopsy,' represents a huge opportunity where LIBS can make a substantive difference," he said. Nevertheless, Barman explained, a number of the biomarkers (notably tumor antigens shed into the vasculature) may not be detected with high specificity using only LIBS-based elemental analysis. "Incorporating biomarker-specific tags (with suitable recognition moieties and LIBS-active elements) may allow LIBS to penetrate into hitherto unexplored territory," he concluded.

Another emerging area of application for LIBS is energy applications, explained Zorba. "The application of LIBS to electrochemical energy storage technologies such as lithium-ion batteries will be particularly important because of its sensitivity to lithium and its spatially resolved and depth-resolved chemical imaging capabilities," she said. "These features will bring a new dimension to chemical analysis that is lacking in energy dispersive X-ray spectroscopy and ICP-OES, which are the commonly used techniques in lithium-ion battery research currently."

In China, coal, metallurgy, and cement applications have a very large market in LIBS, Wang said. "Environmental protection is also an emerging application, but right now LIBS is not as good as XRF," he added.

Challenges

Next, we asked our panelists what challenges must be overcome for these new application areas to work and what progress has been made so far to achieve those goals.

Some of these challenges are the same as the overall challenges facing LIBS. For the energy field, for example, quantification is still a challenge because of the lack of matrix-matched standards for many energy-related systems, which may be inherently complex. "Calibration-free approaches will be very helpful toward advancing LIBS implementation in this field," Zorba said.

A few panelists mentioned the difficulty in getting users to adopt new methods using LIBS. In this context, Russo said, LIBS experts need to support the early adopters. "Educating these users properly and assisting them in proper method development is critical, but it's a time consuming and slow uphill climb," he said. "As industrial customers establish successful protocols, LIBS will be viewed as the standard and can replace acid dissolution in many cases."

Rehse had similar comments. "As with most technologies in medicine, doctors and other medical practitioners are sometimes slow adopters of new technology, because they have to be," he said. "One does not adopt new technology lightly when human health is being considered." Rehse also explained that the Food and Drug Administration needs to approve all such new tests or treatments, which is time consuming and expensive. In addition, a large amount of peer-reviewed literature needs to be present before a critical mass of medical practitioners will accept the validity of a technique. "I would like to say that progress is being made to introduce the technique to the field, but I have not seen any significant inroads yet," he said. "Individual physicians, pathologists, and practitioners I have talked to have all been deeply impressed by what LIBS could do for them, but the lack of well-funded research programs investigating this concept is limiting its widespread dissemination."

To increase user adoption, Baudelet said, researchers must provide very robust data that address the concerns of fields such as geology, manufacturing control, forensic science, and biomedical imaging. "LIBS cannot be adopted just for its own advantages," he said. "It needs to speak the same language as the other elemental techniques already used in the field." Baudelet said that research is being done to address these issues and provide more information from LIBS data, such as elemental maps and error rates, beyond just spectra and the results of statistical tests.

Hark said that field use of LIBS is in its early stages as well, but significant progress has been made, as evidenced by the recent introduction of several commercial handheld LIBS units. "Improved performance will come as more sophisticated systems with more powerful compact laser sources are developed," he said.

Mars: A New Beginning for LIBS?

The Curiosity mission has shown that laser spectroscopy can be operational and successful in very harsh conditions. We asked our panelists if this represents the beginning of a new generation for LIBS.

From Zorba's perspective, the Curiosity mission was less the beginning of a new generation for LIBS and more a platform to educate the public and analysts on the capabilities of this technique, while eliminating the misperception that LIBS is solely a laboratory tool.

Russo agreed. "I don't see this as the 'beginning of a new generation,' but as further evidence that by developing a fit-for-purpose instrument, LIBS can address complex analytical applications," he said.

Rehse believes that the Mars mission helps inform people outside the LIBS community of what LIBS can do. "When people hear that the LIBS unit on the ChemCam has obtained over 2000 data measurements, while some packages have not even taken more than a few dozen during their entire time on the surface of Mars, that immediately grabs people and demonstrates the utility of LIBS," he said

Motto-Ros feels the same way. "ChemCam has a strong educational role as it has become easy to introduce and explain what LIBS can do," he said. "I do several LIBS lectures at my university and I can tell that more and more students wish to become involved in LIBS."

"ChemCam represents a significant home run for the LIBS technology," said Hark. He agreed with our panelists that it significantly raised the visibility and interest in LIBS. "The use of LIBS for deep-sea 'inner space' investigation also captures the imagination and has the potential to spur interest in the method as well," he said.

Baudelet said that another useful aspect of the Curiosity mission is that it suggests the value of sharing LIBS spectra for analysis by any group. "This philosophy of sharing data is really a great way to have research groups that don't have instrumentation get involved in the actual research," Baudelet said. "This could become a new approach to handle LIBS data and develop education in LIBS."

LIBS and Multivariate Analysis

Over the last 10 years, multivariate analysis has become a common part of LIBS data analysis. We asked our panelists the following question: Do you think multivariate analysis helps in the development of the technique itself or does it just help many applications without contributing to the development of general protocols?

Russo explained that multivariate analysis is a valuable tool for data analysis because all of the information in the LIBS spectrum is used for quantitative analysis, which allows matrix effects to be taken into account in calibrations. These unique spectra also can be used for identification purposes, allowing a user to quickly analyze samples and group them into specific sample types without knowing much about the samples, he said. Russo added that the entire optical spectrum has a tremendous amount of data that define a sample. "For simple applications where standards exist and the sample type within an industry does not change significantly, routine analysis can be performed using simple 'area-under-the curve' processing. However, for classification and provenance, the entire spectrum represents a fingerprint or barcode of the material," he said. "Multivariate analysis makes sense as a powerful approach to utilizing all the information in the sample and even helping to understand the LIBS process."

Barman said that multivariate analysis works well for certain applications. "I would say that for specific applications, the use of chemometrics has been integral to the overall development-for example, in suggesting the use of selected spectral bands, which could potentially be implemented without using a conventional spectrometer and gated CCD," he said. "In other cases, the continued trend of the blind use of statistical tools, without a deep understanding of the underlying mechanisms or independent validation, has possibly thrown up spurious results."

Baudelet agreed. He said that multivariate analysis has made a lot of researchers lazy and forget the basics of emission spectroscopy. "A lot of new published papers are full of misidentified spectral lines, leading to erroneous breakthroughs," he said. "The atomic emission spectroscopy (AES) bases should be understood to provide the right input to algorithms."

Rehse agreed that multivariate analysis helps with the development of the technique, in the sense that it has enabled measurements to be made in systems where no good LIBS measurement could previously be made. "In systems like powders or residues on substrates, the ability to perform classification has opened entirely new ways of thinking about the problem of trace contaminants on complicated backgrounds," he said. "Suddenly, people started envisioning applications that hadn't really been thought of before: rapidly identifying 'conflict minerals' or glasses at crime scenes, for example." Rehse explained that this led to whole new ways of sample mounting and preparation and has led to a deeper interaction between statisticians and physical scientists, which he finds very rewarding.

Motto-Ros said that multivariate data processing is clearly an extremely efficient way to perform LIBS spectrum analysis, but it should not replace the work that must be done upstream to ensure the quality of the measurement. "A quote from Mohamad Sabsabi perfectly sums up my feelings about multivariate analysis," he said. '''The best chef in the world cannot make a good soup with rotten fish.'"

Future Improvements in Laboratory-Based Instruments

Next, we asked our panelists how LIBS instruments could and should be improved for use in analytical laboratories.

Motto-Ros said that this was precisely one of the questions that guided his group's research activities over the last three years and that repeatability and reproducibility have been a major focus. "It is one thing to acquire a LIBS spectrum one day, but it is another thing to repeat measurements day after day keeping the signal stable," he said. A key, he explained, lies in fine control of experimental parameters. "Basic control points include the laser pulse energy, the laser focus position with regard to the sample surface, and the position of the light collection with regard to the plasma shape," he said. Motto-Ros recently demonstrated that using these three basic controls, it was possible to obtain stable measurements over several months without any signal normalization. "This promising result should now be confirmed by further studies, but I think we are on the right track," he said.

Wang named similar concerns. "We need a stable system including stable laser energy, laser-material interaction, and consistent signal collection," he said. "In addition, it would be good to have a larger ablation mass."

Baudelet said he would like to see the resolution of miniature spectrometers improved in the future.

Russo came back to the requirement for matrix-matched standards, noting that it is a key factor in analytical laboratories, which handle a variety of sample types. "The majority of LIBS measurements are performed with a nanosecond laser and it is known that the matrix-matching requirement can be relaxed with shorter laser pulses," he said. "The availability of shorter pulse lasers with affordable price tags would relax the matrix dependence as well as provide for nanometer spatial resolution."

Zorba added that wider implementation would be accelerated by a number of features, including fast laser sampling and data acquisition automation, on-demand beam focusing and homogenization technologies, improvements in spatial resolution, and more user-friendly interfaces. "Advancements in compact laser and spectrometer designs, including improved sensitivity, spectral resolution, and wavelength coverage, will also be important for reducing the instrument footprint and for making the technique more attractive to users," she said.

Still others mentioned high definition or hyperspectral imaging for LIBS as a key area to focus on. Barman posed the following question: Can we rapidly image a tissue section to provide information similar to histological analysis? "Combining the elemental basis of LIBS with the imaging capabilities of microscopy should be at the forefront of future instrumentation development," he said.

Rehse had similar thoughts regarding combining LIBS the with hyperspectral imaging. "Combining the atomic information of LIBS with molecular information gives a multidimensional assay that may have great utility," he said.

Rehse also mentioned the high spatial resolution (10 µm or better) of LIBS as another significant advantage that could be utilized in the future. "A good number of LIBS users can now make two-dimensional elemental maps by rapidly rastering the sample or the laser spot, and if repeated shots are fired in one location, the third dimension of depth can be added and full 3D elemental information from a sample can be obtained with a 10-µm resolution," he said. That requires hardware improvements, he said, so that it is possible to quickly obtain these data while rapidly rastering the sample, and then quickly calculate, plot, and display such information to the end user in an easy to interpret 3D elemental map using color-coding. "This is definitely achievable," he said.

The Future of Handheld LIBS

Finally, we asked our experts what they foresee for the future adoption of handheld LIBS.

Baudelet was fairly skeptical about the importance of handheld LIBS. "The fields that are concerned by this application are few, such as mining and sorting," he said. "And the advantages compared to XRF need to be worth the investment."

Russo had a similar opinion, explaining that the challenges for handheld LIBS are many. Good results require the right components and repeatable sample presentation, he said, but for handheld LIBS, the choice of components-the laser and detector-is limited by availability and the variations in sample presentation can be extensive. In addition, the laser spark needs to be robust enough to provide sampling and analysis, and low laser energy generates a very tiny spark, with higher irreproducibility. "There are many variables that make reliable and consistent results difficult with handheld LIBS," he said. "Laboratory systems are successful because lasers and detectors can be selected to fit the application."

As a result, in his view, handheld instruments are currently only beneficial for qualitative applications, for which there are only a few, such as in-field screening. "With a significant investment in laser and detector development, additional field applications can be viable," he concluded.

Zorba agreed that handheld LIBS is currently limited by the lack of compact high performance instrument components such as lasers and spectrometers. "Advances in component development toward approaching the performance of lab instrumentation will be critical to making this technology viable for a variety of applications," she said. "The development of protocols for reproducible sampling and data interpretation will also be very important for future adoption."

Despite current obstacles, other panelists are excited about the future for handheld LIBS. "Handheld LIBS devices have tremendous potential," said Hark. "Given the very wide range of materials measurement needs in the field, handheld LIBS is expected to make an immediate impact that will continue to grow in the future." He added that as people become aware of what handheld LIBS can do compared to other handheld analytical tools, new applications will emerge.

Rehse said that the handheld units that exist are already amazing, given the fact that they have only recently appeared on the market. "I fully expect that they will start to be adopted in the field of emergency response, security, compliance, and quality assurance," he said. "I suspect a lot of future users are not even aware right now that they will become users because they don't even have it in their minds that they could be exponentially increasing the amount of product or item sampling they are performing while doing so in a fraction of the time."

Using LIBS for quality control, he said, would be a paradigm shift, from a scenario where samples go to a laboratory for testing, with results provided a day later, to one where everything can be checked instantly. "If you can hold the 'test' in your hand and in under 5 seconds walk up and down a line sampling everything virtually nondestructively, why wouldn't you?" Rehse said. "This opens up a whole new level of control and information that, as I said, has probably not even occurred to many people yet."

Conclusion

LIBS has developed significantly in the last decade, and the technique has received greater attention. LIBS has even been successfully deployed on Mars. LIBS still faces some important challenges, however, particularly related to the matrix effect, a need for a better fundamental understanding of the laser ablation process, and the lack of reproducibility between measurements. But LIBS experts see ongoing progress in addressing these challenges, as well as excellent potential for new applications. They are confident that as people learn about what LIBS can do, adoption will increase.

Acknowledgments

I would like to extend a special thank you to Matthieu Baudelet for his assistance with the questions and panelist recommendations. I would also like to thank Andrzej Miziolek and Jose Almirall for their additional assistance with panelist recommendations.

References

(1) http://en.wikipedia.org/wiki/Laser-induced_breakdown_spectroscopy

(2) Written communication with R. Hark, 2015.

(3) P. Tsai, D. Jeong, and D. Kleinfeld, "Precise Cranial Surgery with Femtosecond Laser Ablation, Laser Induced Breakdown Spectroscopy and Second Harmonic Generation," presented at the 2013 North American Society for Laser-Induced Breakdown Spectroscopy (NASLIBS) conference, Milwaukee, Wisconsin, 2013.

(4) D.C. Jeong, P.S. Tsai, and D. Kleinfeld, Curr. Opin. Neurobiol. 22, 24 (2012).

Megan L'Heureux is the managing editor of Spectroscopy magazine.