Lasers in Medical Science

, Volume 26, Issue 5, pp 673–687

Prospects for laser-induced breakdown spectroscopy for biomedical applications: a review


  • Vivek Kumar Singh
    • School of PhysicsShri Mata Vaishno Devi University
    • Laser Spectroscopy Research Laboratory, Department of PhysicsUniversity of Allahabad
Review Article

DOI: 10.1007/s10103-011-0921-2

Cite this article as:
Singh, V.K. & Rai, A.K. Lasers Med Sci (2011) 26: 673. doi:10.1007/s10103-011-0921-2


We review the different spectroscopic techniques including the most recent laser-induced breakdown spectroscopy (LIBS) for the characterization of materials in any phase (solid, liquid or gas) including biological materials. A brief history of the laser and its application in bioscience is presented. The development of LIBS, its working principle and its instrumentation (different parts of the experimental set up) are briefly summarized. The generation of laser-induced plasma and detection of light emitted from this plasma are also discussed. The merit and demerits of LIBS are discussed in comparison with other conventional analytical techniques. The work done using the laser in the biomedical field is also summarized. The analysis of different tissues, mineral analysis in different organs of the human body, characterization of different types of stone formed in the human body, analysis of biological aerosols using the LIBS technique are also summarized. The unique abilities of LIBS including detection of molecular species and calibration-free LIBS are compared with those of other conventional techniques including atomic absorption spectroscopy, inductively coupled plasma atomic emission spectroscopy and mass spectroscopy, and X-ray fluorescence.


LaserLIBSElemental analysis techniqueBiomaterials


The characterization of biomaterials is both interesting and challenging for analytical scientists. There are many analytical techniques based on the emission of electromagnetic radiation produced after excitation of atoms, ions and molecules present in the target materials. These techniques generally employ some kind of energy source to excite the species present in the sample to higher energy levels from where they return to lower levels emitting the characteristic radiation which can be collected and sent to a wavelength selector and finally detected. However, most emission techniques cannot be applied directly to intact samples because they require treatment before analysis, and this limits the use of these analytical techniques in environmental analysis, forensic analysis, archaeological analysis, biological analysis and many others areas of applied science as these samples are very sensitive to their surrounding atmosphere. Trace mineral elements play an important role in biologically active materials because minute variations in the amounts of these minerals may adversely affect the metabolism processes in all living creatures. Thus detection and quantification of such minerals in biomaterials is essential to monitor metabolism. Further, the presence of small amounts of toxic and heavy metals in food and food products can adversely effect human health and consequently the detection and analysis of these metals present at trace levels is of the utmost importance.

Several analytical techniques have been applied in an attempt to address these problems. But these methods often require laboratory-scale equipment and sophisticated sample treatment protocols. Recently laser-induced breakdown spectroscopy (LIBS) has emerged as a powerful analytical technique for direct spectrochemical analysis of a variety of solids, liquids and gases with no or little sample pretreatment. During the past decade LIBS has been developed for various applications including biomedical applications. We review the historical development and fundamentals of LIBS, instrumentation for the analysis of samples including biological samples, its applications in diverse fields of biomedical science and its future in applied research. This technique is suitable for applications that cannot be addressed by other conventional analytical methods. In LIBS, intense laser pulses at UV, visible or infrared wavelengths are used to ablate the target material to produce a luminous plasma plume which emits characteristic radiation that helps to determine the target composition. The emitted radiation from the plasma is analysed using a high-resolution spectrometer and sensitive detector. The LIBS spectrum of the target material yields qualitative and quantitative information which can be correlated with the sample identity.

Several reviews [114] and books [1517] have been published that discuss various aspects of LIBS ranging from fundamental studies to applied research. LIBS is probably the most versatile method of elemental analysis currently in use for many biomedical applications [17]. In the biomedical field, LIBS is particularly used to diagnose and classify cancers in vivo by determining the intensity ratios of trace elements in normal and cancerous material [18]. Generally, cancer diagnosis and classification rely upon subjective interpretation of biopsy material, but with the use of LIBS the diagnosis of cancer is easier. Although LIBS has been traditionally considered as an elemental analysis technique it is also being used successfully for molecular identification of materials including biomaterials.

Over last couple of years there has been an exponential growth in the areas of utilization of LIBS which is reflected by an increasing number of publications, and thus its utility as an analytical technique has been proved. We summarize below the work related to the fundamentals, instrumentation and biomedical applications of LIBS, providing up-to-date information on instrumentation and its applications to the direct analysis of biomaterials.

The laser and its applications in biomedicine

In 1917, Albert Einstein first theorized about the phenomenon of stimulated emission which is the backbone of the laser [19]. The first working laser, the pulsed ruby laser, was invented by Theodore H. Maiman in 1960 [20]. However, there were no potential applications known and in fact it was popularly referred to as a tool looking for applications [20]. Within a few years of the development of the laser its medical application particularly in urology was reported in 1968 by Mulvaney and Beck who used the ruby laser to fragment urinary calculi [21]. They were able to ablate the calculi, but the continuous wave ruby laser generated excessive heat and so its clinical use was not extended. By the mid-1980s the use of the laser to treat stone disease had become established, and the era of laser lithotripsy had begun [22]. Thus the development of potential applications of the laser in medical science has given a new direction to analytical scientists. Over last few decades the use of lasers had become standard for the treatment and diagnosis of many diseases including the treatment of a range of ophthalmological and dermatological conditions [2325]. There are many medical disciplines where lasers are successfully used for a variety of purposes. However, there is a necessity for further research in laser applications in medicine in order to achieve optimal outcomes [26].

To improve the medical applications of laser based techniques an understanding of the kinetics and dynamics of laser interactions with biological tissues is essential. Knowledge of laser–tissue interactions will guide the identification of optimal laser parameters to achieve more efficient and safer outcomes [27]. The applications of lasers in medicine can be categorized into two major disciplines, namely diagnostic and therapeutic. The vast majority of applications are in the therapeutic field. In recent years, there has also been much interest in the use of the laser as a diagnostic tool and this has resulted in some exciting developments across all medical specialities [28]. Gaining clinical diagnostic information by the use of a laser probe, for example for the analysis of tissue and biomaterials, may better guide treatment and may also be helpful in optimizing the therapeutic technique [29].

Laser-induced breakdown spectroscopy

Shortly after the invention of laser device using a ruby crystal, Brech and Cross demonstrated the first useful laser-induced plasma produced on the surface of the target [30]. This was the “birth” of the LIBS technique, and in subsequent years significant milestones were reached in the development of this method. LIBS is an atomic spectroscopy technique which is based on the analysis of the spectral emission from laser-induced plasmas produce by high-power laser pulses of short duration applied to the surface of the target material. Generally, atomic emission spectroscopy (AES) uses an external energy source to excite atoms in their ground state. The atoms spontaneously emit radiation when they revert back to the lower energy state, with the emission intensity being proportional to the concentration of atoms in the ground state [31].

In the LIBS technique, high-power pulsed lasers are used as the excitation source (Fig. 1). During the 1980s, the neodymium-doped yttrium aluminium garnet (Nd:YAG) laser was the most common laser system used in most applications involving LIBS. The Nd:YAG laser became popular for LIBS because it was easily configured to produce the megawatt peak power levels required for reliable laser plasma generation from the target materials. The Nd:YAG is a four-level laser system that produces a very high power emission. The energy levels of the Nd3+ ion are responsible for the fluorescent properties and is thus suited for the amplification process. The energy level diagram and laser transitions of the trivalent neodymium ion in the Nd:YAG laser are shown in Fig. 2. The Nd:YAG laser is a classic four-level laser, as illustrated in Fig. 2. The transition from 4F3/2 to 4I11/2 is responsible for laser emission at 1.064 μm (Fig. 2); details of the other transitions of the Nd:YAG laser may be found elsewhere [32].
Fig. 1

Schematic diagram of a simple LIBS system comprising the essential components to produce laser-induced plasma and of the detection system
Fig. 2

Energy level diagram of the Nd:YAG laser based on the doping of the YAG crystal with Nd3+. The Nd3+ ions in various types of ionic crystals act as a laser gain medium, typically emitting 1.064 μm light from a particular atomic transition in the Nd3+ ion, after being pumped into excitation from an external source

Generally, the lasers are usually operated at the fundamental wavelength at 1064 nm, although other wavelengths (532 nm, 355 nm and 266 nm) can be generated using nonlinear frequency conversion crystals. Additional components of a typical LIBS system include focusing and collection optics, a spectrometer, and a data acquisition system. Broadband spectrometers that work in the wavelength range 200–1,000 nm allowing the simultaneous detection of multiple elements are commercially available [33].

To generate a plasma spark, a high-power laser beam pulse of short duration is focused onto or into the target material. An optically induced plasma or spark is formed on the surface of (a solid or a liquid) or in the sample (bulk liquid or a gaseous medium) when the laser power irradiance exceeds the breakdown threshold of the sample [31]. The plasma radiates both a continuum component due to inverse Bremsstrahlung radiation from electron–ion collisions that decays rapidly, and an emission line component that decays more slowly. For a better understanding, the whole process can be divided into three time domains as illustrated in Fig. 3.
Fig. 3

Generation of laser-induced plasma and timing of a LIBS plasma: (a) plasma ignition, (b) emission due to free–free transitions (Bremsstrahlung process) and free–bound transitions, and (c) line emission due to bound–bound transitions

In the first time domain (Fig. 3a), the laser heats and evaporates a small amount of the sample. The ablated materials expand at supersonic velocities producing a shock wave which propagates from the surface towards the surrounding atmosphere. The first seed electrons/species are created, either due to multiphoton ionization or thermal emission of the surface. These electrons/species absorb further photons from the same laser pulse during the duration of the pulse. The species are excited producing emitting plasma that is visible to the naked eye. Depending on the conditions, multiple charged ions are also present in the plasma plume. The time domain (Fig. 3b) is characterized by a broadband emission originating from the Bremsstrahlung of the free electrons and electron–ion recombination and it has duration of a few hundred nanoseconds. Weak lines show up on the strong continuum and they are mostly identified as ionic lines of the plume constituents. The final domain (Fig. 3c) is characterized by an emission spectrum, where narrow atomic lines dominate corresponding to the elements present in the plume, and line strength is proportional to the atomic concentration. This time domain lasts for several microseconds and it is exactly the domain that is relevant for elemental analysis of target material.

Therefore, in summary, the plasma emission can be analysed by spectroscopic methods by time gating. The plasma radiation is initially dominated by a white light continuum which contains little intensity variation as a function of wavelength. After breakdown, the plasma expands outwards and back towards the focusing lens. The expansion occurs at 105 m/s and creates an audible shock wave [34]. Plasma decay occurs by radiative, quenching, and electron–ion recombination processes that result in the formation of neutral species [3].

A gated spectrometer is used to capture the plasma spectrum and generally the spectrometer covers a part or all of the ultraviolet through near infrared range (200–1,000 nm). For simultaneous multielement analysis, an echelle spectrometer can be used, which contains an echelle diffraction grating with coarse grooves and large blaze angles. The grooves have steep sides to cover the full range of wavelengths and a prism is needed to separate overlapping orders of the grating [31, 35]. Generally the detector devices within the spectrometers are charge-coupled devices (CCDs) or intensified CCDs (ICCDs). The spectral line wavelengths and intensities obtained from plasma ablation can be compared with a standard atomic line reference and calibrated against samples of known composition to determine the elemental composition of the sample. The intensities of the spectral lines in the LIBS spectra of the target material provide a quantitative description of the concentrations of the elements contained in the material [15, 33].

Advantages and some limitations of LIBS

The main attributes that make LIBS a very powerful and attractive analytical tool are: its in situ measurement capability, simultaneous multielemental detection, and real-time analysis of materials in the laboratory or in the field. There are several significant advantages that make LIBS more applicable than other techniques, as follows:
  1. 1.

    The need for little or no sample preparation results in less need for toxic chemicals usually required for sample preparation.

  2. 2.

    Versatile sampling of all media (solids, liquids, gases as well as biomaterials), including both conducting and nonconducting materials.

  3. 3.

    Very small amounts of sample (order of micrograms) are vaporized.

  4. 4.

    Extremely hard materials that can be difficult to get into solution can be analysed (e.g. ceramics, glasses and superconductors).

  5. 5.

    With a spatial resolving power of the order of 100 μm, micro regions can be analysed.

  6. 6.

    Multiple elements can be determined simultaneously.

  7. 7.

    The direct determination of aerosols or ambient air is possible.

  8. 8.

    The analysis is simple and rapid.

  9. 9.

    Point detection capability enables the analysis of any kind of material including biomaterials.

  10. 10.

    Remote sensing is possible with the use of fibre optics.

  11. 11.

    Samples can be analysed in a hostile environment.

  12. 12.

    Underwater analysis is possible.

  13. 13.

    Development of field instruments is possible.

  14. 14.

    Stand-off detection is also possible using a telescope for light collection without the need for a fibre optic cable near the sample.

  15. 15.

    LIBS is minimally destructive because the amount of sample consumed is very small (nanograms) depending on the laser pulse energy. Thus it is suitable where only small amounts of material are available.

LIBS has certain limitations:
  1. 1.

    Difficult to get suitable matrix-matched standards, which makes the technique qualitative or at best semiquantitative.

  2. 2.

    Detection limits are generally not as good as those of the conventional techniques.

  3. 3.

    Precision is poor as compared to conventional techniques.

  4. 4.

    Safety measures are required to avoid ocular damage by the high-energy laser pulses.


Studies of LIBS applications in biomedicine

The unique advantage of LIBS in allowing the study of a broad variety of samples without sample preparation is attractive for the analysis of biological samples. Therefore, the LIBS technique is being utilized promisingly for the analysis of biomaterials. We discuss here the use of LIBS for the study of biological samples particularly in the field of biomedical science.

Tissue analysis

Sun et al. [36] reported the use of LIBS for the detection of the trace element Zn in human stratum corneum using an analytical curve with a linear correlation coefficient of 0.998 and a detection limit of 0.3 ng/cm2. The authors found that zinc is absorbed through the skin and its concentration decreases exponentially with depth in the skin. The authors concluded that LIBS is a useful tool for trace element analysis in human skin. De Souza et al. [37] used LIBS to investigate the relative elemental composition of chick myocardium tissue. The analysis showed the presence of elements including Na, K, Ca, and H which were identified separately and compared with the common elements present in tissues. They found that in the extracellular matrix Na predominates and in the intracellular space K predominates together with Ca and Mg. Finally, they concluded that the measurement of the relative atomic composition by means of the laser ablation might lead to a technique for the discrimination of different materials or tissues.

Kumar et al. [18] reported the first experiments to explore the possibility of using LIBS for cancer detection. They analysed malignant and normal tissue from a canine haemangiosarcoma. Canine haemangiosarcoma, which is a model for human angiosarcoma, may be valuable to define and analyse these types of tumours and suggest potential means of improving their classification in humans [18]. They found that the concentration ratios of elements in normal and tumour cells mainly Ca/K and Cu/K were significantly different and concluded that LIBS may be used as an in vivo diagnostic tool for cancer detection.

Zheng et al. [38] have recently investigated the feasibility of using LIBS to characterize animal tissues. Gornushkin et al. [39] used a simple statistical correlation method for solid material identification and Zheng et al. [38] adopted the same technique for tissue identification. Recently, Adamson and Rehse [40] used LIBS for the detection of Al in surrogates of human tissue up to parts per million levels. In this study, tissue was modelled using a 2% agarose gelatin doped with an Al2O3 nanoparticle suspension. A calibration curve created with standard reference samples of known Al concentrations was used to determine the limit of detection, which was less than 1 ppm. The authors concluded that LIBS could be a candidate for the real-time in vivo detection of metal contamination in human soft tissues.

Recently, Myers et al. [41] performed noninvasive, in-situ detection of malignant skin tissue and other abnormalities using a portable LIBS system with a fibre spectrometer and an eye-safe erbium glass laser. They established a medical screening procedure that used a compact “eye-safe” LIBS device capable of real-time, in-situ determination of healthy and unhealthy skin tissue without performing traditional biopsy. Tameze et al. [42] performed empirical analysis of LIBS images for ovarian cancer detection. Using LIBS, plasma images of the blood samples were generated and analysed. They compared the images from blood specimens of cancer-free mice to those of transgenic mice.

Analysis of minerals in the human body

Samek et al. [43] discussed the utility of the LIBS technique for the analysis of minerals and potentially toxic elements present in calcified tissues including bones and teeth to study the influence of environmental exposure and other biomedical factors. They investigated the multidimensional profiles of the elements present in the teeth and bone samples. They quantified the trace elements Al, Sr and Pb using calibration curves, and their results were in good agreement with the results of atomic absorption spectroscopy (AAS). Samek et al. [4446] also performed a quantitative analysis of trace minerals in teeth samples obtained from patients of different ages. The authors concluded that the Al found in the teeth could most likely be attributed to the use of toothpaste with whitening additives as well as the presence of fillings. LIBS has also been used in studies of microleakage in dentistry [47], microleakage between infrastructure and veneer materials in dentures [48], in vitro bone-like apatite formation [49], and carious tooth decay [50].

Recently, the use of LIBS for the rapid identification of teeth affected by caries has been demonstrated by Singh and Rai [51]. They were able to detect a broad range of elements including Ca, Mg, Cu, Zn, Sr, Ti, C, P, H, O, Na and K. They found that the caries-affected part of the teeth contained lower amounts of Ca and P than the healthy part, but higher amounts of Mg, Cu, Zn, Sr, C, Na, K, H and O. They explained the presence of the different metal elements present in the teeth and also discussed their role in the formation of caries.

Collins and Vass [52] used LIBS for the classification of human and animal bone including bone from rabbits, pigs, sheep, bears and cows. They found clear differences between their elemental contents. Niemz [53, 54] evaluated the physical parameters of LIBS for the laser-induced ablation of teeth.

Bilmes et al. [55] used LIBS to identify the trace elements in hominid teeth to investigate their eating habits. Tawfik and El-Tayeb [56] used LIBS to determine the elemental levels of Ca, Pb, Al and Sr in teeth of ancient and modern Egyptians to provide information for studying the aetiology of various diseases. They found higher Ca, Pb and Al and lower Sr levels in teeth from ancient Egyptians than in teeth from modern Egyptians. The high Pb and Al levels in ancient Egyptians indicate that the environment might have been polluted by these metals at that time. They concluded that the study of the elemental levels in teeth could provide information on the environment and eating habits of people of different eras.

Recently, Abdel-Salam et al. [57] used LIBS to estimate the hardness of calcified tissues (human teeth, shells and eggshells) by determining the ratios of the intensities of the ionic and atomic lines of Ca and Mg. They found that the ratios Ca(II)/Ca(I) and Mg(II)/Mg(I) were proportional to the material hardness. They concluded that LIBS can be used for the estimation of the hardness of any calcified tissue.

Corsi et al. [58] determined the concentrations of minerals in human hair by calibration-free LIBS (CF-LIBS), establishing the applicability of this technique for the analysis of biological materials. Ohmi et al. [59] analysed human nail, hair and tooth samples using LIBS and detected Ca with great sensitivity. They found that Ca from the hair of healthy females significantly decreases with age, which is in good agreement with medical reports. Haruna et al. [60] reported the use of LIBS to detect Ca in human hair and nails with high sensitivity, and showed that the determination of Ca in hair may be a method to monitor the daily intake of Ca and to screen for osteoporosis. They also detected Na and C. Branch et al. [61] also used LIBS to monitor the levels of trace elements including Mg, Na, and Ca in human hair and nails. Recently, Hamzaoui et al. [62] used LIBS for the quantitative analysis of pathogenic nails and found a distinct difference in the intensity distribution of the elements Ca, Na and K between normal and pathogenic nails. They used the B2+ → X2+ violet band emission spectrum of CN for the determination of the transient temperature induced by the laser ablation of nails. Pathak and Rai [63] also used LIBS for the in vivo analysis of human nails, and found changes in the intensity of Ca, Na and K in nail reshaped after injury as compared to normal nail. They also applied principal component analysis (PCA) to the LIBS data to differentiate reshaped and normal nails. Ng and Cheung [64] demonstrated the feasibility of LIBS for quantifying Na and K in single human erythrocytes.

Martin et al. [65] used LIBS for the identification of metals including Pd and Ag dispersed in bacterial cellulose membranes using wet and dry metal-doped membranes.

Samek et al. [66] reported the use of LIBS with laser-induced fluorescence spectroscopy for the analysis of B, Ca, Cr, Cu, Fe, Si and Zn, and toxic elements including Al, Cd, Pb and Hg, in the human body. The authors also used this system to screen blood samples, and detected trace amounts of Rb (levels down to 0.3%) in blood [67]. Recently, Rehse et al. [68] used LIBS for the identification and discrimination of Pseudomonas aeruginosa bacteria grown in blood and bile.

Recently, Wu et al. [69] used LIBS for the analysis of body fluids. In this study the plasma was created in a solution (comprising 10% glucose and 0.9% sodium chloride) by focusing the 1064-nm Nd:YAG laser beam with an energy up to 300 mJ, a frequency of 10 Hz, and a pulse width of 10 ns. They reported that the organic matter as glucose and metal elements can be synchronously analysed by LIBS and the metal elements can be determined more sensitively than the organic matter. Their results also indicated that there is an exponential relationship between the intensities of characteristic spectra and concentrations. Further, they concluded that LIBS may be used as a new method for the accurate measurement of trace elements in body fluids.

Analysis of stones in different organs of the human body

Lasers have been used to breakdown urinary and kidney calculi since 1987 [70]. The laser shock-wave disintegrates the calculus into tiny fragments. Fang et al. [71] used LIBS for the quantification of the elemental contents Ca, Mg, Na, Sr, K and Pb in urinary calculi, and they concluded that LIBS offers the possibility to accurately measure trace elements in such stones without the need for any elaborate sample preparation. Recently, Singh et al. [72, 73] characterized qualitatively and quantitatively the different types of gallbladder stone (cholesterol stones, pigment stones, mixed stones). They analysed different parts of the gallstones and found higher levels of metal elements in the centre than in the shell and surface of the gallstones. Singh et al. [74] also reported the use of LIBS for the in situ quantitative estimation of the elemental constituents in different parts of kidney stones (centre, shell and surface parts) obtained during surgery. They estimated the quantities of Cu, Zn, Mg and Sr in the stones using calibration curves. They also used the ratios of the intensities of the different elemental lines to determine the spatial distribution of different elements inside gallstones and kidney stones. The results were in good agreement with the results of other analytical techniques including inductively coupled plasma AES (ICP-AES) and ICP mass spectroscopy (ICP-MS).

Anzano and Lasheras [75] used a μ-LIBS system with a higher energy laser and an echelle spectrograph with an ICCD camera and elemental ratios of reference materials to identify urinary stones. They used linear or parametric and rank or nonparametric correlation methods in their analysis. Pathak et al. [76] used LIBS to characterize gallstones collected from patients from the north-east region of India (Assam). In the different layers of the gallstones in the spectral region 200–900 nm they detected Ca, Mg, Mn, Cu, Si, P, Fe, Na, K, C, H, N and O. Finally, they were able to discriminate the dark layers and the light layers of the gallstones on the basis of the presence and the intensities of the spectral lines of C, H, N, O and Cu. They also used LIBS to investigate the evolution of C2 swan bands and CN violet bands in the LIBS spectra of the gallstones in air and an argon atmosphere. They also investigated the degree of correlation between the presence of major and minor elements in the gallstones and the common diet of the population of Assam.

Salt is an essential dietary mineral in humans and currently the issue of the potential beneficial or damaging effects of salt intake in patients with chronic kidney disease is controversial. Thus, Singh et al. [77] used LIBS for the investigation of common Indian edible salts suitable for patients with kidney disease. They found that Saindha salt (rock salt) is more beneficial than other edible salts for patients suffering from chronic kidney disease.

Martin et al. [78] used LIBS and pulsed Raman spectroscopy to measure the elemental composition of soils and heterogeneous biological matrices. They determined the concentrations of elemental C and N in soils, and the presence of metal contaminants in invertebrates. The pulsed Raman spectroscopy method was used in exploratory studies to assess prominent molecular vibration peaks from the same soils. The techniques greatly facilitated the elemental analysis of heterogeneous environmental and biological matrices by reducing sample preparation and analysis times.

Analysis of biological aerosols and nonaerosolized biological materials

In recent years, the analysis of microscopic particles, cells, aerosols, and especially bioaerosols (bacteria, fungi, viruses, pollen) has received increasing interest because of biological threats to public and defence security. Minute amounts of inhaled bioaerosols can cause disease, toxicity and allergic reactions. Thus, the detection and identification of biological aerosols and agents is an urgent civil and military requirement which will be useful in environmental monitoring.

Morel et al. [79] used time-resolved LIBS for analysing biological matter for the detection of biological hazards. They sought to detect six bacteria including Bacillus globigii as a surrogate for B. anthracis and two pollens in pellet form. Time-resolved LIBS exhibited a good ability to differentiate all the investigated species, whatever the culture medium, species or strain. Samuels et al. [80] analysed bacterial spores, moulds and pollens using LIBS. They used PCA, and found that LIBS provided adequate information to discriminate among the biomaterials. Thus LIBS is able to discriminate between bacterial spores, moulds and pollens.

Kim et al. [81] analysed five bacterial strains Bacillus thuringiensis T34, Escherichia coli IHII/pHT315, B. subtilis 168, B. megaterium QM B1551, and B. megaterium PV361. They identified the major inorganic components in the bacterial samples, including Ca, Mg, K, Na, Fe and P. They were able to differentiate the bacterial strains on the basis of two-dimensional charts of the bacterial components, Ca versus P. They demonstrated the potential of LIBS for the rapid classification of bacteria with low false-positive rate and with minimum sample preparation.

Hybl et al. [82] examined some common biowarfare agent simulants in comparison with some naturally occurring biological aerosol components (bacterial spores, media/protein, fungal/mould spores, and pollen) using a broadband LIBS system. Instead of using pellets or substrate deposited layers, homogeneous samples were aerosolized in a microcentrifuge tube either by making use of the laser-induced shock wave or acoustically by dispersing a dry powder suspension above a loudspeaker. They demonstrated that LIBS has significant potential for classifying bioaerosols and is able to resolve different elemental ratios in biowarfare agent simulants and also in common biological and environmental interferants. Recently, a transportable UV laser-induced fluorescence-cued LIBS test bed has been developed by Hybl et al. [83] and used to evaluate the utility of LIBS for biological agent detection.

Boyain-Goitia et al. [84] were the first to analyse single biological microparticles (pollens from a variety of flowers) by LIBS. They demonstrated that single laser pulse LIBS can be performed on single biological microparticles, and that many more species need to be measured to generate a suitable reference library so that detection and identification can be made reliably in real time.

In recent years, many researchers have focused on the detection and identification of individual bioaerosols using LIBS. Dixon and Hahn [85] demonstrated the feasibility of LIBS-based single-shot analysis of metal-rich bioaerosols (Bacillus spores). Beddows and Telle [86] discussed the prospects for real-time in situ LIBS for the identification and classification of bioaerosols within common urban aerosol mixtures. Comparing LIBS measurements with data from a mobile single-particle aerosol mass spectrometer, they found that mass spectrometer provided statistically relevant data over an extended period of time, highlighting the variation in the background composition. Baudelet et al. analysed E. coli using a LIBS system delivering femtosecond pulses [87], and also compared the results with those from a nanosecond regime [88].

Munson et al. [89] used LIBS to record the emission spectra of bacterial spores, moulds, pollens and nerve agent simulants with the aim of differentiating them using statistical methodologies (linear correlation, PCA) and soft independent modelling of class analogy. They found that spectral averaging and weighting schemes may be used to improve sample differentiation.

Gibb-Snyder et al. [90] achieved size-selective sampling of B. anthracis surrogate spores from realistic common aerosol mixtures by LIBS. Diedrich et al. [91] analysed four strains of E. coli using LIBS with nanosecond pulses and applied a discriminant function analysis to the LIBS spectra obtained from live colonies of all four strains. They showed that LIBS was able to discriminate an environmental strain from a pathogenic strain, which suggests the possibility of using LIBS as a practical diagnostic test to identify strains obtained from environmental assays.

Recently, Gottfried et al. [92] used LIBS for the stand-off detection of chemical and biological threats which led to the development of a combined partial least-squares discriminant analysis (PLS-DA) model for the detection of chemicals, biological materials and explosives using a single stand-off LIBS sensor, demonstrating the potential of stand-off LIBS for the detection of hazardous materials.

Femtosecond LIBS has also been used to analyse different species of bacteria, and the trace mineral elements detected in the bacterial species were Na, Mg, P, K, Ca and Fe [93]. The unambiguous discrimination of these different bacteria was possible using the concentration profiles of the trace elements. Xu et al. [94] demonstrated the feasibility of remote time-resolved filament-induced breakdown spectroscopy of biological materials. The fluorescence from egg white and yeast powder induced by femtosecond laser pulse filamentation in air was detected in the backward direction with the targets located 3.5 m away from the detection system. Remarkably distinct spectra of egg white and yeast powder were found. In summary, they concluded that filament-induced breakdown spectroscopy is potentially a good technique for the remote detection and identification of biological species when combined with time-resolved measurements.

Assion et al. [95] performed element-specific in situ investigations of biological samples using high spatial resolution LIBS. They investigated the analytical performance and ablation process of femtosecond LIBS in comparison to nanosecond LIBS using Ca2+ detection in water, where the water served as a first substitute for biological material. They investigated the ablation process in the outer epidermal wall of a sunflower (Helianthus annuus L.) seedling stem. They found that analytical measurements with high spatial resolution could be performed on biological samples using femtosecond LIBS as demonstrated by in situ measurements of the distribution of wall-associated Ca2+ within the peripheral cell wall of the sunflower seedling stem.

Xu et al. [96] demonstrated the feasibility of remote detection and differentiation of some very similar bioaerosols related to agricultural activity (barley, corn and wheat grain dusts) through nonlinear fluorescence of fragments induced by the high-intensity inside filaments of femtosecond laser pulses in air. They detected signal in the Lidar configuration with targets located at 4.7 m away from the detection system. The molecular bands detected from the species investigated were C2 and CN bands, and the atomic lines detected were Si, C, Mg, Al, Na, Ca, Mn, Fe, Sr and K. The authors used the intensity ratio method to distinguish these samples. They concluded that this technique could be used at long distances and thus could be used as a sensor for similar biological hazards for public and defence security.

In this section we discussed the application of LIBS to the analysis of biological aerosols [8288, 90, 96] and nonaerosolized materials [7981, 89, 9195].

Comparison of LIBS and other conventional atomic spectroscopic techniques

Although LIBS can interrogate samples under conditions which are not possible using other conventional analytical methods, it is difficult to compare its merits with those of other well-established laboratory techniques. It is difficult to compare the limit of detection of LIBS with those of other techniques because these methods chemically reduce the target sample (e.g. a metal) to a solution in which the analyte concentration is in the order of micrograms per litre. LIBS can certainly analyse these liquids directly, but doing so can reduce the concentration compared to the target sample, lowering the detectability for LIBS. However, LIBS may show greater sensitivity for the detection of an element in its native matrix (e.g. in soil) compared to diluting the elements present in the sample by preparing the sample as a solution. But it is still instructive to compare LIBS values with those from other more conventional analytical techniques and the factors that can be used to evaluate the potential of LIBS, such as measurement precision, cost, portability and sample analysis time.

AAS is a laboratory-based technique that uses the absorption of ultraviolet or visible light to measure the concentration of metallic elements in the gas phase following vaporization of liquid or solid samples in a flame or graphite furnace. This is a quite inexpensive method. Atoms absorb or emit radiation of discrete wavelengths because the allowed energy levels of electrons in atoms are fixed. The energy change in the atoms is associated with a transition between two energy levels, which is directly related to the frequency of the absorbed radiation. A graphite furnace AAS uses a graphite tube with a strong electric current to heat the sample. It is a more efficient atomizer, and the furnace dries the sample, ashes the organic matter, and vaporizes the sample to produce free analyte atoms [97]. In flame AAS, the liquid sample is aspirated into a flame using a nebulizer. Solid samples are prepared with an acid digestion method to produce a solution [97]. The flame is lined up in a beam of light of the appropriate wavelength. The flame causes the atom to undergo a transition from the ground state to the first excited state and to absorb some of the light from the beam. The concentration of gas-phase atoms is usually measured using a calibration curve obtained from a standard. The Beer-Lambert law often becomes difficult to apply because atomic concentration and atomization efficiency are usually nonuniform throughout the sample matrix particularly in biomaterials, and the path lengths are also nonuniform. The main disadvantage of this technique is that it analyses one element at a time. On the other hand LIBS can be used for the simultaneous multielement analysis of any kind of sample [97]. In particular, for the analysis of biological samples the spatial distribution of elements give more information about the characteristic properties of the sample and this is destroyed during the ashing and sample preparation protocols in AAS. Additionally, the analysis of the organic components C, H, N, O etc. is essential in the analysis of biologically important materials, and it is very difficult to analyse these using AAS. The LIBS method has the advantage compared to AAS because these elements are the organic markers of biologically important materials and it is important to determine their spatial distribution.

Conventional AES techniques provide quantitative measurement of the optical emission from excited atoms to determine analyte concentrations. Analyte atoms in solution are aspirated into the excitation region where they are desolvated, vaporized and atomized. Excitation sources for atomic emission include flame excitation, direct current plasma, ICP, microwave induced plasma, and electrode arc and spark. These high-temperature atomization sources provide enough energy to excite the atoms into higher energy levels from where they decay back to lower levels and emit radiation. Since the transitions are between distinct atomic energy levels, the emission lines in the spectra are narrow. The multielement spectrum of a sample can be very congested, and spectral separation of nearby atomic transitions requires a high-resolution spectrometer. Since all atoms in a sample are excited simultaneously, they can be detected simultaneously which is the advantage of these methods as with LIBS. But in LIBS the sample is directly vaporized and excited by a laser pulse of high energy which causes dielectric breakdown and creates a hot plasma. ICP is the most commonly used high-temperature (7,000–8,000 K) plasma excitation source that efficiently desolvates, vaporizes, excites and ionizes atoms and helps to reduce molecular interferences. ICP is used in conjunction with AES and MS because both require the sample to be in an aerosol or gaseous form. Samples for ICP are typically prepared using an acid digestion method, although direct laser ablation of solid samples into the ICP is possible to avoid the dissolution procedures of solid samples prior to the determination of the elements [98]. In particular, LIBS is as reliable as ICP in the analysis of most complex liquids. However, LIBS suffers from relatively poor sensitivity and detection limits compared to ICP, but the double-pulse technique seems to be a promising technique to improve the detection limits of LIBS [99].

X-ray fluorescence (XRF) is another atomic emission technique that is a strong competitor to LIBS in some applications that can be laboratory- or field-based. In XRF, X-rays are used to eject electrons from inner orbitals of the atom. When an electron from a higher energy level falls to fill the vacancy, a photon is emitted that has an energy characteristic of the particular element. XRF is used in a wide range of applications including the analysis of biomaterials and is particularly well suited to investigations that involve bulk chemical analysis of major and trace elements. But XRF is limited to the analysis of relatively large samples (typically >1 g) and samples must be in powder form and effectively homogenized. XRF is used for the analysis of materials for which compositionally similar, well-characterized standards are available, and materials containing high abundances of elements for which absorption and fluorescence effects are well understood. Portable XRF systems are available, but suffer from an inability to detect elements with an atomic number below 12 and show an interference effect that can mask the analyte elements. Although X-rays do not penetrate the surface and hence XRF is a nondestructive technique, deeper regions cannot be interrogated which is possible with LIBS.

Overall, ICP has the lowest detection limit of the methods followed by graphite furnace AAS and flame AAS. These three techniques are laboratory-based, but portable field XRF and LIBS instruments are also available. Comparing detection limits, LIBS tends to have somewhat lower limits than XRF for many elements. In addition, LIBS is the only technique that has a point detection capability. Single-shot LIBS is also useful for many medical applications. Recently, single-shot LIBS has been used for the identification of the major and minor constituents of cholesterol gallstones [72]. LIBS can be used to determine the constituents both on the stone surface and inside the stones. The LIBS technique is advantageous for this type of work since it is a rapid technique, is not time intensive, and does not require sample preparation. In this work, LIBS spectra were analysed for element identification and comparison of relative concentrations. Both mineral elements (e.g. Ca, Cu, Mg, Na and K) and organic elements (e.g. C, N, O and H) were determined. The intensity of the atomic lines of C, Ca, H, Mg, N, Na, O and K were shown to be different for pigmented and nonpigmented regions of the stones, thus suggesting concentration differences. These kinds of study are not possible using AAS, ICP and XRF. LIBS can be applied to the analysis of biological samples that are small in size, although this is not without its challenges. Aerosol analysis is a unique application but requires significant work to form plasma on an aerosol particle. Single-shot LIBS has been used to analyse several biological aerosols [83, 85, 90].

Molecular detection using LIBS

Traditionally, LIBS is considered as a physical diagnostic tool based on elemental analysis because information about the chemical composition of the materials is lost as the temperature of the laser-induced plasma is usually greater than 10,000 K. However, much effort have been made to provide the capability to identify the molecular species in pure materials. The identification of explosive materials is based on ratios of the intensity of two spectral lines to determine relative molecular concentrations. De Lucia et al. recorded the LIBS spectra of several explosives including pentaerythritol tetranitrate, cyclotetramethylene tetranitramine, cyclotrimethylene trinitramine, and trinitrotoluene [100]. Such spectra are the start of a LIBS database for molecular identification schemes aimed at identifying parent materials. Methods to distinguish between spectra usually involve either analysing the whole LIBS spectrum or focusing on selected atomic lines; for organic materials these atomic lines mainly include C, H, N, and O, which are ubiquitous in nature which complicates the analysis of organic species.

Portnov et al. [101] have shown that molecular emission may also be used to infer material composition. They applied LIBS to nitroaromatic and polycyclic aromatic hydrocarbon samples to characterize the resultant emission in ambient air. The emission consisted of spectral features related mostly to CN and C2 molecular fragments and to C, H, N, and O atomic fragments. They found that the intensity ratio of the C2 Swan bands to the CN violet system was useful in determining the number of carbon–carbon double bonds in the analyte, which increases increasing numbers of aromatic rings. They combined the molecular band ratio with the O/N ratio, which carries information about the number of nitro groups in the sample. They suggested that these pieces of information could help infer the presence of polycyclic aromatics and nitroaromatics in a particular sample.

Nitrogen is associated with numerous moieties in biological materials, including amino acids, proteins and enzymes, and thus may provide additional discrimination. However, since in the nanosecond time scale all molecular information is essentially erased in the LIBS plasma, the intensity of the CN bands observed in LIBS with nanosecond lasers is dependent on the C concentration, the N concentration, the matrix (which influences recombination chemistry) and the plasma parameters, which influence the cooling rate in the plasma [102]. The CN and C2 concentrations are also dependent on the atmosphere (N from the air contributes to the CN signal, and O from the air decreases the C2 since more CO and CO2 is formed) [102]. Hence these CN emissions are probably only a reliable marker of the C/N elemental ratio in a particular plasma volume, and not of the original proportion of C and N bonds.

As was discussed above, Boyain-Goitia et al. [84] did some work on LIBS and Raman detection of individual pollen spores attached to needle tips. In their LIBS experiments, where CN, C2 and Ca, and several trace elements, were observed, they found that there was a large degree of variation between samples, and while it was possible to normalize spectra and minimize the pulse-to-pulse variation, they found it difficult to identify features or patterns in the spectra associated with particular pollen types. They concluded that multivariate and pattern-recognition techniques should be applied to LIBS analysis of bioparticles to improve discrimination. They came to a similar conclusion regarding their Raman measurements, in which they were able to see particular vibrational features associated with plant structure.

Pathak et al. [103] recorded the LIBS spectra of gallstone samples and used PCA to classify them (cholesterol type, mixed type and pigmented type). They prepared a LIBS library (a set of LIBS spectra of training samples of each category) and PCA was used to differentiate the gallstone samples. Their results (Fig. 4; PC1 contributed 92.88%, and PC2 5.47% of the variance) clearly demonstrate the ability of PCA based on LIBS spectra to classify various types of gallstones. They also utilized the point detection ability of LIBS to study the spatial distribution of the major and trace elements including Mg, Mn and Ca etc. in the different parts (centre, shell and surface) of the mixed gallstone.
Fig. 4

PCA plot of training (plus) and test samples (asterisks symbols) of different types of gallbladder stones

Pathak et al. [104] also used PCA analysis in combination with LIBS data to distinguish the caries-affected parts of teeth from the healthy parts. A library of LIBS spectra (wavelength range 200–500 nm, with a spectral resolution of 0.1 nm) of different parts of teeth was prepared. PCA was performed on the data matrix (10 × 5,589). The first two principal components of the PCA analysis (PC1 contributing 96.70% of the variance, and PC2 contributing 1.60% of the variance) classified the different parts of the teeth within the dataset. PCA performed on the training data matrix (10 × 5,589) discriminated healthy and caries-affected parts of the teeth which clearly demonstrates the ability of PCA to discriminate various parts of teeth sample based on data generated with LIBS spectra.

In addition to PCA, there are also other methods also, for example PLS-DA, which have been used to analyse the entire LIBS spectra for the detection of chemical, biological and explosives materials. As also discussed above, Gottfried et al. [92] used a combined PLS-DA model for the detection of chemicals, biological materials and explosives using a single ST-LIBS sensor demonstrating the potential of stand-off LIBS for the detection of hazardous materials.

The most recent publications indicate that due to its unique capabilities, LIBS can be used to detect biological hazardous materials, and LIBS spectra provide an ample amount of useful information pertaining to the measurement of the molecular and cellular moieties. In such applications LIBS might presently play a supporting and leading diagnostic role. Improvements in LIBS sensitivity provided by new dedicated LIBS hardware, and in statistical methods (e.g. chemometrics, PCA analysis, and PLS-DA etc.) are expected to lead to improvements in LIBS-based discrimination of chemical and biological samples in the future.

Quantitative analysis of biological samples using LIBS and CF-LIBS

The ultimate aim of any spectrochemical analysis technique is to provide the concentration of a species present in a sample with high precision and accuracy. A quantitative analysis begins with determining the response of a system for a given concentration or mass of the analyte of interest. This usually involves developing a calibration curve, plotting graphs of the analyte signal versus the absolute concentration of the elements in standard samples. The calibration is usually strongly dependent on the analysis conditions, and several parameters can affect the LIBS signal, including the laser pulse energy and repetition rate, the lens-to-sample distance, detector parameters, and the parameters that are dependent on the sample and sampling procedures. The direct sampling of the materials with little or no sample preparation is one of the advantages of LIBS over other analytical techniques, because the physical and chemical properties of the sample can have a strong effect on the ability to obtain quantitative data. Depending on the application, some instrumental parameters can be held constant during data collection to maximize analytical performance, but the sample matrix effect is a major factor affecting the quantitative data of LIBS. Generally, certified reference materials (CRMs) having a known concentration of the analyte element in question and having a similar matrix to that of the unknown sample being analysed are required to produce the calibration curve. It is very difficult , particularly for biological materials, to obtain CRMs and with a similar matrix.

Fang et al. [71] used LIBS to analyse and identify the elemental constituents of urinary calculi. They quantified the elements Ca, Mg, Na, Sm, K and Pb from calibration curves. Recently, Singh et al. [74] applied LIBS to study cross-sections of kidney stones, and quantified the elements Zn, Cu, Sr and Mg from calibration curves derived from materials with a similar matrix. They have also verified their results by reference to the results of ICP-MS. However, sometimes it is very difficult to obtain materials with a similar matrix and in such cases other methods should be applied.

However, much work has been done to resolve sample matrix issues with LIBS. One approach is CF-LIBS in which the laser-induced plasma is assumed to be in local thermodynamic equilibrium and optically thin allowing the plasma matrix to be overcome. The system yields satisfactory quantitative LIBS data for elemental concentrations without using calibration curves and CRMs. Starting from the relative intensities of spectral lines, a family of Boltzmann plots corresponding to all constituents in the plasma is constructed. The concentrations of the constituents are then calculated from the intercepts of the lines on the y-axis. Finally, the concentrations of the observed constituents are added up to 100% of the material in the laser-induced plasma. The plasma is often not optically thin for strong lines and then spectral line self absorption becomes a major issue for the technique. Many lines terminating in the ground states of the elements often indicate self-absorption or self-reversal, and may be excluded from the analysis to avoid an effect on the quantitative results of LIBS. It is also important to determine an accurate plasma temperature and electron density. All the emission lines of the respective elements must be monitored and the oscillator strengths of the atomic lines must be known with good accuracy.

The application of CF-LIBS to mineral analysis of hair has been reported by Corsi et al. [58]. They measured the concentrations of the main minerals present in human hair from several subjects by CF-LIBS and compared the results obtained by CF-LIBS with the results obtained from a commercial analytical laboratory, and found good agreement. Finally, they explored its feasibility of CF-LIBS for the fast and inexpensive determination of heavy-metal poisoning in hair. Recently, Singh et al. [73] used CF-LIBS for the quantitative measurement of the metal present in the different parts of gallbladder stones (centre, shell and surface) of the different kinds (cholesterol, mixed and pigmented) . They also verified their results with the results obtained from ICP-AES which were in good agreement. Singh et al. [77] also utilized this method for the quantitative analysis of different kinds of salt samples and suggested that Saindha salt (rock salt) may be beneficial for patients suffering from chronic kidney disease. Pandhija and Rai [105] used CF-LIBS for the quantitative measurement of the concentrations of elements including Ca, Sr, Mg, C, O, Na, K and Fe in coral samples which contains organic constituents identical to those of the human skeleton.

Summary and future prospects

In this review we present the most recent developments in LIBS in the field of biomedicine. In the past decade there has been a burst of research activity in the use of LIBS for the analysis of trace elements in biomedicine matrices. As noted at the beginning of this review, LIBS is an effective technology with a wide range of potential applications in the detection and monitoring of major and trace elements in the human body, and LIBS technology has great potential for clinical practice. Many of these applications cannot be addressed using conventional analytical methods such as AAS, ICP, and XRF, but can be solved using LIBS. For the quantitative analysis of biomaterials where CRMs are available to prepare a calibration curve, the utility of CF-LIBS for determining the concentrations of the major and minor elements present in biological samples has been proven. Improving instrumentation, understanding the laser plasma, and data analysis are currently active areas of LIBS research.

Copyright information

© Springer-Verlag London Ltd 2011