LIBS - Science method

Hi Qing Xing Peng

If you want to identify specific CH₄ characteristics from your recorded LIBS spectra, from what I can see, there are two main approaches:

Using atomic emission lines:

Hence, by analyzing the C:H intensity ratio from the LIBS spectra, you can determine whether it is consistent with methane's molecular structure (1:4). Note that, due to various experimental factors, the LIBS intensity ratio would likely deviate from the exact 1:4 ratio. Therefore, it is necessary to establish baseline ratios by developing a calibration curve from known methane concentrations. From this calibration, a correction factor can be derived as follows: (C:H = 1:4) / (I_C:I_H).

The intensity ratios of these molecular bands can also provide insight into methane presence and concentration.

Additionally, consider the detector time delay at which these spectra are recorded: diatomic molecules recombine later (typically, of the order of a few tens of microseconds) than atomic species. Thus, optimizing the gate delay of your LIBS detection system is crucial to achieving reliable results.

The answer to this would be copy/paste of an entire spectroscopy textbook (except for XRD, because it's no spectroscopy). Especially when it comes to the entire basic theory, a forum post like this is not the reight format. So please read a basic introduction and if you have a morep precise question, return here.

Has your sample signal been normalized? The differences do not appear to be very pronounced. Additionally, why have you chosen to use Ti 441.492 nm? It is clear that this is not a commonly used spectral line for Ti. There are many strong peaks for Ti that could be used for univariate calibration, such as Ti I 365.35 nm, Ti I 375.29 nm, Ti I 399.86 nm, Ti II 323.45 nm, Ti II 334.94 nm, etc. Why have you not selected these? Are there other interfering peaks? Self-absorption occurs when the Ti concentration is too high, and in such cases, it is advisable not to use easily excited characteristic peaks as they are more prone to self-absorption. You can simply determine whether self-absorption is occurring by checking for peak center dips or splitting. If you need to quantify the degree of self-absorption, it can be calculated using the appropriate formula.

Also, I am dealing with a single isolated peak. So, there is no need for deconvolution.

You can try Standard Normal Variate (SNV) normalisation. You should subtract the mean of each spectral feature (in this case intensity for each wavelength) and then divide it with standard deviation. In other words, you find the mean and std of all intensities for each wavelength, and then do the aforementioned subtraction and division.

This should bring each spectrum to the mean of zero and std close to one. It should look better.

Averaging is advisable to reduce LIBS reproducibility error. Every LIBS spectrum is actually average over 10 or 20 successive laser shots. Very often authors do 10 shots on one place, then repeat it for lets say 20 other places on the target. Then the final spectrum is the average of all 200 spectra. On these spectra, calibration models are often constructed.

LIBS (Laser-Induced Breakdown Spectroscopy) differs from XRD (X-ray Diffraction) and XPS (X-ray Photoelectron Spectroscopy) in a few key ways, particularly the information they provide and how they interact with the sample.

Here's an analogy: Imagine you are analyzing a cake.

In short, LIBS excels at elemental analysis, XRD for crystal structure determination, and XPS for surface chemical state investigation. They are complementary techniques often used together for a comprehensive material characterization.

Hi, Prof. Lianbo Guo has done many works on the determination of self-absorption coefficients using the LIBS spectrosccopy, for example，Accuracy improvement of single-sample calibration laser-induced breakdown spectroscopy with self-absorption correction.

Hope this helps.

Best regards,

Xianshuang

If tablets break, you may reduce the energy of your laser pulses. Otherwise, a femtosecond laser will be gentler than a nanosecond one.

Plasma emission occurs due to three types of transition in plasma that are free free, Free bound and bound bound transition. All the transitions provides various emission out of which bound bound transition provides line emission and others are continuum emission. All these processes are manly dependent on the plasma density and its temperature. Therefore, a continuum emission also can provide information about plasma density and Temperature.

It may be the result of line absorption in plasma.

Muhammad Khisham have a look: https://gmd.copernicus.org/articles/14/675/2021/gmd-14-675-2021.pdf

SEM-EDX can perform elemental analysis on the surface of the sample

I find your intentions very interesting and very noble.

As you may have seen, there are no online databases of LIBS spectra built on standards that I know or remember. but reading in the comments you will find several groups, among which I also add myself, willing to provide you with their material and all the help you need.

The reasons depend on the fact that the libs spectra vary greatly according to the different instrumental parameters and measurement conditions.

However, there is a very useful NIST portal, where you can simulate spectra of single elements or composites by also varying the parameters of the plasma.

It might come in handy if you plan to build a training database for a neural network. The help you will find in the NIST database is in the ability to simulate extreme conditions regarding plasma parameters and even with reference to LIBS spectrum resolutions. Information that can be very useful in the training phase of the neural network.

I hope that you will be able to carry out this idea of yours which seems to me really interesting and worthy of consideration by the scientific community that gravitates around LIBS spectroscopy.

Hello,

We have performed LIBS in argon atmosphere for quite the same reason, see: . Here, we have used a vacuum chamber with a "static" filling of argon at defined pressure (i.e., evacuation of the chamber followed by filling with argon up to a desired pressure instead of permanent flushing by an argon gas flow). See also: . This approach worked fine!

I hope this helps!

Best

Christoph

Hello, The larger the observed plasma volume, the better. I thus reommend to align the observation path as perpendicular to the sample surface as possible. In doing so, even a single best form lens may be sufficient in some cases... But generally, the amount of required light also depends on the particluar application and the used spectrometer...

Hello,

I do not really know about papers where such an equation was published. But this is really an interesting question.

However, it cannot be answered that easily since the function T vs. t strongly depends on the plasma, i.e. the laser parameters (fleunce etc.), the material (solid, liquid etc.), the ambient conditions (pressure, gas etc.) and the ablation regime (phase explosion, LTE or not etc.).

In 2014, we have published a set of data (see Table 3 in [C. Gerhard, J. Hermann, L. Mercadier, L. Loewenthal, E. Axente, C.R. Luculescu, T. Sarnet, M. Sentis, W. Viöl: Quantitative analyses of glass via laser-induced breakdown spectroscopy in argon, Spectrochimica Acta Part B 101 (2014) 32-45, DOI: 10.1016/j.sab.2014.07.014]. I have plotted the values reported there; see the attached graph.

For the given specific experimental conditions described in that paper, the dependency of temperature over time can be estimated by the exponential decay shown in the attached figure…

I think that in most cases, it is an exponential decay as a general approach.

I hope this helps?

Best,

Christoph

Hi Muhammad

That's of course a tricky one.

Ideally you need a sample containing no detectable amount of your analyte (call it a blank) but with the same composition of the sample (at least close enough that the LIBS plasma has nearly the same composition). This would allow you to calculate the SD of the baseline emission intensities for your analyte.

Alternatively you may model the spectrum, fit a baseline under the emission line of your analyte and use the SD of the fit. This could however underestimate the baseline signal in case of spectral interference.

In practice you may also use the sample with the lowest concentration in your calibration set and analyse it as unknown. If the results are in agreement with the reference value, you can use this as the upper boundary of the actual LOD value.

Hope this helps.

Bodo

Dear Dr. Q Iraqi

Sorry above reply is wrongly pasted here.

Actually LIBS is very much suitable to identify and quantify the elemental analysis from solid or liquid sample. Ratio of the elements can be identified by the ratio of their line intensities. However Boltzmann plot is used to find the plasma temperature.

thank you Ijaz Durrni

In the case of Nano second laser, interaction time is more that increases the thermal diffusion length in the material that is less in the case of Femto second laser. This produces more material ablation in the case of nanosecond laser than the femto second laser ablation. This makes the plasma emission more intense in ns-LIBS than fs-LIBS. More melting of material in ns- LIBS makes it in efficient for depth profiling. In fs-LIBS ablation mechanism is different and fast that makes very sharp hole in material. Thermal diffusion is also very less that makes it very much suitable for depth profiling. However, ns-LIBS is much suitable for elemental analysis. Another most important application of fs-LIBS is known as FIBS (Filamentation induced breakdown spectroscopy). FIBS can be able to provide elemental analysis from a sample located at more than ~ 90 meters away. This distance is many times more than the distance reported for Stand off LIBS.

@ Ion Laurentiu Tobos - Could please explain a little more... I dint understand your answer.....@Ion Laurentiu Tobos

Your question is not clear. What do you mean by spectrometer angle.

Thanks Sandipta Kumar Sahu your reply provided me a new insight. I have removed -DGMX_PREFER_STATIC_LIBS=ON and rebuild OpenMPI with its library.

Please see following paper

Thank you Tomáš Vrzal and Steve Driscoll for your suggestion

Regarding increase/decrease of electrolyte resistance, it depends on the electrolyte system being used and the chemistry involved in electrochemical reaction. Like in Lithium Sulfur chemistry, sulfur converts into polysulfides which subsequently dissolve into electrolyte altering its resistance. Secondly, you have to look whether change in resistance is small or big. Minimal changes are difficult to interpret while significant changes can be explained by understating the chemistry of the system. So YES

Regarding diffusion resistance, its common to observe that after multiple lithiation the electrode becomes porous making it easy for lithium to diffuse. Moreover, if micro-size particles are used for electrode preparation, they can break down into smaller particles after multiple lithiation/delithiation, causing diffusion resistance to decrease. So YES

Please look at the following below links which may help you in your analysis:

https://www.sciencedirect.com/science/article/pii/S0039914013000143

https://www.ncbi.nlm.nih.gov/pubmed/21962699

Thanks

In light of above discussions, It is clear that only two methods are possible for getting concentration of unknown samples, First is based on calibration curve and other on is calibration free technique. For getting a good calibration curve with minimum error, one should plot a calibration curve of required element in the similar matrix samples in similar experimental condition.

Dear Mr Lee,

I read that your project’s goal is to build a portable LIBS technique for detecting heavy metals in food and medicine. I also read all the recommendations you got from colleagues and your answers in return. I think that you are going in the right direction and you have all you need in your hands in your Department of Instrument and Engineering in your University in Shanghai.

From my side, I am sending you a good article from well-known D.Cremers et al. published in J. Viz. Exp. (2013) demonstrating the LIBS with correlation between the detection limit and laser pulse energy and without correlation (see a copy in attachment). Then I recommend to look at the collection of recent LIBS-related papers in the latest issue of Applied Spectroscopy vol.72 (2018) http://journals.sagepub.com/topic/collections-asp/asp-1-laser-induced_breakdown_spectroscopy/asp?pageSize=20&startPage=0

Good luck!

When laser interact with the target surface it creates plasma. The plasma near focusing point is very hot and emits bremsstrahlung radiation, which is know as background emission. As the plasma expands away from the target surface it cools down along with decrease in the intensity of background emission. After plasma cooling first the line emission due to ions becomes observable and its intensity increases with time. After further cooling line emission from the neutral atoms becomes observable. Now one has to decide and optimize the gate delay and gate width of the detection system to get a better S/B signal either for ion or neutral line emission of interest.

Signal to noise ratio is dependent on many factors as has been discussed above. However a better detector can be able to reduce the noise. One has to take care to reduce the other noises coming from other sources.

Dear Maulana,

I recommend that you read the review of V. Strelko "Role of carbon matrix heteroatoms at synthesis of carbons for catalysis ana energy applications" published in Journal of Energy Chemistry 22 (2013) 174-182.

Good luck,

Sergii Kuksenko

Dear Sagheer

Please read carefuly that paper. If they have used argon gas in the experiment. They must have explained the reason for using that gas. However in general argon gas is used in LIBS experiment to enhance the signal (sensitivity) of the LIBS. The shock wave generated at the boundary of plasma and argon gas compresses the plasma, which increases the plasma temperature and density resulting in comparatively higher LIBS signal.

Dear Cristhian

I agree with Prof. Baig and Dr. Marijn. LA-ICP-MS system as well as its maintainance is costly. It has been proved that LIBS works better even for the geological sample and is faster than the LA-ICP-MS. Finaly you have to decide keeping all the points in your mind in the present situation.

Hi! You could also consider Applied Photonics Ltd. We have more than 20 years experience in the design, development and manufacture of a very wide range of LIBS systems. They are truly modular, therefore easily upgradeable and expandable and with really competitive prices. You can visit our website www.appliedphotonics.co.uk and contact us for more details at mail@appliedphotonics.co.uk so we can discuss the available options that could suit you.

The Galactic GRAMS software used to have a very nice interactive interface to perform this type of curve fitting. I am not sure which Company owns "GRAMS" nowadays

The sample matrix is a major factor affecting the accuracy of LIBS quantitative results. The “matrix effects” are due to changes of emission line intensities of some elements when the physical properties and/or the chemical composition of the sample matrix varies. These effects primarily influence the production of atomized material by ablation and the excitation properties of the laser plasma. There are two kinds of

matrix effects: physical and chemical. The so-called “physical” matrix

effects are related to the physical properties (e.g., grain size, texture, reflectivity and hardness) of the sample surface on ablation parameters.

These effects may result in an alteration of the amount of ablated

mass which may cause a variation of the line emission intensity even

if the concentration is the same in the various matrices. The “chemical” matrix effects occur when the emission behavior of one element is altered by the presence of another one. For example, an element present in equal concentration in two different host materials may exhibit different LIBS emission intensities. These effects make it very challenging to find matrix-matched standards to be used to perform quantitative LIBS analysis of natural samples. ciao Giorgio

It looks like coma optical aberration in your Czerny-Turner type spectrometer with spherical mirrors occured due to off-axis optical path. You may try to affect slightly the line shape by turning an input optical fiber through some angle around the axis. If it have no effect - the problem is inside the box. In the last case the best solution could be a request to manufacturer.

Hello,

There are many papers describing the "free-calibration" LIBS, where the plasma parameters are used to elementa concentration:

I hope this information helps

best regards

Luis

Many many thanks dear sir.

Regards

Saifullah Jamali

Dear Muhammad,

electron temperature was determined from Boltzmann plot and electron density was calculated from stark broadening .

Best,

Hello Saifullah,

To determine the elements or impurities in your sample, it is necessary to have the data in order to enter it in an identification software.

I recommend you compare the position of your peaks with the NIST database.

If you are interested, I could also send you an elementary identification software that we have developed in our group for the LIBS technique

regards

Luis

LIBS is a spectroscopic technique capable of doing many tasks. As already stated in the previous answers, it can measure plasma temperature and electron density. They can give a good characterization figures of the plasma. This plasma (LIBS) can be used to perform the direct chemical characterization of solid samples without prior sample preparation. LIBS can also be used for nano-material preparation under liquid to be an environmentally friend preparation procedure. LIBS has many other applications in science and technology.

I think Mohammed Ubaid Hussein mixed up two different things here - the electron density in a plasma, which is the number of electrons per unit volume (and the entity, Muhammad Usman Khan wants to measure) and the electron probability function, which is indeed the quantum mechanical propability to find an electron with velocity v in a certain place at a certain time.

@Muhammad Usman Khan: If you want to measure the electron density with LIBS you will have first to make sure that your spectrometer is calibrated for wavelength and absolute intensity of your emission line. If this is the case, you can determine n_e, for example, follow the procedure outlined in this paper:

yes, we use the new techniques in the industrial line of cell through NEUI.

Dear Harikrishna,

Thank you for answer.

Yes, safe Memory Partition is available only in the main Controlling microcontroller.

I fully agree with you for the abstraction path but it's critical if I have another/multiple supporting µCs only for Extended the I/O Signal. Then what would be safety concept? How I can add/connect/ linked my I/O Signal from the supporting µC to the main µC's in AUTOSAR architecture!

Best regards

Eva

Please check this option. Is an amazing equipment

The only commercial software that I am aware of that includes a CF-LIBS module is sold by Marwan Technology as part of their MOBI instrumentation [link below]. It may be available separate to the instrument.  [I haven't used this software myself]

See Dr. Tavassoli's group:

In Germany it is Fraunhofer Institute for Laser Technology in Aachen. Mr Reinhard Noll is the key researcher, see also his researchgate page "Reinhard NOLL".there You will find his works.

As a coarse method to check the calibration, you can look at the light from a `common` fluorescent tube (the type that is used for lighting a room).  There are some Hg lines that are typically found and can be used for a quick check.  Look at the attached wikipedia link, they show some typical spectra with the peaks identified.

Hi, For massive amounts of data, initially a variable selection is performed based in the loadings of PCA calculated using matlab, after this approach the Pirouette software is a good option. Alternatively all analysis can be performed using matlab.

Regards.

Another useful one [which is not behind a paywall] is:

Lumps are inhomogeneous large agglomerates. In case of CMC they form when the powder is contacted with water. CMC on the outside of the lumps will start to swell and form a highly viscous layer around the lumps. Then it will be difficult to disperse them. You may avoid this by first forming a paste of CMC in a nonswelling but water miscible solvent. I think glycerol is commonly used.

If I am not wrong, the laser induced breakdown ablation is generally used for spectroscopic analysis by which one can identify the chemical nature of the studied target material. It is also known as LIBS. By using a short pulse duration it is possible to distinguish the basic process from the secondary effects if any.

What is the state of your Cr(CO)6? Is there a possibility that the complexes in particle phase is ablated and form phase-selective LIBS? And the breakdown process only occur in the positions where particles exist.

In that case, only Cr in particle phase are ablated but not gas phase. You can try to observe the atomic emission of Cr in particle phase. I predict that if the Cr(CO)6 is changed to gas phase, there will be no similar phenomenon

Indeed, we do observe similar process in TiO2 and V2O5. You can see my papers as attached below.

fs lasers have very small energy per pulse, making it only possible to ablate very small spots, making very small plasma with little emission. You need a good optical setup and high sensitivity spectrometer, compared to ns LIBS. But the small ablation per pulse can be advantage in some cases like thin layer depth profiling etc. I dont have experience with attosecond laser but I guess its the same

thanks very much for your reply but ı dont understand this code for example which is testfeature,trainfeature and class labels for y,x,cmd

A few years ago it was found an heuristic  equation that describes the product of ablation as a function of the fluence for metals and crystals. Perhaps, it can be  serve for your case.  JOURNAL OF APPLIED PHYSICS 104, 103112, 2008

Thanks Jeffrey M Girard.

Most basically McWhirter's Chriterion must be satisfied, that says

ne ≥ 1.6 *10^12 T^(1/2) ∆E^3

Hi there,

Without looking at the detailed specs of your laser and spectrometer I can't give you an exact answer. One way would be to synchronize both your laser and your spectrometer with a separate pulse from your delay generator. Probably your laser will be able to accept a trigger input that determines the time delay between the flash light and the the Q-switch, the laser pulse exiting the cavity at the end of the trigger pulse. If not, your laser may have an output that sends a signal just before each pulse, which you can use as a sync input to your delay generator. You then need to set up a waveform that will trigger your spectrometer to capture a frame; the spectrometer will have a sync input for that. Presumably the laser pulse will be so much shorter than the shortest integration time you can achieve with the spectrometer that in order to get the timing right you could start capturing before firing the pulse. But this is just me guessing. Ah - before I forget - your spectrometer may actually have a sync OUTPUT that can be used to trigger your laser. Some spectrometers provide such an output especially for breakdown or fluorescence spectroscopy. Anyway, the manuals will tell you. Hope this is of some help. If I've forgotten something I'm sure someone will add it...:P.

Good luck,

Marco

You should also know that the Herzberg databases (diatomic and polyatomic) can be found in the NIST site

This is what you find for H2

and this for ethene

You can also search the spectra.

I use a femtosecond LIBS system. Before I answer the question, please let me know the condition of your lab in terms of humidity and temperature stability.

You can also use a Collison nebulizer to generate aerosol droplets of a solution or suspension (e.g. http://www.bgiusa.com/agc/collison.htm). The inlet pressure is usually around 20 psig or higher. The droplet diameter is consistent and is a function of the inlet pressure.