Rock Analysis With a Laser

Send to Kindle

By Ely Silk

So, you’ve just come back from your nature walk carrying some interesting looking rock samples.  You probably figured that with all those field guides and charts at home, identification will be a breeze.  Rummaging around your closet, you find that hardness of minerals kit you got as a kid on one of your museum field trips.  The kit often included a streak plate.  Assuming the kit and plate are properly used, my experience indicates that you can narrow the approximately 6,000 possible choices of rocks and minerals down to 5,980 or so.

If you’re planning to run chemical analyses on your samples, you will first need to finely crush representative parts of the unknown rock and subject the resulting powder to extremely corrosive acids—if you are able to buy them anymore.  This is followed by a series of qualitative tests for different cations.  These tests are a chore to set up and also involve chemicals that are getting tougher to acquire.  You then have to dispose of the spent chemicals.

Traditional spectral analysis using a carbon arc and an associated spectrometer is certainly possible, but it requires lots of experience to get accurate results.  Some elusive components in your specimens may be quite volatile, and their spectral lines could vanish before you spot them…or think you spot them.

If only there were a better way to quickly and reliably determine the unknown’s chemical composition, identification could be simplified and…fun!

With the advent of pulsed lasers, in 1962/63 a new technique was introduced:  laser-induced breakdown spectroscopy (LIBS).  Though it got off to a slow start, it is now the subject of intense investigation and application in analytical labs worldwide.  NASA has placed a LIBS system aboard its Mars Science Laboratory Curiosity rover.  Typically, commercial systems range in price from $50k on up.

Image by Ely Silk.

Why so expensive?  A typical LIBS apparatus consists of a series of narrow-band spectrometers with resolutions of 0.1nm or less.  Each spectrometer is equipped with a highly sensitive photodetector, such as a high-resolution CCD chip.  Enough spectrometers are used to adequately cover the spectral range of interest.  As an example, to cover the range from 200nm in the ultraviolet to 800nm in the near infrared may require six or seven spectrometers.  Alternatively, a single echelle spectrometer can be used.  This type of spectrometer can image the entire spectrum on one detector with very high resolving power.  You will also need a powerful pulsed laser usually emitting at 1064nm, focused to produce a plasma plume on the analyte, the material being analyzed.  The laser pulse duration should be on the order of 10 nanoseconds or less.  Add to this, broad-range fiber optics, standard optical components and mounts, electronics to delay the detected light pulse, sample mounts, a computer with spectroscopy software, etc., etc., etc.

What is a citizen scientist to do?  Multiple spectrometers are used because the resolution is inversely proportional to the spectral range covered.  You can, of course, purchase a standard USB spectrometer that covers the entire range from 200nm to 800nm or beyond, but its resolution may be no better than 1nm.  Nevertheless, I decided to try that approach to see what could be done within its limitations.  The spectrometer I use has a stated resolution of 0.8nm.  But then there’s the laser requirement.  I happened to have one lying around that could do the job.  I bought the system years ago for some experiments with dye lasers.  However, that laser is getting old and less energetic.  Besides, this won’t help experimenters who need to buy a laser.

I hunted around and finally decided to evaluate the relatively inexpensive imported cosmetic or tattoo lasers.  Whereas a commercial pulsed Nd:YAG laser marketed for LIBS work can run upwards of $25k, the imports are priced at under $3000—often way under.  I acquired or built the necessary electronics for handling the pulses produced by the plasma plumes.

The following chart shows a sample spectrum.

The total cost for the system shown here is under $7000.  This is for brand-new, off-the-shelf components direct from the manufacturers.  Just think of the possibilities of being able to perform analyses with no muss, no fuss.  Place your specimen and zap it.  The spectrum immediately appears on the screen!

I just added a section to my Web site that goes into greater detail on my experiments with both electric spark spectroscopy and LIBS.  The section includes a very brief history of spectroscopy as well.  I also touch upon the nature of the analysis that the experimenter must perform to figure out what those peaks on the computer graph are saying.  Here’s the link:

www.viewsfromscience.com/documents/webpages/methods_p6.html

One of my main objectives in building my Web site is to develop techniques and lower-cost procedures for different areas of science to meet the needs of citizen scientists, as well as professional scientists, researchers in developing countries, small labs, and educational centers.

I hope that this discussion helps convince you to get involved with a technology that could make your chemical analyses enjoyable, if not a breeze.  LIBS is a fantastic and fun way to perform chemical analysis on materials ranging from glass and ceramics to rocks, meteorites, works of art (arrivederci, Mona Lisa), and archaeological treasures.  And that’s just for openers.  The applications are endless.

Finally, let me leave you with this warning: Always use proper eye protection when using any laser, especially a pulsed Nd:YAG laser.  It can permanently blind you before you can say, “Who turned out the lights?”

I wish you happy (and careful) experimenting!

This entry was posted in Amateur Science, Geology, Invention, Physics, Projects. Bookmark the permalink.

5 Responses to Rock Analysis With a Laser

  1. This is a fascinating project.I have a spectrometer and may look into this.

  2. Marky says:

    It looks like you are using the standard atomic emission (AA or AE) spectral lines. Is this correct?

    Are there any special interpretation issues? I see amplitude used. Most instruments use absorbance. How would you convert your amplitude measurements to absorbance?

    Very good project. Thank you for the information.

    • Ely Silk says:

      As they were produced by atomic emission, the peaks are measured as amplitudes. Interpretation involves sitting down with spectral line tables and meticulously evaluating each peak. Absorbance scales are necessary for instruments that measure atomic absorbance. Typical chemistry lab spectrometers that are used for determining the concentration of liquids are calibrated in units of absorbance. There’s no way to convert emission amplitudes into absorbance. You may be thinking about transmittance rather than emission. If I were measuring, say, the light transmission of glass filters, then units of transmittance or absorbance would be the right ones to use. I hope this helps.

  3. Marky says:

    Thank you for the reply.

    The AA and AE instruments I’m familiar with used a logarithmic scale in both aborbance and emission modes. I’m not familiar with LIBS and didn’t know if that was normal or not.

    Is there a website that gives more specific instructions on the construction of the instrument?

    With the dramatic reduction in price your instrument shows, you should be getting calls from some instrument makers soon. Your project is phenomenal. Congrats.

    • Ely Silk says:

      Absorbance is tied to a fixed light source in the spectrometer. It is meaningful to give absorbance (or transmittance) values with reference to that light source. So, you can say that the liquid in a cuvette absorbs 40% of the light at 589nm or such. Emission is an open-ended scale. You can have emission that is barely visible up to a quasar in intensity. Again, if you tie it to a particular calibrated light source, then it may be meaningful to calibrate an emission scale. With LIBS, your light can vary all over the place and, even if you could, it wouldn’t be meaningful to have a calibrated scale. However, if you are planning to do quantitative work, then you can reference a particular amplitude at a particular wavelength between samples. This will give you an idea of relative concentrations of a particular element. But, for that to work, the system must be calibrated with different samples because not all emitted lines behave the same way when concentration varies.

      As far as constructing an instrument, I hope that you first visit the link to the LIBS section of my Web site given in the article. In the references tied to the LIBS section, you will see books that go into detail about these systems. (I always remind anyone interested in working with pulsed Nd:YAG lasers to observe extreme precautions. Just as the laser beam can quickly vaporize rocks, it can instantly destroy an eye or two. Eye protection for the user and others who are nearby is mandatory. I don’t go near my system without goggles!)

Add Comment Register



Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>