1. The Beginning
Long ago I watched a movie called 'Medicine Man' (1992) starring Sean Connery [G5], playing a scientist deep in the Amazon who had stumbled upon the cure for cancer from an unknown ingredient harvested from the jungle. Amongst his well equipped jungle laboratory he had what he called a gas chromatograph, which had a CRT display on which vertical lines represented various chemical compounds. Compound 37 was the magical ingredient, but he couldn't identify the source. The movie wasn't a great hit, but it did inspire me in two ways - adventure and scientific discovery.
A decade or two passed whilst I was engrossed in various motorcycle projects followed by my own jungle adventures and then in 2009 I came across the movie again. I thought it would be really cool to own a gas chromatograph to see what things were made of, and started to look for one. It wasn't long before I discovered they looked nothing like the one in the movie, they required chemicals to work together with a significant depth of knowledge of chemistry, and they were very expensive.
I also discovered there were many established scientific methods that did much the same thing, each dismantling the fabric of matter into its elemental and chemical composition by various destructive means - chemical, flame, RF, Microwave, X-ray and laser, to name a few. All were expensive, but the one that seemed the simplest was the laser based LIBS system - Laser Induced Breakdown Spectroscopy, aka LIPS Laser Induced Plasma Spectroscopy.
The Mars Lander spacecraft rover has a LIBS system on board. The process is relatively simple - a laser is focused at an object and where it hits, the high temperature converts the material to a gaseous plasma state in which light is momentarily emitted in the form of wavelengths corresponding to the atomic signature of the elements that make up the object. A spectrometer spreads the light onto a linear array of optical sensors numerically corresponding to the emitted wavelengths, and a computer reads the array and determines which elements are present. http://www.msl-chemcam.com
Artists impression of NASA Curiosity Rover firing its ChemCam laser. Credit: NASAJPL-CALTECH
LIBS explained: http://en.wikipedia.org/wiki/Laser-induced_breakdown_spectroscopy
Spectrometers explained [O22]: http://www.astro.caltech.edu/~george/ay122/Ay122a_Spectroscopy.pdf
However I could not afford even one of these. The cheapest I found was the PortaLIBS system from Stellarnet.us (link and specifications further down) which at the time consisted of a small laser and up to 6 spectrometers each costing $3000, with a total system price of around $30,000 (2016):
It has always been the case that if I really wanted something I could not afford, I would try to build it for myself, and thus my project was born.
ADDENDUM
During the time I have been researching and building my project, I have found few amateurs such as myself exploring LIBS and associated laser physics at the necessary depths. Most of my knowledge has been acquired either from determining the operation and use of various instruments on eBay USA, or from scientific papers published on the Internet. I have found Internet servers typically save data for around 6 years before deleting it and many of my research notes reference observations from past papers that are no longer visible.
Partly as a result of this, I decided to collate all of my notes onto this website, as well as embed documents I found along the way, so that others in my position should be able to learn from my own research and save a lot of time.
Speaking of time, my project has no timescales and at any time this website provides a snapshot of the current stage of my research. Expect therefore, that ALL pages may change content. In particular, you will find some pages written in the present tense but others in the future and if you return later, issues I had encountered before may have been revisited and a resumé or conclusion presented instead.
To those without a scientific background: please don't be put off by the huge number of acronyms - I've tried to put them all in my References Glossary subsection. I would have preferred a hover help that automatically shows what they mean, unfortunately Wix doesn't support this. As a workaround to save endless hunting back and forth, I recommend you keep my glossary page open in a second browser tab:
DIY LIBS SYSTEMS
2022 - [L46] Finally a viable atomic line-based DIY LIBS setup using a Nd:YAG tattoo laser, BW TEK BRC100 spectrometer and DIY LIBS software. Analysis is by visually comparing captured spectrographs by overlaying them with NIST elemental library spectrographs.
https://www.reddit.com/r/physicsgifs/comments/s43zfa/homemade_libs_laser_induced_breakdown_spectrometer/
The above link leads to the following YouTube video: https://youtu.be/2C7Hajgg9nU
This video describes the LIBS process and his design; he explains in detail how to calibrate the
spectrometer which is also possible using the C# software. The signal enters The spectrometer via a 100µm slit. An IR PD detects the laser, triggering the Toshiba linear CCD gating pulse and the electrical signal is transferred to an Arduino type processor board. The author's replies in the comments section provide additional information: 'The triggering gets delayed by the software. Right now it is using a 5ms delay. I would have to calculate the theoretical resolution. The TCD1304 has 3648 Pixels. At the moment the spectrometer is only sensitive in the 400-600nm range.'
This extra page with YouTube video https://www.youtube.com/watch?v=D1HU-7zP9xo briefly describes the
hardware: 'The linear CCD is connected to a Nucleo [-F401RE Arduino] board using Esben Rossels PCB'.
This page describes the original spectrometer hardware: https://erossel.wordpress.com/page/2/
The initial software was written in Python by Esben Rossel and is located here:
https://tcd1304.wordpress.com/pyccdgui/ and here: https://github.com/jfsScience/jfsOtterVIS
'The software is an extension of the pyCCDGUI.py from Esben Rossel and is based on his
spectrophotometer (OtterVIS LGL) which uses the TCD1304 chip.'
The next version was written in C# by 'g3gg0.de' and is located here:
https://www.g3gg0.de/wordpress/uncategorized/a-homebrew-libs-software-written-in-c/
This page also has further descriptions of the hardware and software.
The STM32F401RE Nucleo-F401RE µC board: https://www.st.com/en/evaluation-tools/nucleo-f401re.html
2014 - A DIY LIBS setup using a miniature Nd:YAG from an M1 tank rangefinder and a Nikon DSLR to capture atomic line colour bands reflected from a DVD:
https://www.youtube.com/watch?v=ZzBSCrS8FAY&t=113s
COMMERCIAL LIBS SYSTEMS
Below are three low cost commercial LIBS systems I found in 2016. These all employ multiple spectrometers each covering a short bandwidth range at a resolution of typically 0.1nm. The number of spectrometers varies between 4 and 8, and the Stellarnet November 2016 website price for a full bandwidth 0.1nm resolution 5-channel 190-1000nm system is $35k.
I say they are all low cost, but I do not have prices for the other two. Generally I have found if a price isn't displayed you can immediately assume it isn't cheap but I am assuming as they all utilise similar technology, they are competitors with likely similar prices.
When it first appeared in 1995, my LIBS spectrometer alone probably cost a whole lot more than a modern multiple-spectrometer system as it is a Roper Scientific / Acton Research / now Princeton Research (obsolete) Spectrapro SP-275 triple grating Czerny-Turner monochromator with an intensfied PhotoDiode Array (turning it into a spectrometer), which would have been very expensive:
http://en.wikipedia.org/wiki/Monochromator
Grating G1 = 150g/mm Blaze = 500nm lined, Range 600nm (e.g. 200nm - 800nm)
Grating G2 = 300g/mm Blaze = 300nm lined, Range 300nm (e.g. 350nm - 650nm)
Grating G3 = 1800g/mm Blaze = holographic, Range 40nm (e.g. 480nm - 520nm)
G3 has a resolution down to ~0.08nm, but this only covers a very short range in the order of 40nm, and to identify elements, it is necessary to manually scan multiple bandwidth ranges to achieve the same coverage as modern multiple spectrometer systems that cover all possible wavelengths. Thus the operator must be aware of the range of wavelengths likely to identify the element and must pre-select this range before attempting analysis. This time-consuming mechanism was the first that appeared on the LIBS market, before it became affordable to parallel up multiple spectrometers. Top notch systems are more likely to employ an echelle spectrometer which is capable of covering the entire range.
http://en.wikipedia.org/wiki/Echelle_grating
Stellarnet LIBS PortaLIBS (full range $35k, November 2016 website prices) (up to 8 spectrometers)
http://www.stellarnet.us/systems/porta-libs-2000-and-plasma-monitor-configurations
Specifications DWARF-Star NIR Spectrometer $13,125
Laser Type: Pulsed Nd-YAG @1.06um
Laser power: 25mJ (4ns pulse)
Laser & rep rate: Kigre MK-367, 1Hz
Power to sample: 6 MegaWatts
Laser lifetime: >300k shots
Spectrometer channels: Expandable to 8
Spectrometer ranges: 190-1000nm wavelengths
Optical resolution: HR=0.1nm or SR=0.2nm
Detector range: 2048 pixel CCD array
SR wavelength range: 200nm=200-400/400-600
HR wavelength range: 100nm=200-300/300-400
Additional channel: SR=$3000 HR=$3500
Plasma chamber: 2 x 2 x 3 inches
Dimensions: 7 x 14 x 18 inches
Typical setups:
LSR1 $19,995: 1 channel, 200nm range (SR = 0.2nm resolution)
LSR2 $22,995: 2 channel, 400nm range
LSR3 $25,995: 3 channel, 600nm range
LSR4 $28,995: 4 channel, 800nm range
LSR5 $31,995: 5 channel, 1000nm range
LHR1 $23,995: 2 channel, 200nm range (HR = 0.1nm resolution)
LHR2 $27,995: 3 channel, 300nm range
LHR3 $30,995: 4 channel, 400nm range
LHR4 $34,995: 5 channel, 500nm range
- - - - - - - - - - - - - -
Ocean Optics LIBS 2500 plus (up to 7 spectrometers) (a used LIBS 2000 on eBay went for $10k in 2015)
http://oceanoptics.com/product/laser-induced-breakdown-spectroscopy-libs
Specifications
Laser options: 200mJ or 50mJ standard; other energy options available
Wavelength range: 200-980nm (individual channels are for the 200-980nm range)
Optical resolution: 0.1nm (FWHM)
Integration time: 1 ms; variable in free-run mode
Trigger delay: User-configured in intervals as short as 500ns
Trigger jitter: ~20ns
Trigger level: TTL not to exceed 5.5V
Dimensions: 13 x 6 x 5.5 inches (7-channel system only)
1 LIBS-CH-A Spectrometer channel with 200-305nm wavelength range
2 LIBS-CH-B Spectrometer channel with 295-400nm wavelength range
3 LIBS-CH-C Spectrometer channel with 390-525nm wavelength range
4 LIBS-CH-D Spectrometer channel with 520-635nm wavelength range
5 LIBS-CH-E Spectrometer channel with 625-735nm wavelength range
6 LIBS-CH-F Spectrometer channel with 725-820nm wavelength range
7 LIBS-CH-G Spectrometer channel with 800-980nm wavelength range
- - - - - - - - - - - - - -
Avante LIBS, sold in UK by Anglia Instruments, no prices
http://www.angliainst.co.uk/libs-systems.asp
Specifications
Laser Type Big Sky Ultra Q-switched Nd: YAG (Class 4)
Wavelength 1064nm
Energy/Pulse 50mJ/pulse
Repetition Rate Up to 20Hz
Spectrometer spec Wavelength Range 200-900nm
Resolution <0.1nm
Detector Type CCD, 2048 pixels per channel
Integration Time delay Minimum 128µsec, 21ns steps
Integration Time Minimum 1.1ms
Dimensions 7 x 4.3 x 1.73 inches (1 channel)
THE START OF THE PROJECT
I started the project around Spring 2009 and since then I have learned and amassed a huge amount of information (my own project log numbers over 500 pages), much of which is collated on this website. Until that date I had designed, built and programmed many electronics projects both at work and for myself, but none with lasers.
From the start it was obvious I would need a means of verifying wavelengths and after some research I bought what I considered the best value performance instrument: a parallel port Stellarnet EPP2000 12-bit Super Range 200nm - 1080nm spectrometer with a 25µm slit giving 1.5nm resolution.
Later I had it upgraded to USB with 16 bits and the external electrical trigger option fitted, and it came back in the form of a Black Comet SR:
Over the many years equipment has been bought on eBay from a staggeringly large number of countries: Australia, Austria, Canada, China, Belgium, Bulgaria, Finland, France, Germany, Greece, Ireland, India, Israel, Italy, Korea, Lithuania, Netherlands, Philippines, Poland, Russia, Singapore, Thailand, the UK and Ukraine, but mostly from the USA.
EBay has proven to be a great source of information, not just in locating project parts, but also by exposing the world of laboratory lasers and analytical instruments, many of which I had never seen before: each time I came across a new instrument I went away and found out how it worked. I have often likened the project to the equivalent of a couple of degree courses, except unlike a real course I keep all of the equipment, the downside being the paucity of answers to my many searching questions.
Just after I bought the EPP2000, I saw an auction on eBay for a PCM403 Plasma Chemistry Monitor, below, consisting of an Acton Research Sp-300i spectrometer coupled to a PC running software that could identify up to 80 compounds from their spectral signature. It came with a mercury (Hg) calibration lamp and it even had installation floppies. I realised the potential to match it with a laser to build a LIBS system.
Unfortunately the bid starting price was $7000, well beyond my budget - I was still reeling from the $3000 I paid for the Stellarnet, but given its quality and the exceptional level of support I've received over the years, I don't regret this.
I sadly watched the auction end, instead Googling 'LIBS' and reading as many papers I could find on the subject. I learned 1064nm Nd:YAG lasers were often used, and Stellarnet's own PortaLIBS system employed a 4" long 25mJ Kigre MK367 1064nm Nd:YAG [D1]. By chance, I found a couple for sale in Belgium and bought them, together with a power supply the seller had kindly knocked up to verify their operation. The Kigre MK367 needs about 4.5 Joules at 673V to reliably fire its flash lamp at up to 1/3Hz, and having its own internal passive Q-switch, produces a 6MW (6 megawatt) 4ns wide pulse of light.
Impressive specs, but much of it meant very little to me, and there was scant information on the web about the MK367 or how to drive it.
I contacted Kigre, to whom I am indebted for the most generous advice and guidance that cemented my determination to build my own LIBS system. President Michael Myers is a laser pioneer and a hero figure for me, his name forever cropping up in ground-breaking papers together with Chris Hardy his chief engineer, who provided far beyond my initial request for help, despite both being fully aware my project has no commercial objectives.
Chris forwarded a copy of Perkin Elmer's design guide for flash lamps, the optical energy source inside the MK367 long before laser diodes became available, together with suggestions how to drive it. The MK367 issue 4 datasheet [D1] gives more information than was available in 2009. The Perkin Elmer / EG&G Flash Lamp catalogue is also a very useful resource, containing a considerable depth of information as well as example circuits and equations [O11]. Flash lamps are peculiar non-linear devices to drive and these equations are essential to prevent damage and potential injury from an exploding flashlamp.
Essentially I needed to charge a capacitor to a high voltage that would appear across the flashlamp, which would ionise when a 10kV pulse was applied to its trigger.
An inductor between the capacitor and the flashlamp terminal would modify the shape of the current waveform taken from the capacitor when triggered, 'tuning' the pulse to the flashlamp characteristics and removing the danger of explosion.
The combination of the capacitor and inductor is referred to as a Pulse Forming Network (PFN).
Chris confirmed a 20µF capacitor and an air-cored 20µH inductor would suffice. Armed with this information, I could build my own fixed voltage power supply, but my system is more about experimentation than simply building it and moving on to something else.
I therefore went back to eBay and bought an adjustable Wilmore 1515 1200 Joule capacitor charging PSU so I could experiment with different flash lamp voltages. Due to its age this massive but highly reliable PSU was only $200. I also bought a Power Designs 2k-10 2kV 10mA PSU [I41] for $80, to experiment with different flash lamp trigger voltages. Next I bought a couple of large EG&G TR-132C trigger transformers [D28], $30, and four huge 20µF 1kV Arcotronics capacitors for a bargain $20. All of these are overkill for the little MK367, but deliberately over-spec to accommodate future experiments. The trigger transformer only needs about 140V, but the 2kV PSU is selectable in 10V steps, with a precision potentiometer for fine adjustment, ideal for experimentation.
Lastly, I purchased a cheap Chinese LC tester kit on eBay, added a box and a cheap Russian precision 1nF 0.5% film capacitor for calibration, and wound my own 20µH inductor with 22AWG magnet wire on a cotton reel.
PHILOSOPHY
At this point you may well ask why I am contemplating using flashlamps when the age of semiconductor diode lasers is upon us? The answer is simple, it is for education and fun, and for the challenge.
On September 12 1962, President John F. Kennedy announced the decision to aim for the moon with the rallying words: 'We choose to go to the moon in this decade and do the other things, not because they are easy, but because they are hard, because that goal will serve to organize and measure the best of our energies and skills, because that challenge is one that we are willing to accept, one we are unwilling to postpone, and one which we intend to win, and the others, too'.
This too is my philosophy. I choose to learn and I choose to face difficulty in this process.
Not being able to afford the latest gizmos also plays a certain hand in this decision.
New text box