Discover the First Dedicated Gunshot Residue Solution

Jeroen Smulders is the product manager of the Thermo Scientific™ Phenom™ Perception GSR and Phenom XL at Thermo Fisher Scientific. With a background in physics, Smulders has also worked for FEI and Phenom-World. Jeroen and Rogier Miltenburg, who has a background in biomedical engineering and is the product specialist for Phenom XL and Phenom Perception GSR, speak about the Phenom Perception GSR to AZoM

Please tell us about the Phenom Perception GSR and how it came onto the market.

The Phenom Perception GSR is quite small; it fits on the desktop. It doesn't look like a traditional SEM (scanning electron microscope), but it does have all the bits and pieces a floor standing electron microscope does.

The sample holder fits up to 30 GSR samples, plus reference samples that you can then use for calibrations, and so on. There is also a 24-inch monitor, where the Perception user interface is found. From here you control the whole microscope as well as the GSR workflow. You can do everything in that user interface; microscopy, EDS analysis, automatic GSR runs, reporting. Everything happens on that screen.

Phenom-World launched the GSR solution in 2015. At that time, it used the GSR software from Ken Mason. Mr. Mason has over 20 years’ experience in GSR and we collaborated with Mr. Mason to get a good understanding of gunshot residue and what customers want, and how it is different from general electron microscopy.

We introduced this in the market and had that product for a few years. To further development, we were already working together with FEI, with whom we had a product in collaboration. That was the Explorer 4 Analyzer which is also an automation platform for automated particle search.

In 2018, Thermo Fisher Scientific bought Phenom-World and we integrated to become part of Thermo Fisher Scientific. At that time, it was a very logical step to see how the collaboration of the Explorer with the Perception software could be extended to GSR as well. Together with all this knowledge, that all got combined under the umbrella of Thermo Fisher Scientific, we developed what is now the Thermo Scientific Phenom Perception GSR.

So far, it's been very well received in the market. It's already used in over 10 countries as we speak.

The Phenom Perception GSR

For those with less experience with GSR, what is it and could you give more information about the software?

GSR is carried in the cloud of smoke that is released when a gun fires. That smoke contains many particles, and some of these particles, but definitely not all of them, are very specific, very characteristic for the discharge of a gun. Traditionally, that was the lead (Pb), barium (Ba) antimony (Sb) particles, and they are the particles that we look for traditionally. Nowadays, there are some other chemical compositions as well that could be covered, but Pb Ba Sb is the most traditional chemistry.

How we analyze that is fairly well described in several standards, for instance, the ASTM (American Society for Testing and Materials) and the ENFSI (European Network of Forensic Science Institutes) describe what the best practices are and what the guidelines are on how to analyze these particles properly. That analysis is performed as follows.

You start with a pin stub typically with a carbon sticker, with the actual sample on top of that. That sample is typically 12 millimetres, although some variations can exist. Then, that pin stub is divided by a grid where, image by image, it will be scanned by the SEM. These images have a fairly dark background from the carbon and some bright particles, depending on how many particles there are.

These images are then thresholder by the backscatter contrast, and when we find something that is above the backscatter threshold, we measure that with electron beam. We measure the morphology, and we measure the size. Once that falls within the area of interest, we take an EDS spectrum and that EDS spectrum is automatically analyzed for the chemical composition of that particle. As mentioned before, if that then turns out to contain lead barium and antimony, we call it a characteristic particle for GSR.

That is in a nutshell how the process of gunshot residue analysis works. Obviously, there are many, many more details and settings that you can do, which are done in the software user interface.

What must be considered to get good results?

We try to make the results as user-independent as possible, but it should also be microscope independent. To make sure that the microscope gives you the proper results, and that you can get a good, reliable result as an operator, there are several areas that the microscope has to perform well on. One of those is the lifetime of the source, and the stability of the electron source. This is because you have the backscatter contrast for different shades of particles; if your backscatter varies, then your contrast varies and therefore you cannot just set a simple threshold anymore.

The signal to noise ratio is quite important so that you have a crisp and sharp image and so that you get that image within a reasonable time. A lot of the parameters of the microscope have to be properly calibrated, such as field of view, the scanning area, the stage, and so on, so that when you scan an image and move the stage, you don't miss any areas. You certainly don't want to scan too many areas more than once. So, all of these calibrations and settings have to be very well aligned, including the backscatter, the EDS detector. Also, the classification files on what it means when you get a certain chemical composition, and how you classify the actual particles needs to be accurate.

In fact, the stage position is one of the most important things, because when you move the stage one field of view, you will want to be sure that it is exactly one field of view and not a little bit more, because if you have a gap, there could be particles there that you are missing.

Traditionally, you can say that there are three types of electron sources. The tungsten sources, the crystalline sources (LaB₆ and CeB₆), and FEG sources. They all have different characteristics when it comes to the brightness, spot size, and bundle diameter. For instance, how long does the source last? And finally, what is the price of those sources?

What are the best sources for GSR analysis?

Tungsten is considered the step-in model of the sources. It doesn’t have a very high brightness, it has a fairly large spot size, so the resolution is somewhat limited although good enough for GSR, but not the highest end. It has a relatively short lifetime, depending on the settings that you use, but the big advantage is that it is quite cheap.

The CeB6 source, the crystalline source has about 10 times higher brightness than the tungsten with a much smaller beam diameter, so you get a higher resolution. Its lifetime is much longer and typically, the end of the lifetime is much smoother so you can plan exchange. The price is higher than the tungsten, but it is not quite as expensive as an FEG. Really, the FEG is considered top of the range. It is quite expensive, but it has a long lifetime, is very nice and bright, and gives the best resolution. The long lifetime of the FEG source is the main reason customers purchase it as the gain in resolution is overkill for GSR.

We carry the CeB6 source. We think that is the perfect compromise in the sense that you do get the long lifetime and the stability of the sources but still at a reasonable price. So, in that sense, it's the best combination for GSR analysis.

One of our customers imaged a particle after more than 3,000 operating hours on that CeB6 source. I have to note that this is just one example, and we cannot say that all our sources last that long, but it is a nice example of how long it can last and how many hours you can analyze with this source and still get good results.

Another example, although not specifically about one part of the microscope, but of the total package, is the report that you can get from Plano if you analyze Plano samples. The Plano sample is an artificial GSR sample that contains particles in a whole range of sizes. If you analyze these particles and make a report out of that you can see if you have missed any particles, counted any particles twice, have measured the size accurately, got the chemistry right, and so on. We can say that we can consistently find all these particles with the Perception GSR, and still at a high speed as well.

All particles are positioned within the accuracy of a few microns, which is very accurate. That means that when we find a particle, we know exactly where it is. That is very useful for repositioning the particles when you revisit the particle for confirmation, the fact that when you drive to that particle, it will be right there, right where you look for it, so you don't have to search for the particle, it will be just right in the middle of the image.

This is a combination where all the bits and pieces of the microscope really have to work together and have to be well aligned together to get to these results. Just having a good backscatter or just having a good source is not enough. It's all the parts of the microscope that have to be properly aligned.

How does the Perception workflow operate?

With Perception, what you would want to do first, is define your areas of interest in your Nav Cam overview. You can have up to 30 GSR samples in one sample holder. So, for each sample, you define the area that you want to analyze.

Then you apply a recipe on these areas. So, in the recipe, which you can make yourself in the software, you can drag all of these functions into the recipe builder so there are multiple parts of the recipe. What it does at start-up, which it runs every time after each sample, and data runs, you put the samples you want to analyze in and you can decide what you want it to do during shutdown.

You only have to make this recipe once. You can define all of the settings to fit your laboratory standards, and then you can use that recipe each time for the same sample area, so you don't have to design a new recipe every time you want to analyze a GSR sample. After that, if you have your recipe and your sample areas, you basically just press start.

When you press start it will automatically do all of the calibrations for you and then scan the sample areas. You will see all of the squares drawn on your sample, and it will go through them one by one to check all the particles that you want to find and see if there are any GSR particles on it. Then it will give you the results which just give you all of the particles that it found and in which category it fell.

For GSR, it is also important to do revisiting. This also built in the Perception software. With the data set that is generated from your runs, you can easily revisit those particles for confirmation. So, in this case, we see a GSR particle here, and we draw an EDX area that we want to analyze. This will generate a new spectrum to make 100% sure that this particle that we’re seeing is a GSR particle. In the report, it will show you the picture of the GSR particle with the spectrum, which is needed to show that a particle is a GSR particle.

Take us through the Perception software.

This is the screen you see when you open the software. We have our Aluminium Copper interface, and the Nav Cam overview.

You can adjust all your image settings; so, you can adjust the brightness, your contrast, you can zoom in, zoom out. You also have the EDS panel. The nice thing about Perception is if you click, it will start doing EDS automatically, so you don't have to take an image, you just click, and it will automatically do the point analysis.

You can also set different analysis parameters for each stub you have in your sample holder. You can change what size you want to detect, and to what magnification you want it detected, and how long you want to run EDS on your samples, what video levels you want to detect, as well, so you can also remove all the particles that have a very dark video level. Usually that’s not what we're interested in because GSR particles are very heavy elements, so we want to detect the bright particles.

Then you set up your recipe, you set up what stubs you want, click save, go to your automated feature analysis. You can also put extra comments in here. So, it's easy to say what the suspect is, or what client, or what project number, and then you just press play to start running all your stubs according to your recipe.

When the whole progress is done, we just get a data set of the particles the Perception software detected in a stage map. Here is where all the particles that were found during the run. F

What is easy in this software is that you can just double click, and then you can revisit the particle, and it will be in your field of view to confirm that this particle is actually a GSR particle. Next, users can run a manual EDS and be able to whether they have a GSR sample. After they have added one GSR particle, they can confirm it, then they can make a report. From this whole data set, we can generate a Word file with all the data from the run. You can choose which data you want in there you can include your quality control as well as a summary of all your particles. You can also choose individual particles that you want in the report. You can set it up however you like. This marks the last step.

However, I would also like to talk about Perception’s x-ray mapping feature. In the x-ray mapping view, we can detect which elements we would like to see. You can see that aluminium is found in the blue colour, for instance, and copper is found in the green colour. This way you can quickly analyze your field of view to see what elements you have in there.

The summary of this walkthrough is that we made a dedicated solution for GSR in the sense that we really optimized all the parameters with all the knowledge about the hardware and software about the EDS detector integrated inside the electron microscope, along with all the knowledge of the software and the automatic running of the software and all the calibrations. Everything combines within our group, within our company, in house basically, to really give you full solution for GSR, and optimizing all the different parameters that come with that. That's on a small footprint, too.

Everything you need pretty much fits on one meter of desk space, and it doesn't require anything special, it doesn't need anti-vibration or electromagnetic shielding. It's just 220 volts and you're good to run it. Being so small, it's still super reliable in not only giving you the accurate and proper results on the samples but also in being operational.

Phenom Perception GSR Solution

On-demand webinar: GSR analysis with desktop SEM

What is the time it takes to analyze all 30 GSR samples?

That's an interesting question, because, first it's not easy to answer, and second you can load 30 GSR samples into the stage and into the microscope, but it is highly exceptional. In most labs, they continue with a case by case, and any case would typically have four to six or maybe eight samples. But that's pretty much the maximum. There are very few laboratories that mix cases and load multiple cases in one go and really fill the stage with 30 samples. That being said, it can be done. The software can handle it, the microscope can handle it. It will be running for a while.

To give some concrete numbers, let's say that each sample has 1,000 to about 3,000 particles, they're medium loaded samples you would say, they take an hour to maybe one and a half hour, so then the microscope would be running for around 30 to 40 hours, which is a nice weekend job. If you start that on Friday, it will be done on Monday. You probably won't make it overnight except if the samples contain very few particles. But, for a weekend, it can do that. As I said, yes, it can be done, but it's not very common.

Is the user able to change their source in-house?

Sources can be changed in-house; we do strongly advise to have that done by a service engineer, because of the experience and of the sensitivity of the high vacuum at the source area and the risk of contaminations if inexperienced users exchange the source. That being said, there are exceptions to that. But it really depends on the environment where the system is placed and the level of knowledge of the user.

Is it a high maintenance system?

The maintenance intervals are quite long and are quite predictable. That is partly to do with the source but also the rest of the electron microscope and especially the hardware, which are so extensively tested that we are really sure that it can take the heavy caseloads of GSR work, where you typically run the microscope for whole nights on end. Where normal electron microscopes are usually used for an hour or two, GSR microscopes are typically running pretty much 24/7.

Despite all of that, it is still very fast. The stage movement takes a few hundred milliseconds. It's fast, and that gives you very responsive and fast feedback on the system. It’s not only the stage that is very fast, but also with the bright electron source we can make images at a high speed. We have very short pixel dwell time, and besides this, it gives you a fast image, and less charging on your sample because the beam will only pass the sample very shortly.

On top of that, you also get a very high x-ray count rate. We don't have to put the beam on top of any particle for a very long time to get an x-ray spectrum; we can do that typically within a second or two.

Can all varieties of ammunition be detected?

We have mainly referred to lead, barium, and antimony ammunition, but indeed you can also detect other ammunition. There are classes already in the class file that cover some of those, but more importantly, the class file can be adjusted by the user. So, if in future some new ammunition is invented that contains different chemistry, which is today unknown to the software, the user can still update those files and make sure those are also considered for GSR. So, in that sense, the software is future-proof in that you can keep those classifications up to date.

Where can our readers go to find out more?

To find out more please visit thermofisher.com/phenom-gsr

About Jeroen Smulders

Jeroen Smulders is a product manager for the Thermo Scientific Phenom Desktop SEM product range at Thermo Fisher Scientific. He plays a leading role in the development and implementation of new software applications for Thermo Fisher Scientific products and shares his best practices with customers to increase their output and get the best results.