In this webinar, we look at the capabilities of TEM and its methodologies. The Q & A from the webinar provides a brief FAQ about this type of analysis.

Q: How large a field of view can you prepare for TEM?

A: While we cannot image the entire prepared sample at once, we typically prepare TEM sections by FIB with an electron transparent region that is about 10 µm in length. We actually prefer an area that is smaller than that just to maintain mechanical stability, but that’s the typical limit of FIB preparation. In a TEM sample that is prepared by ultramicrotomy, an ultrathin sample can be up to 2 or 3 mm in diameter.

Q: Can any material be prepared for TEM?

A: Most likely not. There are some restrictions around what we can put into the transmission electron microscope. Again, the most important restrictions are that it needs to be vacuum compatible and that it needs to be able to stand up to the FIB preparation if we go the FIB route. However, some polymer samples that may be vacuum compatible might be ‘sectionable’ with our ultramicrotome but may degrade under an electron beam. Some could even be too soft to section by ultramicrotomy. There are many, many materials that can be used for TEM, however.

Q: Can TEM be used to measure crystallinity index as XRD?

A: In a single crystal sample, we get well-resolved diffraction patterns that we can use to measure lattice parameters and similar aspects of crystallinity.

Q: Is EDS analysis in the TEM or STEM more time consuming than in the SEM?

A: The answer to that question is no. It typically takes about the same amount of time once you have your sample in the instrument to actually generate the EDS analysis.

Q: Can you do grain boundary imaging and analysis in the TEM?

A: Typically, the answer to that question is yes. Again, it can be somewhat sample dependent, but the huge advantage in the TEM or STEM is that the sample is so thin that there is virtually no spot blooming compared to bulk EDS analysis in the SEM, allowing us to perform EDS analysis on a nanometer rather than micron scale. So if you have a problem along those lines, again, we would certainly encourage you to get in touch with us and we can see about designing an analysis to address that.

Q: Can you do grain or particle size analysis from TEM images?

A: Again, the answer to that, in general, is yes. However, you will need to contact us to confirm.

Q: Can you make a TEM sample from a four-inch wafer?

A: Yes. On a four-inch wafer, if there are small structures or small defects that we need to find, the capability exists to drive to those positions and then excise those regions of interest from the wafer as lift-out samples and prepare them for subsequent TEM analysis.

Q: What is the difference between high resolution TEM and STEM?

A: STEM in its name implies that you are scanning the beam: Scanning Transmission Electron Microscopy. In that technique, we would form a finely focused probe beam, step it across the sample in a raster fashion to generate the image (as is done in an SEM) but we take advantage of the higher beam voltages and different electron optics to provide additional signals and information about samples, as compared to SEM. In a high resolution TEM image or TEM in general, the beam is not scanned; we rely on the electron optics to form a global image of the sample, much as you would in a light microscope.

The above webinar discusses analyzing surface chemistry on micro-devices and its methodologies. Here you can examine a Q & A from the webinar of FAQs about this type of analysis.

Q: Does surface roughness have an effect on depth profiles?

A: Surface roughness does indeed have an effect on depth profiling. But with our new instrumentation, our ability to rotate a sample at multiple points coupled with our better charge neutralization helps to mitigate the effects that you would see from surface roughness. It may cause your sample to sputter etch at a bit slower rate than expected, but rotating will help decrease the effects that we see from that. Charge neutralization also helps to mitigate the effects one sees. So yes it’s possible, but there are caveats.

Q: How deep can you profile? How long does it take to sputter?

A: That depends on the sputter rate of the material under analysis. Again, we do a quality check four times a year on our instrumentation. And we know that silicon dioxide sputters at a rate of 10 nanometers per minute in our instrumentation. We have parameters that can increase or decrease that sputter rate and make it much faster or much slower, but if that’s a benchmark, it’s limited. Thus, with the 10-nanometer sputter rate, it just limits how deep we can get into the sample.

And some materials may sputter faster or slower, so it all depends on how much time you want to spend to get, say, a micron into the surface.

Something to keep in mind as you’re sputtering past a micron is that interfaces may be intermingled as you’re getting deeper into the surface because of the crater that’s being created. It depends on the data that can be collected, as well as the electronics, and the set-up of the instrumentation. Those are all factors to keep in mind as you’re sputtering one to two to three microns into a sample. Time is a big factor– how long do you want to spend to sputter a micron or more into your sample.

It’s possible, and we can do it. But once you get into the two to three micron range, there are several things to keep in mind. You may see artifacts as we sputter that deep into the sample. As long as you remember that there may be artifacts and that that data may have to be taken with a grain of salt, you can sputter as deep as you want, as long as you have the time to do it.

Q: Can you give me an example of when weight percentage analysis would be needed?

A: Weight percentage analysis is often needed when a client comes in with an idea of a product that’s been on the surface. It’s not always possible, but sometimes they say, hey, we know the weight percent of this product. Can we see some of that as we analyze our sample? So let’s say we have a particular product of poly dimethyl silicone, and can we see that reflected on our sample. Sometimes that can be calculated to weight percent that can be found. Weight percentage can also be useful or beneficial if you’ve tried to create a product similar to a competitor’s and you want to see if that works out. Sometimes that can be back calculated. Sometimes that comes with a caveat.

Q: How can you produce maps and line scans similar to what you showed in your depth profiling system?

A: A map is actually made up of multiple points so we analyze each point and software will combine the data to give it to you. So you are losing some sensitivity to data because it is a multidimensional analysis. We recommend doing a point analysis to get your overall composition. We take multiple points, and just like a pixelated picture the software combines all those pixels.

The above webinar discusses the capabilities of fast mapping and its applications. Here you can read the Q & A from the webinar for a series of FAQs on the topic of fast mapping.

Q: Do you have to define elemental windows before starting the map?

A: The answer is no. Because you’re getting an entire spectrum from each pixel in the map, you don’t have to define anything ahead of time or after the fact – that’s locked in. You can always go back and change the elements you’ve mapped for, so that way you can dynamically change it as needed. You can even add elements partway through. You can subtract them if you find that they don’t make sense because they shouldn’t be there. But again, if you don’t agree with the spectrometer, you can always check your work by extracting synthesized spectra containing the same elements, and verifying that any decisions that you came to are actually accurate and correct.

Q: Is this method able to be used for polymer or non-metallic materials?

A: The answer is yes. You can certainly use this to look at polymeric materials. As with all SEM work on polymers, you typically have to coat the sample with a conductive coating. Otherwise, you will end up with charging problems because the sample cannot dissipate the charge from the electron beam.

Q: How deep into the sample does the beam penetrate?

A: The beam penetrates anywhere from one up to tens of microns, depending on the density of the material and the energy of your beam. The scale in the webinar example is 10 microns in the lateral dimension, but the same holds true in the vertical direction. Thus, when you’re hitting a low atomic number material like aluminum with a 30 kilovolt beam, most of the information will come from the outermost few microns but you can still get some data from much deeper into the sample surface. So if you’re looking for very sensitive detection, you can still pick up those very weak traces of elements when they’re coming in from a distance. As you can see with the iron comparison, the penetration depth is much smaller because of the higher atomic number of that material. And so you also get a smaller analytical volume with most of the information coming from the top 2 microns. But you can still get some data from as deep as 5 microns, but it’s not the highest intensity data that’s coming from the fringes of those “spot blooms” where the e-beam spreads out into the material.

The nice spring weather has come and gone. With that nice spring weather also came pollen…everywhere! Although it is often seen as a nuisance for triggering allergies, pollen is very interesting to look at magnified. You may be surprised at how different various pollen types look.

The familiar yellow haze on our cars and everywhere else can be annoying. But the variation in the appearance of different types of pollen at higher magnifications can be surprising.

Pollen close up: