Obtaining 3D Chemical Maps by Energy Filtered Transmission Electron Microscopy Tomography

The overall goal of this protocol is to obtain, visualize and quantify the 3D chemical maps obtained by energy filtered electron tomography. This method can help answer key questions in material science, chemistry and catalysis fields, such as how the components of a given material are mixed at nanometric scale. The main advantage of this technique is that it provides information concerning the chemical distribution in the bulk and at the surface of the sample.

Energy filtered imaging can separate chemical elements that are very difficult to distinguish using other imaging techniques. The experimental protocol described here shows the means of obtaining 3D chemical maps but also it tries to provide ideas to quantify some of the 3D chemical parameters. The 3D chemical maps can be only generated for samples that are not beam sensitive since the recording of filtered images requires long exposure times to relatively intense electron beam.

To begin prepare a sample such as silica alumina or titania alumina, as described in the accompanying text protocol. Next use an electron microscope to record energy filtered tilt series. These images will be used to calculate each chemical map of the search chemical element.

Open the three filered tilt series consisting of two pre-edge series and one post-edge series in image J for analysis. Now open the specialized EFTEMTomoJ plugin and click on then click on Image or Stack to select the tilt series. In the table that appears enter the energy at which the tilt series was recorded as the energy shift.

Also enter the slit width and the exposure time of each filtered image. Check all three tilt series under mapping and ensure that they will be used for computing the chemical maps using the power law. Then click on Next.

Choose the first image as the reference image and then click Apply followed by Next. At this point a preview image of the proposed alignment appears. Visually verify that the three images recorded at the same tilt angle are well superimposed, showing no shift between the images.

This protocol was performed on the version 0.9 of the EFTEMTomoJ plugin. At this moment the filtered images recorded at the same tilt angle are aligned. Next select the EFTEMTomoJ window, the images corresponding to the background and to the chemical edge.

Then select the model of the signal extraction as power and click on Create Map. Save the tilt series of the chemical map and then repeat this process for all of the other chemical tilt series. In order to quantify this drift, first open the zero loss aligned tilt series.

Then open the chemical maps aligned tilt series, go to Image, Color and select Merge Channels. Select the file corresponding to zero loss for the red color, the first chemical element for green. Uncheck Create Composite and check Keep Source Image.

At this point a stack is created at each tilt angle for all the recorded images. Next go to Plugins and select Align RGB Plans. Here red is the reference image.

Select green and using the arrows overlap it over the red one. Once in position click on Next and repeat this process for all of the other angles. Go to Image, down to Color and select Split Channels to split the RGB stack into tow stacks.

Red corresponds to zero loss and green corresponds to the chemical maps with the drift corrected. Finally be sure to save the tilt series, then repeat the actions for the second chemical element and remember to save the file. For 3D chemical mapping again sit at a computer and open image J.Open the TomoJ plugin by going to plugins and selecting TomoJ.

This will select the load angle form file. Because all the tilt series are already aligned navigate directly to reconstruction and calculate the volumes of zero loss as well as the chemical volumes using a reconstruction algorithm. It is recommended to use an interactive algorithm to reconstruct the chemical volumes.

Also using this software it is possible to reconstruct the volumes using the GPU of your computer. Once all volumes are computed use the Merge Channels options to apply different colors to the obtained volumes and overlap them in a single volume to obtain the 3D chemical map. Finally follow along in the accompanying text protocol for additional details on 3D modeling and quantification of the chemical information, surface area and pore size distribution.

EFTEM Tomography was used here to analyze three different samples of titania alumina catalyst supports with different component ratios. It was found that titania at high concentrations forms clusters that are embedded in the alumina. These models show the chemical distribution of titania and alumina on the surface of the samples.

It was found that independent of the proportion of titania and alumina in the sample, the surface of the sample is covered with titania in a proportion of about 30%Alumina has a high specificity to the surface and the goal is to have titania supported by alumina. After modeling it was found that the sample containing 10%titania has a layer that is only about 10 nanometers thick on the surface. This variance could play a role in the sample's catalytic applications.

It is important to mention that energy filtered electron tomography is a time consuming technique. While attempting this procedure remember that energy filtered imaging is sensitive to the defraction contrast, but also the thickness of the sample and its resistance to the electron beam must be verified. After its development this technique paved the way for researchers in materials science to better correlate the morphological and chemical parameters with the properties of that studied material.

We hope that after following this protocol the new users of electron tomography will have a better understanding of the step to follow and on the difficulties encountered in reconstruct a volume, segment a model and quantify its parameters.