Cryogenic electron tomography (cryo-ET) provides the unprecedented ability to study biological matter in its native state at sub-nanometer resolution. Cryo-electron tomograms are 3D reconstructions from 2D images collected using a cryogenic transmission electron microscope (cryo-TEM) over a range of orientations. A notable advantage of the technique is that the sample preparation workflow produces specimens unperturbed by dehydration or staining since samples are vitrified in their native environment.
A key requirement for cryo-ET samples is that they must be very thin, around 100-300 nm, to be electron transparent. Larger objects such as cells must be thinned to make them electron transparent. This is typically performed in a cryogenic scanning electron microscope equipped with a focused ion beam (cryo-FIB/SEM) to fabricate lamellae, or very thin slices of the sample.
How does ice contamination impact cryo-ET?
According to a survey conducted by Delmic, the lamella milling step in the cryo-FIB/SEM is one of the most challenging aspects of the cryo-ET workflow because it can result in amorphous ice growth on the sample surface . This amorphous layer of ice can grow as fast as 50 nm/hour. Typical milling sessions take place over several hours in which multiple lamellae are fabricated. Therefore, within 4 hours, the amorphous ice layer on the first lamella can easily be thicker than the lamella itself. The amorphous material coating the lamellae will result in reduced contrast and resolution of the TEM images and degraded tomographic data.
How are lamellae made for cryo-ET?
For small entities like single particles, viruses, isolated organelles, or small bacterial cells, thinning is not required to make them electron transparent. Whole cells and larger objects need to be vitrified and then thinned to around 300 nm or less to make them suitable for TEM.
The preferred method of thinning samples to fabricate lamellae is using a cryo-FIB/SEM. The SEM in cryo-FIB/SEMs is utilized for imaging the sample throughout the milling process to monitor the thickness and orientation of the cut.
Focused ion beams of heavy ions like Ga+ are used to precisely cut the lamellae at specific locations. A cryogenic fluorescence light microscope (cryo-FLM) can be used to identify locations of interest on a fluorescently labeled sample. A cryo-FLM system like the METEOR, which can be integrated into the cryo-FIB/SEM, simplifies this process significantly.
While the cryo-FIB/SEM operates under high vacuum, residual moisture in the chamber can easily condense on the cold sample leading to amorphous ice layer growth especially when sessions take place over longer periods. A common approach to mitigate ice contamination is to coarsely ablate the target regions to around 500-800 nm followed by fine polishing to produce the final lamellae . This reduces the sitting time after the final polishing step which reduces the accumulated layer of amorphous ice. Even with this strategy in place, the milling step remains a bottleneck in the overall cryo-ET workflow. This limitation undermines the yield of high-quality samples produced during automated milling sessions, which take place over several hours, often overnight.
How does the Ice Shield work to reduce ice contamination in the cryo-FIB/SEM?
Preventing amorphous ice formation during lamella milling is critical for improving the overall efficiency and throughput of cryo-ET workflows. Minimizing ice contamination would also lead to higher-quality cryo-ET data because of improved image contrast and tomographic resolution. Ultimately, addressing this critical bottleneck in the workflow would benefit not only the direct users of cryo-ET, but those who benefit from the 3D structural insights revealed by the technique.
The CERES Ice Shield is a commercially available and patented solution from Delmic that minimizes ice contamination in the cryo-FIB/SEM. How does it work? During milling, a liquid nitrogen-cooled shutter is automatically extended between the sample and the SEM column to act as a cryogenic pump . This works to reduce the partial pressure of water vapor in the sample vicinity. A small bore in the shutter allows the ion beam to pass through thus protecting the sample from parasitic ice growth during milling. The shutter automatically retracts during SEM imaging to allow access to the sample. As a result, the cryo-FIB/SEM modified with the CERES Ice Shield will have a significantly lower partial pressure of water vapor than standard cryo-FIB/SEM instruments, allowing automated milling sessions to take place without the concern of amorphous ice contamination.
A case study: ice growth rates before and after Ice Shield installation
In a case study performed by Delmic and its collaborators, the ice contamination rates were characterized before and after the installation of the CERES Ice Shield. It’s important to note that every cryo-FIB/SEM will have a different baseline level of ice contamination, therefore the reduction in the ice contamination rate is the most important factor when considering the Ice Shield’s performance.
To measure ice contamination rates, several SEM images of grid holes were acquired over a four-hour period once the stage was settled. The images were taken at a shallow angle, allowing the ice film to be visualized, measured, and quantified. The same procedure was repeated on three separate cryo-FIB/SEMs and in all cases, it was found that the Ice Shield reduced parasitic ice growth to non-detectable levels.
Additionally, the partial pressure of water vapor was compared with and without the Ice Shield which correlated to the reduced ice contamination rate. Without the Ice Shield, the partial pressure was roughly 3 x 10-7 mbar. After installation, it dropped by two orders of magnitude to 3 x 10-9 mbar.
In summary, the CERES Ice Shield is a simple yet immensely effective solution that is addressing one of the challenges facing cryo-ET users. With productivity and functionality in mind, the Ice Shield effectively eliminates parasitic ice growth during milling while not impacting any other aspect or function of the cryo-FIB/SEM.
 Lau, K., Jonker, C., Liu, J., & Smeets, M. (2022). The Undesirable Effects and Impacts of Ice Contamination Experienced in the Cryo-Electron Tomography Workflow and Available Solutions. Microscopy Today, 30(3), 30-35. doi:10.1017/S1551929522000621
 Turk, M. and Baumeister, W. (2020), The promise and the challenges of cryo-electron tomography. FEBS Lett, 594: 3243-3261. https://doi.org/10.1002/1873-3468.13948
 Sebastian Tacke, Philipp Erdmann, Zhexin Wang, Sven Klumpe, Michael Grange, Jürgen Plitzko, Stefan Raunser, A streamlined workflow for automated cryo focused ion beam milling, Journal of Structural Biology, Volume 213, Issue 3, 2021, 107743, ISSN 1047-8477, https://doi.org/10.1016/j.jsb.2021.107743.