How to prepare HexAuFoil^® sample supports

Plasma cleaning

HexAuFoil^® grids should be plasma-cleaned or glow-discharged to increase their wettability prior to use. In general, for best results, they will need glow discharging for significantly longer than other sample supports, such as UltrAuFoil^® holey gold supports. However, the gold foil is not volatile when glow-discharged or plasma-treated, so the grids may be subjected to more extensive plasma treatments than standard carbon foils, without any risk of degrading the  surface. A good rule of thumb is to approximately double the plasma-cleaning/glow-discharge times compared to UltrAuFoil^® grids. A table of suggested settings is provided below for some common glow discharge instruments (as originally published in Naydenova and Russo, 2022).

Suggested plasma cleaning settings

Vitrification settings

In general, HexAuFoil^® grids can be directly substituted into your vitrification protocol, with the exception that blot times should be extended compared to similar samples on a standard larger hole grid. For example, if you currently blot for 10 seconds, with HexAuFoil^® grids, you should initially try a 15 second blot. We recommend applying sample to the foil side of the grid and blotting from the same or both sides, but this is sample dependent, and any variation can be used.

Typical settings for vitrification with a HexAuFoil^® grid on some common plunge freezing instruments are given below by way of example (taken from Naydenova and Russo, 2022).

Suggested vitrification settings

Microscope alignment, calibration, and data collection

We strongly recommend very careful alignment of eucentric height and calibration of image shift immediately before data collection to ensure the best results. Due to the high consistency and dense design, data collection time can be halved even including the additional time taken in set up.

Software
EPU
EPU has been updated to handle HexAuFoil® hexagonal mesh grids with increasing functionality from 3.1 onwards.

SerialEM
SerialEM is also able to automatically identify HexAuFoil holes, if using version SEM_4-0-18 or later.

Data collection settings
Sample data collection settings are provided below, from a Titan Krios. Aberration-free image shift (AFIS), fringe-free illumination with no specimen tilt is preferred to take full advantage of the dense hexagonal array. Data collection is optimized with a single exposure per hole, one entire hole per image. Use magnifications corresponding to pixel sizes around 0.5–0.8 Å/pix on the detector to achieve this. The beam should be concentric with the hole with diameter ~500 nm, slightly larger than the 290 nm diameter holes, but small enough to avoid exposing adjacent holes. With a 600 nm centre to centre distance, and ±100 nm error in beam position, beam diameters up 700 nm are acceptable.

Larger beam sizes
Users can use a larger beam diameter (1μm), but does require some holes to be skipped, to avoid double exposure. This adaptation was used in Naydenova, Jia and Russo, Science, 2020, and some example settings are provided below (from a Glacios)

Alignment and Calibration
The regular hole lattice may be used for beam diameter and image shift calibrations. In addition, during processing, the edges of gold foil provide a built-in magnification calibration using the gold lattice reflections (2.347 Å). As with UltrAuFoil® holey Gold supports, a separate grid with an amorphous foil (normally carbon) will be required for other alignments including coma and astigmatism.

Additional care should be taken with image shift and eucentric height calibrations, as at these high magnifications, small errors can lead to large amounts of gold in the image. It may be useful to repeat the calibrations a couple of times to improve accuracy.

Data processing

In general, data processing protocols for HexAuFoil® grids are unchanged from data collected on traditional sample supports. However, the following changes to standard protocols should be considered:

1. The presence of the gold foil in every micrograph provides a built-in magnification calibration using the gold lattice reflections (2.347 Å). These edges are also high-contrast fiducials for motion correction.
2. The lace of electron beam induced movement of the vitreous ice means only whole-micrograph motion correction without subdivision into patches is necessary.
3. With no movement other than stage drift, bayesian polishing can still be used to confirm the absence of correlated particle movements within each hole but is not necessary.
4. Final high-resolution reconstructions can be used to extrapolate the structure factors of the molecule to those of the undamaged structure, at zero electron dose.