Atomic structure of a glass imaged at last | Hacker News Reader

Atomic structure of a glass imaged at last

nature.com

109 pointspseudolus19 comments

Warning for anyone browsing: The article is very interesting, but there's no "image", just diagrams. It's not clear if there even is an "image" as such, or if it was some kind of more general experimental confirmation. Maybe there's an image is in the source paper, but I don't have institutional access and the paper by itself is 200 USD.

Imaging doesn't necessarily produce 'images' in the common meaning of the word.

You may think that those 'images' of proteins you see sometimes are 'real images', because they are grey and grainy, but they are just computer-generated diagrams.

I'm not really sure if this comment has any substance. What do you want/expect to see?

"have used a method called atomic-resolution electron tomography to determine the position of every atom in a nanometre-scale sample of a metallic glass" assuming the accompanying diagram is exactly that, it's as much of an image as you will ever get, of anything, afaic.

Edit: formatting

And yet, the number of otherwise informed, intelligent people I've talked to who believe that the frequently-used image of the Covid virus is an actual picture is astonishing.

https://www.niaid.nih.gov/news-events/novel-coronavirus-sars... says images from a scanning electron microscope. What is the difference between that and an "actual picture"?

Thank you, I was trying for a while to find a way to phrase it. All images are reconstructed... if it was helpful I'm sure they could have added back the noise they removed.

I think this is the preprint article: https://arxiv.org/abs/2004.02266

looks like it. page 14 has a figure with a "representative experimental image, where some crystalline features are visible." the nature paper adds that the scale bar is 2nm.

Suppl. Fig 3. on page 31 shows why the raw images (each taken with a slightly different orientation of the nanoparticle) are not emphasized by the paper: they all look pretty much the same to the human eye. To get useful info, you need to do some mathematical processing of the subtle differences between them.

The last 9 pages are even more interesting visually

'Imaged' is probably the wrong word to use based on its connotations outside the field. What this paper represents is a full 3D reconstruction of the atomic configuration of disordered solid, and so the result is really a point cloud not pixels or voxels.

This sort of atomic scale reconstruction is really hard to achieve outside of simulation, and some very cool things can be done with it. The 'Short-Range-Order' of disordered or semi-ordered structures is really difficult to probe, but may play a signficant role in properties.

The journal subscription is $200 but the paper is $9 to rent or $32 to buy without it.

It does look like there's images in the source paper but I'm only going off the fuzzy preview: https://www.nature.com/articles/s41586-021-03354-0.epdf?no_p...

_rent_ interrobang

I guess that's a short-term subscription or something?

Usually something like 48 hour read-only, and restricted to some BS viewer that doesn't allow printing.

Fairly trivial to bypass usually, worst case is just printscreening it.

Edit: Ah -

    View the article PDF and any associated supplements and figures for a period of 48 hours.
    Article can be synced in the ReadCube Cloud.
    Article can be viewed offline in ReadCube's desktop and mobile apps.
    Article can not be printed.
    Article can not be viewed or saved outside of ReadCube.

You might consider the image shown to actually be more impressive than a regular 2D image since it is a tomography done in 3D. There is far more information in this simulated image than in say, an SEM image of graphene. Also, it’s hard to say what an “image” actually is at nanoscale.

Holy cow, the goddamn cookie acceptance nag fills all but one line of my browser window.

Here, have a free copy of my "kill fixed" bookmarklet. Sorry I don't have it in more readable code, but it's small enough to parse through so you can verify nothing nefarious resides in there.

    javascript:(function()%7B(function%20()%20%7Bvar%20i%2C%20elements%20%3D%20document.querySelectorAll('body%20*')%3Bfor%20(i%20%3D%200%3B%20i%20<%20elements.length%3B%20i%2B%2B)%20%7Bif%20(getComputedStyle(elements%5Bi%5D).position%20%3D%3D%3D%20'fixed')%20%7Belements%5Bi%5D.parentNode.removeChild(elements%5Bi%5D)%3B%7D%7D%7D)()%7D)()

As with most websites with obnoxious cookie banners, the banner doesn't show if you don't have JS enabled on the website. The article even reads fine without it, including the pictures, something that can't be said of the new hipster blog-hosting sites that lazy-load images with JS instead of letting the UA do it.

Why do almost all scientific articles - especially from nature and new scientist - not have images or videos to show the damn thing everyone is so excited about?

Can't afford Elsevier.

Reply on news.ycombinator.com