It's too small to be useful in tradition lens applications. Even at lower X-ray/Gamma energies when the refractive index is much bigger ( see: http://henke.lbl.gov/optical_constants/getdb2.html) you can't really use make reasonable lens.

It would be far far easier to increase the "commonly" available divergent sources' intensity than attempt to recover losses due to divergence with such a weakly focusing system. Hell, you could likely achieve a much bigger intensity increase by moving the source closer.

Can anyone even think of the medical imaging system they are envisaging. I assume they mean a phase-contrast approach but that needs a intense 700kV source with a small spot size and thin structured grating that absorb 700kV photons (which I don't think exist thou it's maybe just about within the capacity of man to make them).

The only application I can think of for this is for X-ray telescopes and astrophysics where you can't increase the source but you can have massively long imaging systems. However I will bet that their are not many interesting astrophysical events that emit only 700kV+ photons and the attenuation of the massive silicon lens may well counteract the benefit of the focusing.

The immediate never mind long term carcinogenic effect of very low dose radiation is not settled either way even in the simple case of cells. You can trudge through the related links and responses to them at :
http://en.wikipedia.org/wiki/Radiation_hormesis
if you want to bore yourself.

The problem is it's very difficult to detect the difference between a small negative effect and a similarly sized positive one. Plus due to complicating effects like the bystander one it is likely that the pattern, flux and nature of the applied radiation will complicate the results.

At such low radiation levels in more complex organisms to know for certain you would need to have an unethical massive experiment using millions of animals with half being irradiated in a precisely controlled way. Rightly so this doesn't happen and so researchers have to look at smaller samples often with massive variations between them. Which is why the issue is not settled.

O and all radiation isn't bad, look at radiotherapy :-)

Couldn't agree more on the terminology. Eg phase-shift,interference,diffraction,scattering all used to describe similar things in different fields.
Yes I did wonder about double slit experiments (started in Atomic physics) but more so about the pattern's mere existence for single photons. The question of its shape seems trivial by comparison. :-) But thats for another board.

Okay sorry my mistake I thought you meant a direct phase measurement.

IMHO they are exactly doing a classic optic phase-constrast technique. Not exactly the same geometry as in microscopes due to the vastly superior optical beam available. But very similar and identical to the one often used for accelators X-ray CT which is why they reference this research. If you think this is wrong thats fine, I just disagree.

Ah okay, I understand what you mean now. However I still think your wrong as whilst you can measure phase of photons in the RF/Microwave domain there is no method of measuring phase directly with optical photons never mind 100keV electrons.

Indirectly by observing the imaging pattern (like a classic optical fringe pattern) the phase-shift can be done. Thou you might have to manipulate the beams to make the pattern more obvious as in classic interferometers or classic phase-contrast microscopy. Or you can look from "far" enough away so that the scattering from the phase-shift is enough to produce an image from which you can infer the phase-shift itself. This is the technique the Sheffield electron folk, some telescope folk and the X-ray folk can use. There is actually a smooth continuum of techniques depending on how far away your detector is, now transparent you object is and now you block or deal with the transmission image that "contaminates" the pure scatter/phase-shift image. But lets ignore the other fields for now as this is a semantic debate.

In this particular case they are not measuring the phase of the electron wave. They use a Gatan, Inc. ORIUS SC200 CCD Camera. Such a detector doesn't even detect the electrons directly. Instead it detects the 10 or so optical photons emitted from the scintillator placed in front of a CCD when each electron strikes it (the optical photons are of course emitted with random phase). So all they can possibly detect is an electron intensity profile. As this is a far field diffraction pattern you can mathematically reconstruct the sample that would produce it.

They themselves point on page 2 left column that this technique has been about for 40 years in the optical world and admit X-ray researchers have looked into it before.

I have done optical super resolution microscopy myself but have never heard about the MRI equivalent. Then again I don't really know much about MRI research so that's not so surprising and I obviously defer to your knowledge on this. Similarly I have seen synthetic aperture talks but only those done with Ultra sound or telescope based data.

Interesting now all the very different imaging modalities have vastly different implementation rates for similar techniques. Whilst the method and technology of acquisition is responsible for some of this I suspect a lot of it is researchers simply not knowing what is going on in other fields. (As I myself demonstrate with MRI)

Okay take your point on the MRI. I'd describe that as phase encoding spatial information rather than interference but I accept your way is just as valid. It's certainly not the most common method in X-ray phase-constrast CT. However this method is used for X-ray CT at synchrotrons.

That's not true. When a physicist (like me) talks about an interferometer generally we mean a device that can measure phase shift or relative phase by comparing a beam with unknown phase to a separate known phase-shift reference beam.
See: http://en.wikipedia.org/wiki/Interferometry

Such interferometers (which normally use a point detector and therefore aren't measuring a pattern) have been about for well over a hundred years
See: A. Michelson, E. Morley. American Journal of Science: 333–345. (1887). So I simply don't know what you mean by "traditionally done".

Inferring phase-shifts from diffraction patterns is different (but obviously mathematically related) and more limited. For example a light beam going normally through a thin sheet of glass has no diffraction pattern. However it has a measurable phase-shift with an optical interferometer.

When you say we can only recently measure phase at high frequency for light I think I am missing something. If you are clever with the optics you can make all phase-shifts manifest as intensity variations. This is exactly the principle of a phase-contrast optical microscope widely available since the 60s.

Saying you don't need a lens is just wrong. At the very very least you need a lens to focus the beam onto the specimen.

O and if you actually read the paper you'll see that on the right hand column of page 5 all the current experimental limitations on resolution are listed. That the lens doesn't feature doesn't mean it follows that it is just down to wavelength, it's not. There is no lens in any hospital X-ray systems and your 100 keV diagnostic X-rays does not give you ~0.01nm resolution images. There is a world of difference between physics based resolution limits and that which can actually be obtained in most cases.

The method is nice, worthy of a nature paper and will improve on the future. However those that are saying that this is it "they got this" for imaging are just plain wrong. It's a new imaging modality we get one every couple of years or a few a year depending on your definition.

nMR measures the spectrum of RF emitted when nuclei excited into alignment by resonant RF pulses relax out of alignment with an externally applied magnetic field. MRI which is the correct name for the imaging modality measures the pattern of this RF emission. Maybe I am missing something but now exactly is this an interferometric process?

As to how this IS used in a research CT scanners google "phase-constrast CT". O and you want to do this in future to reduce patient dose or highlight soft-tissue boundaries which aren't obvious on a standard CT.

It's not a type of interferometry as they don't measure phase-shifts directly. They measure the diffraction pattern and infer the phase-shifts that would result in such an image. Basically measuring the Fraunhofer diffraction pattern and computationally reconstructing a sample that would create such a pattern.

As such if the sample was a plane normal to the electron beam they would see nothing in this method. The imaging is not done in the Fourier domain thou the reconstruction can be if you prefer. For the imaging it is merely a standard electron imaging array which is why on their webpage they can show "The [diffraction] image before it is reconstructed using a computer"

You don't actually need a lens for phase constrast measurement. That is why it is possible to do exactly this method for hard X-rays for which reasonable lens don't exist. See the following for example: http://www.nature.com/nature/journal/v467/n7314/full/nature09419.html

Electron imaging is only possible in vacuum and delivers massive doses typically to microscopic samples. So I doubt it will replace medical CAT or MRI scans any time soon.

Basically what they have done is phase contrast transmission electron imaging. This is quite an achievement in itself and well done to them. However they most certainly did not invent this "technique" (and I doubt they actually claimed that). The method is well known from X-ray phase contrast imaging research.

They even wrote this: "The technique is applicable to microscopes using any type of wave and has other key advantages over conventional methods. For example, when used with visible light, the new technology forms a type of image that means scientists can see living cells very clearly without the need to stain them, a process which usually kills the cells."
Em, yes but optical phase-contrast is damn well established. O and Frits Zernike who got the Nobel prize for doing exactly this in 1953 might be pissed off.

Wrong! You can infer the phase by measuring the deflection angle due to the phase-shift. Naturally this doesn't work for normal angles of incident but the images are still dramatically improved.

Unmistakable, really? It might interest you to know a crucial part of the UK process is using obvious decoys.