Reference:

Review of Scientific Instruments, 50(1), 26 (1979); "X-ray magnifier"; Boettinger, W.J.; Burdette, H.E.; Kuriyama, M.

A method for the magnification of x-ray radiographic images is described and demonstrated. This magnifier employs two successive asymmetric diffractions of an x-ray beam from highly perfect silicon crystals. The two diffractions magnify the beam in two perpendicular directions. A device with a magnification of 25x is demonstrated for Cu K(alpha) radiation. This device preserves and sometimes improves the resolution inherent in the radiographic technique. The x-ray magnifier is particularly useful in circumventing the relatively poor spatial resolution of electro-optical imaging systems needed for real-time observations. Basic limits on magnification and resolution using this method are described

http://mbbsdost.com/X-ray-magnifier-The-Review-scientific-instruments-Boettinger-WJ-Burdette-WJ-Kuriyama-WJ--1979-Jan/pubmed/5608650/googlescholar

https://www.ncbi.nlm.nih.gov/pubmed/18699332

APPENDIX:

 "This magnifier employs two successive asymmetric diffractions of an x-ray beam from highly perfect silicon crystals. "

Here is what I think is the key difference between what we have done and what others have done in the past kind of like the "X-ray Magnifier". Please correct me if needed.

In the case of the X-ray magnifier, the magnification is achieved by expanding the diffracted beam (the transmitted beam in the case of radiography) using ACT (asymmetric crystal topography), a well vetted technique from way back, twice to achieve beam expansion in 2 directions, correct? Clever!

In topography this would also create a double convolution on the in-coming diffraction topograph (for magnification), wouldn't it?

The real objective of viewing the oblique angle topograph of the "incident beam" from the Bruker D8 HRXRD was to observe and quantify the 2D intensity pattern emanating from the Ge (220) double bounce beam conditioner to simulate the image of the incident beam on the sample at various "oblique" angles of incidence. Then deconvolute from the expanded ACT image.

So, in the OAT technique, the fundamental difference being the use of the 2D detector surface (or dental film and/or nuclear track plate as in the past :-) and its relative angular position w.r.t. the incoming beam. For instance in the Berg-Barrett method the diffraction topograph is typically expanded using diffraction geometry for the asymmetric reflection and then the "film" (in our case the 2D detector plane) placed as parallel as possible to the sample surface to maintain one-to-one correspondence with sample VOXEL and 2D detector PIXEL.

I do not yet see any literature reference that uses the oblique angle topography method to expand a symmetric reflection topograph, as we intend to do soon once we have the ability to pivot the detector plane about an axis at the center of the detector plane. Meanwhile, our observations of the very narrow "line source" (1x20mm at 90 degree) emanating from the HRXRD beam conditioner is ample evidence of feasibility, no?

Here is an example of CONTRASTING detector plane at 90 degree and 5 degree:

https://www.flickr.com/photos/85210325@N04/30277126503/in/dateposted/

Have any of you ever looked at the main beam coming out of your HRXRD instrument? Bruker, Panalytical, whatever. Do you even think it is relevant to image instrument Io?

Especially, if one wishes to make interpretations from 2D XRD signals without going to a "point source" and raster mode thus bring the advantage of real time imaging of Nano structure for mono crystalline specimen.

I also believe this method would apply to XRR and powder diffraction.

1. In the first case of XRR, the potential of topographic XRR analyses is exiting, isn't it? The XRR signal covers most specimen surfaces entirely at those shallow angles of diffraction close to "zero" degree 2-theta along diffractometer plane.

2. In the case of powder diffraction, it would be feasible to dramatically enhance the angular resolution using the oblique angle principle without expanding the diffractometer radius, wouldn't it?

The trick is, engineering a fixture to accommodate this "pivot" axis on the detector surface parallel to the diffractometer axis. Thus providing the ability to position the detector surface at an oblique angle to the diffractometer radius. That's exciting isn't it? :-)

Kind of like ADT - Asymmetric Detector Topograph! To clearly distinguish from ACT - Asymmetric Crystal Topography.

Boettinger, W.J.; Burdette, H.E.; Kuriyama, M. et al have many other contributions to ACT as well, I'm sure.

"This technique of image magnification via oblique angle of incidence has been known for several decades" - Correct! However, oblique angle of incidence on the sample surface but not on the detector surface, yes?

This is "Breaking News" on RG. Real time on-line publication takes the cake compared with the delayed gratification of conventional peer reviewed publications. We are making history by publishing this unique, real time, never before published NDE XRD method for magnifying images on RG :-)

I prefer the real time review of thousands of "expert" peers in this group to a bunch of tenured academia making subjective decisions. Here, even the reviewer may be "reviewed", right?

Oblique Angle (of detector plane w.r.t. diffracted beam) vs. Magnification:  Mag.= [1/Sin(angle)]

5o = 11.5x,

4o = 14.3x,

3o = 19.1x,

2o = 28.7x,

1o = 57.3x,

0.5o =114.6x,

0.2o = 286.5x.

The images attached below are for a 3o oblique angle. Both grey scale and pseudo color images are shown. The magnification at this angle would be 19.1x. Since the scale is missing from these images, use 36 Pixels = 1 mm on the detector surface. Both these images are 816x816 Pixels = 22.4x22.4 mm2:

https://www.flickr.com/photos/85210325@N04/31069934415/

https://www.flickr.com/photos/85210325@N04/31034180576/in/photostream/

https://www.flickr.com/photos/85210325@N04/31155851532/in/dateposted/

Simple question! Why hasn't anyone considered the advantage of this method for POWDER DIFFRACTION?

Oh Boy! The angular resolution achievable by using the OAT technique (tilting the detector w.r.t. the diffracted beam) for powder diffraction applications appears to be phenomenal.

An example : At a 5 degree angle of incidence, the 25 mm diameter detector would cover a total angular (2-theta) window of about 1.724 degree (@ SDD=100mm) and each 27.5um pixel on the 2D detector would subtend about 7.6 Arc Sec (0.000613396 degree)!

I'd think, that is phenomenal! Especially, without changing the SDD (sample to detector distance) i.e., the diffractometer radius. What about you? What is your opinion? :-)

Hypothetically speaking, if one used a low divergence "point source" on a crystal to create a fixed expansion with ACT (turning the "point" into a "line"), then it would be feasible to use the OAT method to match the orthogonal magnification precisely and eliminate the "asymmetry in magnification" mentioned earlier in the "Magnifier". The fact that the incident angle of the beam on the detector plane in the case of OAT may be adjusted continuously (unlike ACT) would be a key advantage, wouldn't it?

Any of you that are privileged owners of a PixCel, or Vantek or Pilatus or Mars or other two dimensional detectors should be able to independently verify this claim. The rest of you "proletariat" could use good old dental film. Go ahead, I challenge you to show & tell us all :-)

I'm not sure who to give credit for the ACT method of beam expansion. Sure was around before Kuriyama or Boettinger or Tanner or Bowen. Maybe Berg, Barrett, Lang? Some help and suggestions would be appreciated into the history of ACT :-)