X-rays pass through everything, right? After all, that's how a dentist checks your teeth and airport security examines your bags.

Not quite. X-rays are photons, as are light and radio waves, but with such a high energy that they penetrate most materials. So using mirrors just like those on the Hubble will not work. But we can use a similar design by employing grazing incidence reflection.

Hold a sheet of glass upright and it's just glass. Tilt the glass almost edge-on and it it's a mirror. When the angle is shallow enough - the "critical angle," which varies with the material and the photon energy - light reflects rather than passes through. That's why you see glare on automobile windshields.

We do the same with X-rays if we coat the glass with a smooth metal finish. But what shape shall we give the mirrors?

A technician inspects one of the polished, coated AXAF mirrors.	AXAF's mirror assembly consists of four mirrors nested within each other.	The completed AXAF high-resolution mirror assembly is readied for installation in the X-ray calibration facility.

We start with the same kinds of shapes used by Hubble's two mirrors, a parabola for the primary mirror and a hyperbola for the secondary mirror. These are lines that always curve but gradually flatten out so the curve never closes. Light that enters straight into a parabola or hyperbola is reflected to a focal point and form an image of the object that sent the light. No telescope design provides a perfect image, so a second curved mirror is used to remove most of the distortion and provide a high-resolution image over a wider field of view.

Hubble's mirrors use the parts of the curves near the bowl-shaped apex. For X-ray telescopes, we take a section from farther up the curve, where it is almost flat. That way, the X-rays strike at a very shallow angle - less than 1 degree on AXAF - and are focused rather than passing through the mirror. This design is called a Wolter Type I for the optical scientist who developed it.

The mirrors are precisely ground glass and coated with iridium to reflect X-rays. The smoothness of the mirrors is equivalent to polishing the Earth to within 2 meters (78 inches) of being a perfect sphere.

But we have introduced a problem. By using a section distant from the center, we no longer have a large area to collect X-rays. It's a thin ring that would make faint pictures. So we build four sets of mirrors, each smaller than the one before it, nest them inside each other, and align them to within 1.3 microns (about 1/50th the width of a human hair) so they focus at the same point. This collects more X-rays and produces a brighter, sharper image than just one mirror.

Go here to read about the AXAF mirrors, or return to AXAF main story.

Updated Feb. 4, 1997

return to top of page

Authors: Dave Dooling
Curator: Bryan Walls
NASA Official: John M. Horack