The EPIC-MOS instruments on-board XMM-Newton



For many more images of both XMM-Newton hardware and astronomical observations, please see the ESAC XMM-Newton gallery page.


XMM-Newton has 3 X-ray telescopes on-board, each with its own European Photon Imaging Camera (EPIC). Although the X-ray telescopes focus' the X-rays using mirrors, the configuration is more reminiscent of a refracting telescope (explained below).

Two of the X-ray telescopes use EPIC-MOS (Metal Oxide Semi-conductor) cameras, developed and built at the University of Leicester's Space Research Centre, and the third telescope uses an EPIC-PN camera, built at Germany's Max-Planck-Institut für extraterrestrische Physik (Garching) and the Astronomisches Institut Tübingen. Both EPIC-MOS cameras consist of seven CCD's, with each CCD made up of 600x600 pixels - the EPIC-MOS are essentially two 2.5Mpx digital X-ray cameras! The EPIC-PN camera is of a different design, and contains 12 CCD's of 64x189 pixels.

What do the cameras measure?

For each X-ray, the EPIC cameras record the following fundamental data:

» the time when the X-ray arrived,
this allows astronomers to see the rate at which X-rays arrive at the telescope, and so how the X-ray brightness of the astronomical object changes over time. A graph of brightness against time is known as a light-curve. Light-curves allow astronomers to see if the target suddenly brightens, or if it is fading.
» where the X-rays hit the camera's sensor,
this enables an image to be produced, allowing astronomers to see exactly where the X-rays are originating from in space.
» the energy of the X-ray
measuring the energy of the X-rays allow astronomers to understand the physical processes that are occurring at the target. Astronomers can directly measure such quantities as the temperature of the target, how many X-rays are absorbed by intervening gas, what chemical atoms are present, and much more - all by looking at the energies of the incoming X-rays.

These quantities can also be use in combination. For example, astronomers will often make a light curve of the low energy X-rays, and another light-curve of the high energy X-rays, and compare the two. If a cloud of gas drifts over the source of X-rays, it preferentially absorbs low energy X-rays - so if an astronomer sees a dip in the low energy light-curve, but no change in the high energy light-curve, then they will presume that and absorbing cloud of gas has drifted over the source of X-rays. Also, by comparing a low energy image with an high energy image, an astronomer can instantly see where the hotter and cooler gas is physically located.

Example observations by EPIC-MOS

OY Carinae: a binary star system

This light-curve shows the number of X-rays counted over time - and the X-ray source appears to suddenly turn off, and back on again! OY Carinae is actually a pair of stars, and the X-rays are originating from just the smaller, white dwarf star. As the two stars orbit one another, the second larger star passes in front of the white dwarf star, once every 90 minutes, completely hiding the white dwarf (and the X-rays) from view for about 4 minutes. For more details, see the HEASARC picture of the week archive.

Geminga: a rapidly moving neutron star

This image is of Geminga, a neutron star moving rapidly though space at 120 kilometres per second, leaving behind a pair of tails of X-ray hot gas, stretching 3 million million kilometres across the sky. The neutron star itself is only 20-30 kilometres across and is the dense remains of an exploded star. Geminga lies at a distance of about 500 light-years away. Its rapid movement creates a shock-wave that compresses the gas of the interstellar medium and its naturally embedded magnetic field by a factor of four. For more details, see the European Space Agency website.

Capella: a bright giant star

Our Sun produces X-rays in its outer atmosphere (the corona), due to the intense magnetic field heating the gas up to millions of degrees centigrade - hot enough for the gas to emit X-rays. Some stars, such as Capella, have an even more active corona than the Sun does.
A spectrum is simply a graph of the number of X-rays observed with a given energy. In this case, the energy of each X-ray is given in kilo electron-volts (keV); 1 keV is equivalent to the average energy of an X-ray from a gas at 11 million degrees kelvin (or Celsius). Atoms in a gas emit (and absorb) X-rays at very specific energies, depending on the atomic element, temperature, and density of the gas. So by looking at the peaks in a spectrum (called spectral lines) astronomers can determine the exact nature of the X-ray emitting gas. For instance, in Capella's spectrum, the spectral peak (line) in the graph at just below 2 keV indicates that magnesium is present, the line at 1 keV is from iron and neon, and the line between those two at 1.35 keV is also from iron.
For more about coronally active stars, see the HEASARC picture of the week website.

Modes of the EPIC-MOS Camera

The number of images (frames) taken per second by the EPIC camera's is limited by the speed at which each frame can be read-out - there are 360,000 pixels to read per CCD! The CCD's are continuous exposed to the sky, and at the end of each frame, all 600 columns are rapidly transfered to a read-out storage area together, within a few milli-seconds. The data in this storage area is then read-out row-by-row (taking 2.6 seconds in full-frame mode - 600 times longer than the transfer time), before being prepared for transmission to the Earth.

In full-frame mode, the data from all 360,000 pixels are read-out from the storage area in 2.6 seconds, and so this limits the number of images that can be taken per second. However, astronomers can choose between receiving a full 600x600 image once every 2.6s, or a smaller 300x300 image every 0.9s (large window mode), or a tiny 100x100 image every 0.3s (small window mode). For very high time resolution, astronomers use the EPIC-MOS timing mode, which takes an 100x1 pixel image every 1.5ms!

The ability to use these modes are, however, dependent on the target. A source could be either too faint, such that an X-ray is not even detected within a frame time, or too bright - if two X-rays arrive at the same pixel at the same time (called pile-up) confusion can arrive (has one high-energy X-ray arrived, or two low energy X-rays?!). In full-frame mode, pile-up can occur if more than one X-ray is detected from a source every 1.4 seconds.

The Mirrors

XMM Mirrors

X-ray observatories cannot use the conventional mirrors that we are used to in everyday life, since X-rays would just pass straight through the gaps between the atoms!
The reason is that different wavelengths of light have different interaction cross-sections (or interaction area; a measure of the likelihood that an X-ray will interact with an atom). The larger the wavelength of light, the larger its interaction cross-section, and the larger the chance of an interaction i.e. an reflection.
Optical light has a large wavelength and a large interaction cross section, and so is easily reflected by (interacts with) a polished surface. X-rays, however, have a short wavelength (high energy) and a small interaction cross-section - so they can pass between the atoms of a mirror without interacting with them at all.

How X-ray mirrors focus X-rays
X-ray mirrors focus the X-rays from space (coming in from the right) by redirecting the X-rays, through reflection, into a focus point on the cameras at the far left.

Imagine a fence (which represents a mirror), with evenly spaced posts (the atoms within the mirror). A large football (our optical photon with a large wavelength and so a large interaction cross-section) will be easily reflected off the fence. However, a small tennis ball (our X-ray, with a small interaction cross-section) will pass straight through the fence. So how do we focus an X-ray?

Somehow, we need to get the incoming X-rays to interact with (i.e. "hit") more atoms. We can do this by turning the mirrors edge-on, and so the X-rays are more likely to hit an atom thus bringing the X-rays to a focus point. The X-rays are skipping off the atoms like a stone skipping over water. As you can see with the fence analogy, an incoming X-ray will see more atoms (or fence posts) when the mirror (fence) is placed edge on.

A large football (representing optical light with a large interaction cross section) reflects back to us off the fence posts (the spacing of the fence posts represents the spacing of atoms in a mirror).
However, the small tennis ball (representing an X-ray photon with a small interaction cross-section, or "size") passes straight through the fence (atoms).
So how do we focus an X-ray (or tennis ball)? Place the fence (or mirror) edge-on, and the tennis ball (X-ray) will be more likely to hit a fence-post (or atom) to enable it to be brought to a focus.

Here is how to imagine it if you prefer to think of photons as wave packets of energy. We have two mirrors, one vertical and one almost edge-on. The optical wave (red, top) reflects off the mirror atoms (green) - but the (blue) X-ray passes straight through. However, the X-ray cannot avoid hitting an atom when the mirror is placed edge-on, bringing all the X-rays to a focus point.

Indeed, arguably the most complex part of X-ray observatories is the manufacture of the mirrors. The biggest problem with X-ray mirrors are the fact that only a tiny fraction of the incoming X-rays are brought to a focus - most just pass between the mirrors. In order to get around this problem, the designers of the XMM-Newton observatory placed 58 nested mirrors within each other (below, left), in an attempt to catch as many X-rays as possible. Whereas, designers of NASA's Chandra X-ray Observatories opted for only 4 mirrors (below, right), but four mirrors so accurately made that the resulting images are much sharper that the images taken by XMM-Newton.



XMM Mirrors Chandra Mirrors
The XMM-Newton Mirrors Chandra Mirrors

Page maintained by Dr Tao Song

Share this page: