X-ray Microcalorimeter Spectrometer (XMS)
X-ray Microcalorimeters
At its most basic, calorimetry is the process of measuring the energy put into a closed system (or taken out of a system) by measuring the change in temperature of that system. X-ray microcalorimeters work in the same fashion, but they measure very small energy differences.
Since X-rays are very energetic, measuring the temperature difference of a material after absrobing an X-ray photon may not sound too difficult. However, if a 6 keV photon were to be absorbed by a typical US penny, its temperature would only change by 3 parts in 1018 (that's 3 parts in a billion-billion)! (This example is from the GSFC X-ray microcalorimeter group webpages.) To overcome this problem, a microcalorimeter uses materials which have a much smaller heat capacity than copper and runs at a very low temperature (< 1 Kelvin).
There are three main parts to the microcalorimeter pixel: an absorber, a "thermometer", and a heatsink. An X-ray entering the detector may be absorbed by the absorber, causing the absorber's temperature to rise. The "thermometer" then reads the temperature change in the absorber due to the X-ray. The "thermometers" on the Constellation-X XMS will be transition-edge sensors (or TESs), though a temperature-senstive transistor, called a thermistor, can also be used. The whole system is linked weakly to the heatsink, which allows the "thermometer" and absorber to return to their original temperatures to be ready to detect another X-ray.
Cartoon showing the different parts of an X-ray microcalorimeter.
Constellation-X XMS
The X-ray Microcalorimeter Spectrometer (XMS) will be an array of 32 by 32 microcalorimeter detectors. Each detector will be 250 microns on a side, and will consist of a TES (see below for more on TESs), an X-ray absorber, and a membrane thermal link to the heatsink. The heatsink will operate at a temperature of 50 milli-Kelvin, or 50 thousandths of a degree above absolute zero (see below for a bit more on how we achieve such low temperatures).
The Suzaku/XRS microcalorimeter array. The pixels are each 625 micrometers on a side.
Sensing Tiny Temperature Changes: Transistion Edge Sensors
As mentioned above, the "thermometers" that will measure the temperature difference in the absorber due to an incomming X-ray will be transistion-edge sensors, or TESs. A TES is a thermometer made from a superconducting film. Typical substances have resistance to electricity. Even the powercord for your laptop has some resistance to electricity, though it's much smaller than the resistance of the plastic coating protecting you from touching the wire. Superconductors, however, can be made to have zero resistance to electricity, if you lower their temperature. Each superconducting substance has a different temperature at which it will become superconducting, and this temperature is called Tc, or the "critical" temperature. The transition from non-superconducting to superconducting happens almost instantaneously at that temperature – at half a degree above Tc or even a tenth of a degree above Tc, the material will not be superconducting. There is, however, a very small range of temperatures, usually about a milli-Kelvin around Tc (or one one-thousandth of a degree around Tc), where we can watch the resistance decrease. The TESs are set to a temperature in the transition from normal to superconducting. When an X-ray is absorbed, the temperature of the TES changes by just a small amount, but it drives the material to normal resistance, producing a huge, and measurable, change in the electrical resistance of the material.
Getting Cold: Adiabatic Demagnetization Refrigerator
In the lab, we can use solid substances, such as neon and helium, to achieve fairly low temperatures (much as you might use ice to cool a beverage, since ice is just solid water). However, neon is solid at 27 Kelvin, and helium becomes solid at 0.95 Kelvin. These are still much too hot for the work with microcalorimeters, which requires temperatures of 0.050 Kelvin.
To cool our detectors that extra bit, we use adiabatic demagnetization refrigerators, or ADRs. This method uses the properties of paramagnets and thermodynamics. Paramagnets are substances that have regions with different magnetic spins. You've probably played with a magnet, and know that there is a "north" and "south" pole on each magnet. Even if you cut a magnet in half, the two halves still have a "north" and a "south" pole. In fact, you could continue cutting that magnet in half a hundred times, and each of the pieces would have a north and a south pole.
Inside a paramagnet, there are small regions, each with a north and a south pole, but these poles are randomly oriented (see image). When the material formed, they "froze" in these random orientations, and it is not worth it, energetically, for those spins to try to align themselves. However, if you put that paramagnet into a strong magnetic field, all of those little north and south poles would line up with that magnetic field.
ADRs work by putting a paramegnetic "salt pill" into a strong magnetic field produced by an electromagnet. As more current is run through the electromagnet, the magnetic field ramps up and the spins of all those tiny magnetic spins in the paramagnet start to align themselves with the magnetic field. Once the spins are all aligned, the paramagnet is isolated thermally from its surroundings. It is still in the magnetic field, but is not in contact with anything that could transmit heat. Then, the magnetic field is ramped back down. As the magnetic field decreases, those spins start to return to their original orientations. From the laws of thermodynamics, we know that as the chaos (or entropy) of a system increases, the temperature decreases. Thus, we can lower the temperature of that paramagnetic salt pill.
To cool the detector, then, we connect it thermally to the paramagnetic salt pill. As the temperature of the salt pill decreases, the detector will also decrease in temperature. Using this technique, we can achieve temperatures of 0.050 milli-Kelvin.
In the presence of a weak magnetic field, some of the magnetic spins (which can be thought of as like the needle on a compass for small sections of a paramagnetic salt) in a paramagnetic salt will align with the magnetic field, while others will not. However, when a strong magnetic field is applied, all of the magnetic spins would line up with the field. In an adiabatic demagnetization refrigerator, the paramagnet and detector would be thermally isolated. Then, as the magnetic field was ramped down, the spins would become random, causing the temperature to decrease.
