The Detector

Microchannel Plate Detectors

The detector for our payload was a Microchannel Plate Detector (MCP), commonly used in the UV and X-ray wavelength regimes. The detector was built by Oswald Siegmund at Sensor Sciences, who has made numerous high quality MCPs for a variety of major missions.

MCP's are great because they are fast, low-noise and read-noise free, as well as being photon counting (they can detect and record individual photon events).

MCPs roughly work by converting photons into electrons and then multiplying them to allow for detection. A microchannel plate is a thin wafer made of glass that has thousands of small pores inside of it, like in the diagram on the left. The top of the plate is coated in a photocathode that converts photons into electrons, and there is a large voltage differential between the top and bottom of the plate. When a photon comes in, it hits a particular pore, creating an electron. Due to the voltage, that electron is sucked into the pore, where it hits the wall and generates a secondary electron. Those electrons continue down the pore, hitting again, multiplying, hitting again, multiplying, etc. Most designs have a second plate immediately under the first to increase the gain even further. By the time the electrons exit the back end of the bottom most plate, they are numerous enough to be detected by an anode that sums up their charge.

These anodes are ingenious solutions to the problem of locating the position of the original photon in X-Y space. A typical design features two wires, each coiled in a set of parallel lines that covers the detector active area. The two coiled wires are perpendicular to each other, as shown in the image below. When a photon hits and generates an electron cloud, that cloud is transferred to the anode, and travels along it. Eventually, the cloud reaches each end of the anode, and is registered by the detector electronics. By timing the difference in arrival times of the cloud at each end of the anode, the cloud's position along the anode can be determined. By having two perpendicular anodes, this allows a cloud's X/Y position to be locked down.

The detector registers photons by recording their x position, y position, time of detection, and pulse height, which loosely translated to how many electrons the incoming photon generated, and can be used for diagnostics. Since the detector registers time (unlike, say, a CCD), spectra can be analyzed temporally. For a CCD, a cosmic ray or similar set of events near the end of an exposure might ruin an image, but in an MCP you can filter out that segment of data and use the rest of the spectrum. This allows for some useful and interesting diagnostics to be performed. For instance, I could plot my count rate as a function of time, and get a sense of how much my spectrum was being affected by the atmosphere as my rocket took data at different altitudes in its parabolic arc. Very cool.

Our detector was 200mm x 200mm, making it the largest MCP detector ever flown in space. Part of the science motivation for the payload was flight testing of these extremely large UV detectors, as a precursor to future flagship missions like the LUVOIR UV concept. The detector survived multiple launches and a litany of vibration tests while performing perfectly, showing that these large detectors have a real place in future UV astronomy.

The detector photocathode was Potassium Bromide (K Br), which was chosen to maximize signal in the 600-800A bandpass so as to maximize our chances of a useful spectrum there. (Since the 700A+ wavelength regime is unknown for hot stars and is central to our science case of estimating B star ionization potential).

Our rough resolutions were approximately 100 micrometers in X and 150 in Y. These detectors can generate a much higher resolution, but that's all that was needed for our science case.

The detector photocathode is highly sensitive to moisture and hydrocarbons, and would degrade rapidly upon exposure. From the moment the photocathode was deposited, through build up, integration, testing, flight, and recovery, the detector never saw any environment besides pure gaseous nitrogen or high vacuum.

I installed the detector into the DEUCE payload in August of 2017. See below for photos of the detector.