Design of a Thick Gas Electron Multiplier based photon pre-amplifier

In this paper we present the design of a photon pre-amplifier based on a photo-cathode coated Thick Gas Electron Multiplier (THGEM). Such device is crucial in application where a weak light signal produced in a radiation detector must be amplified so that it can be carried to a photo-detector by means of optical fibres. An example of a device where a light signal must be amplified is a gamma-ray Cherenkov detector for fusion power measurements in magnetic confinement devices. In such application the active part of the detector must be located very close the plasma, typically in a harsh radiation environment where standard photodetectors cannot operate. The photon pre-amplifier allows to increase the signal generated in the active part of the detector so that it can be easily detected by the photodetector located outside the harsh environment. We present the conceptual design of a THGEM based photon pre-amplifier supported by Garfield++ simulations. The device working principle is the following: primary photons impinge on the photo-cathode and extract electrons that are accelerated by the THGEM electric field. Upon collisions with the accelerated electrons, the gas molecules in the pre-amplifier are brought to excited states and de-excite emitting scintillation photons. Since each electron excites multiple gas molecules, the scintillation photons outnumber the primary photons, leading to the amplification. In addition, we present the first observation of measurements of Nitrogen gas scintillation in a THGEM device. We devised an experimental setup consisting of a vacuum chamber containing a THGEM and an alpha particle source. The vacuum chamber is filled with pure nitrogen and is coupled to a photomultiplier tube via a view-port to detect the scintillation photons generated in the THGEM. For sake of simplicity the electrons that induce the scintillation are generated by the ionization track of an alpha particle rather than by the THGEM photo-cathode coating. A good qualitative agreement between simulations and experiment has been found, however no quantitative conclusions can be made due to the lack of N2 excitation cross sections in the Garfield++ code.