Evaluation of the absolute single-photon detection efficiency of HRPPD Journal: Nuclear Instruments and Methods A

Yifan Jin, Alexander Kiselev, Sean Stoll · 2025

Read Evaluation of the absolute single-photon detection efficiency of HRPPD Journal: Nuclear Instruments and Methods A by Yifan Jin, Alexander Kiselev, Sean Stoll (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2506-16490.txt.

Introduction

Yifan Jin, Alexander Kiselev, and Sean Stoll present a 2025 paper that measures how well a new photon detector catches single photons. The device, a high-rate picosecond photon detector, is bound for ring-imaging Cherenkov cameras that need fine detail on place and timing. Its quantum efficiency had been reported, but the real rate folds in collection losses too.

Using a pulsed diode laser, the team counted photoelectron pulses at one pixel near the center of the device. The detection efficiency came out at about 17.1 percent at a wavelength of 398.6 nm. Quantum efficiency at the same spot was about 24.4 percent. Dividing the two gives a collection efficiency near 70 percent, in line with hopes. The check anchors performance claims for the coming collider detectors.

Catalog record

CREATORYifan Jin, Alexander Kiselev, Sean Stoll
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE22 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2506-16490

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Evaluation of the absolute single-photon detection efficiency of HRPPD Journal: Nuclear Instruments and Methods A Yifan Jin, Alexander Kiselev, Sean Stoll Abstract Pixelated High Rate Picosecond Photon Detectors (HRPPDs) by Incom Inc. are promising photosensors for use in Ring Imaging CHerenkov (RICH) detectors, where a high gain, sub-mm position resolution and sub-100ps timing resolution are required in a single photon mode. Quantum Efficiency (QE) has been measured for the first batch of EIC HRPPDs both by Incom and EIC research groups at Jefferson Lab and Brookhaven Lab, with peak values at ∼ \sim 365 nm typically exceeding 30% ( [ 1 ] ). In this study, we present a first direct measurement of HRPPD Photon Detection Efficiency (PDE) being equal to 19.3 ± \pm 0.2 (stat) ± \pm 0.8 (sys) % at 398.6 nm, in a photoelectron pulse counting mode using a picosecond diode laser. HRPPD QE at the same spot and at the same wavelength was evaluated to be 24.3 ± \pm 0.1 (stat) ± \pm 0.5 (sys) %, leading to a Collection Efficiency (CE) estimate of ∼ \sim 79%, which is consistent with the expectations. 1 Methodology To measure the PDE, we used picosecond diode laser light pulses in a configuration shown in Figure 1 (left). Figure 1: A schematic view of our PDE measurement setup (left) and a QE measurement setup (right). The laser beam was split into two paths: direct light focused onto the HRPPD photocathode, at a location matching the center of one of the anode pads (pixels), and light reflected at 90 degrees impinging on a reference photodiode. While a high photon flux is desirable for accurate current monitoring by this photodiode, a much reduced photon flux is required for the HRPPD to ensure a low photon counting mode. To resolve this discrepancy, a neutral density (ND) filter was placed between the beam splitter and the HRPPD. Assume that the number of photons incident on the HRPPD in a single laser pulse follows a Poisson distribution with a mean \lambda , and that each photon is detected independently with a probability p , corresponding to the PDE. Then, the probability that the HRPPD does not detect any signal from a given pulse is: P(\text{no signal})=\sum_{n=0}^{\infty}P(n)(1-p)^{n}, (1) where P(n)=\frac{\lambda^{n}e^{-\lambda}}{n!} . P(\text{no signal})=\sum_{n=0}^{\infty}\frac{(\lambda(1-p))^{n}e^{-\lambda}}{n!}=e^{\lambda(1-p)}\times e^{-\lambda}=e^{-\lambda p} (2) p=\frac{ln(P(\text{no signal}))}{-\lambda} (3) Note that the mean photon number \lambda does not need to be small for this expression to hold. In other words, operation in a single-photon mode is not necessary for this method. The mean number of photons in a single laser pulse ( \lambda ) incident on HRPPD can be derived from the following expression: \lambda=\frac{I\cdot R}{e\cdot QE\cdot f\cdot A} , where I is the current as measured by the reference Hamamatsu photodiode (see Figure 1 ), QE is its quantum efficiency at the laser wavelength, R is a ratio of the transmitted and the reflected light after the beam splitter at this wavelength, e is the elementary charge, f is a laser repetition rate, and A is an attenuation factor of the ND filter. Since neither the beam splitter ratio R nor QE of the reference photodiode is known precisely enough, the R/QE ratio was measured by a Newport NIST traceable photodiode installed in place of an HRPPD, as explained in detail in Section 3 . The attenuation factor A of the ND filter was also measured separately. Once the R/QE ratio and the A factor are calibrated, Formula 3 gives a PDE estimate. We also evaluated a QE at the same HRPPD active area spot, by measuring a photocurrent at the same photocathode voltage (see Figure 1 right), and extracted the Collection Efficiency (CE) as a PDE/QE ratio. It is worth noting that the photon detection efficiency (PDE) is typically determined using a different method ( [ 2 ] ). In that approach, the pulse amplitude at the anode is measured and then divided by the gain, obtained from a spectral fit, to estimate the number of photoelectrons generated after the photocathode. In contrast, we count signal pulses in a binary manner, thereby eliminating the dependence on gain. 2 Experimental setup A newly established HRPPD test stand at Brookhaven National Laboratory (see Figure 2 ) was used to perform the measurements. The whole set of essential equipment, together with the typical settings, is listed below. Figure 2: Experimental setup. 2.1 HRPPD and readout backplane We used HRPPD #16 (serial number assigned by Incom Inc.) for these measurements. It was mounted on a side wall of the dark box, as shown in Figure 3 . The HRPPD was equipped with a Q02c passive readout backplane and a pair of 32-channel M02c MCX adapters, which allowed us to fully instrument a single 8x8 anode pad spot. Pads of the other fifteen 8x8 pad spots were terminated to ground via 50-ohm resistors using S02c plugin cards. Figure 3: HRPPD #16 mounted on a dark box side wall. One can also see the interface board with a pair of plugin MCX adapters. 2.2 Dark box equipment An optical setup shown schematically in Figure 1 was mounted on an XYZ motion control system (three Thorlabs MTS50/M translation stages) as shown in the same orientation in Figure 4 . A UV-resilient UM22-600 multi-mode fiber with a core diameter of 600 \mu m by Thorlabs was used between the dark box patch panel and the optical head for both PDE and QE, as well as for the calibration measurements. The light was focused by a LA1422 N-BK7 plano-convex lens with a focal distance of 40 mm, with its intensity and angular spread limited by a SM1D12C diaphragm with a chosen \sim 3.5 mm diameter setting, and split into two paths by a CM1-BP145B5 pellicle beam splitter, all by Thorlabs. Light reflected at 90 degrees was detected by a Hamamatsu S1226-8BQ reference photodiode installed in a custom support inside a 1" diameter tube, at a distance roughly the same as the distance between the beam splitter and the HRPPD window. Light passing through was detected by either an HRPPD during the measurements (as shown in the picture) or by a NIST traceable Newport 818-UV-DB photodiode for calibration purposes. According to the factory test report, the photodiode responsivity was equal to {1.6864E-1 A/W at 390 nm and 1.7957E-1 A/W at 400 nm} in a configuration without an external OD3 filter and {4.0227E-4 A/W at 390 nm and 4.2081E-4 A/W at 400 nm} with the filter installed. The uncertainty of this calibration is supposed to be 1.65% at the wavelength of interest (398.6 nm, see section 2.4 below). After a simple linear interpolation, conversion to quantum efficiency units, and error propagation, we obtain a QE of the Newport photodiode at 398.6 nm to be equal to 55.4 \pm 0.9 %, and the attenuation factor of the factory OD3 filter at this wavelength equal to 426 \pm 10. None of the elements shown in Figure 4 was either changed or moved between PDE and QE measurements. In particular, we used the same fiber, diaphragm diameter setting, and XYZ-coordinates of the optical head. For calibration, either an ND filter or a Newport photodiode was mounted downstream of the beam splitter as explained in the subsequent sections. Figure

Read the complete text