Development and testing of cost-effective, 6 cm×6 cm MCP-based photodetectors for fast timing applications
Introduction
Photomultiplier tubes (PMTs) [1] are current-source amplifiers with high gain, high quantum efficiency (QE) and the ability to resolve single photons. PMTs have a time resolution of a few hundred picoseconds [2], but the complexity of their discrete dynode configuration limits their capability to provide precise spatial information and makes them susceptible to magnetic fields. Multi-anode PMTs with a more compact structure have similar performances as standard PMTs and allow for better position information as well [3]. However, both standard and multi-anode PMTs are not suitable for applications which require simultaneously very high time (<50 ps) and position resolution (<50 μm). Recently, the demands of nuclear and particle physics experiments call for the next generation of photodetectors with capabilities better than those of traditional photomultipliers [4], [5]. Therefore, alternative technologies for fast single photon detectors have been developed, such as the microchannel plate photomultiplier tubes (MCP-PMTs).
MCP-PMTs [6] are compact photodetectors capable of picosecond level time resolution [7], [8] and sub-mm position resolution [9], [10], and are an evolution from the basic principles of traditional PMTs. As a key element, a MCP consists of millions of parallel conductive glass capillaries (6–40 μm). Each capillary is an independent secondary-electron multiplier that can be considered as a continuous dynode structure. Such a structure provides local electron multiplication with a very small path length, resulting in an extremely fast time response. Compared to other photon sensors, such as traditional PMTs and solid-state detectors, the new generation of MCP-PMTs described here is able to simultaneously provide high-gain (106–107), high QE (>20%), excellent time and position resolutions. In addition, MCP-PMTs have shown potential in applications involving strong magnetic fields [11], [12], [13]. All these characteristics make them an excellent candidate for the next generation of photodetectors. If MCP-PMTs could be made considerably more cost-effective and robust, they would be widely used in a variety of applications in the fields of particle and nuclear physics, astrophysics and medical imaging.
The Large-Area Picosecond Photodetector (LAPPD) project [14] is a US Department of Energy (DOE) funded collaborative project, with the goal of developing low cost, commercializable methods to fabricate 400 cm2 thin planar photodetectors based on Atomic Layer Deposition (ALD) coated MCPs [16]. The production process of the MCP substrates, developed by Incom Inc.,1 is based on the use of hollow capillaries in a glass drawing process, taking the place of the traditional chemical etching process. The MCP substrates are functionalized by an ALD coating method at ANL, which controls the necessary resistive and secondary emission properties [15]. The ability to engineer the material properties offers the opportunity to improve the performance and reduce the cost for MCP-based devices. During the past few years, the LAPPD collaboration has made substantial progress on the MCP fabrication and the photodetector development [15], [16], [17], [18], [19], [20], [21], [22], [23], [24].
Argonne National Laboratory has built a photodetector processing system, capable of producing 6 cm×6 cm active area, glass body, cost-effective photodetectors. Two critical issues were addressed for the production of robust photodetectors: indium-based hermetic sealing and MCP outgassing. Recently, a series of long-lived prototype photodetectors were produced, representing a significant milestone for the production of hermetically sealed functional devices. A testing facility was built for characterizing the MCP-based photodetectors. It uses a pulsed blue laser (405 nm)2 with duration time of 70 ps FWHM and allows for testing the fast time response as well the gain and uniformity of the photodetectors. In this paper, we present the development and testing of the 6 cm×6 cm MCP-based photodetectors.
The paper is organized as follows: Section 2 describes the design of the photodetector processing system; Section 3 presents the structure of the MCP-based photodetectors; Section 4 describes the laser test facility and the experimental setup; Section 5 describes the data analysis method; Section 6 presents the performance of the photodetectors. Finally, Section 7 gives our conclusions and outlook.
Section snippets
Small single tube processing system
The LAPPD collaboration initially planned to build a photodetector production facility for producing 20 cm×20 cm MCP photodetectors in an all-glass package one detector at a time. Considering the high cost and complexity of building such a large system, ANL has built a smaller system that would serve as an R&D facility and as an intermediate step towards the production of the full-size detector and their commercialization. The small single tube processing system (SmSTPS) has the ability to
Stack components
The modular design of the MCP-PMT is shown in Fig. 2. This design inherits its design concept from the LAPPD detector [26] but with a small form factor, based on an all-glass hermetically sealed tube package that consists of the following components:
- 1.
A glass bottom plate on which is silk-screened a stripline anode readout [25].
- 2.
A glass side wall.
- 3.
A resistively matched pair of MCPs separated by a grid spacer.
- 4.
Three glass grid spacers.
- 5.
A glass top window with a bialkali (K,Cs) photocathode.
- 6.
An indium
Experimental setup
We have established a laser based test facility for characterizing the photodetectors, using a pulsed laser emitting blue light with a wavelength of 405 nm and a typical pulse width of 70 ps FWHM. The experimental setup is shown schematically in Fig. 4. The optics are implemented in two stages that are covered by light-tight dark boxes.
Optics in the first dark box is designed to produce a narrow laser beam. The blue light emitted from the laser diode, with an emission angle of 15°, is attenuated
Data analysis
The waveforms are recorded by a Tektronix DPO7354 oscilloscope with a 40 Gs/s sampling rate and a 3.5 GHz bandwidth, and the data analysis is done offline in software. The data analysis is done in several steps, including waveform processing, pulse selection and timing discrimination.
Prior to the pulse selection, a Fast Fourier Transform (FFT) algorithm is applied to the raw waveforms. In the frequency domain, the highest signal component is about 1 GHz (see Section 6.2), well below the analog
Results and discussion
A number of 6 cm×6 cm photodetectors have been produced and characterized at ANL. The QE of the photocathode has previously been demonstrated to reach a 15% at λ~350 nm [24]. The test results presented are based on a typical Argonne prototype (see Fig. 3) with 20 μm MCP pore size.
Conclusion
Argonne National Laboratory is currently focusing on the development of cost-effective, small form-factor (6 cm×6 cm), MCP-based picosecond photodetectors with scalability to extend then to large area. Recently, a significant milestone of producing the first fully processed and hermetically sealed MCP-based photodetector has been reached at the small single tube processing system. A series of functional 6 cm×6 cm prototypes have been produced and characterized. The R&D and production experience
Acknowledgments
We would like to thank Ronald Kmak (ANL) for the design of the vacuum chamber. We also thank Joe Gregar (ANL) of the Argonne glass shop, for his talent work on the frit seal. We are deeply grateful to Matthew Wetstein (University of Chicago) and Bernhard Adams (ANL) for their advice on detector testing. Work at ANL was supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences and Office of High Energy Physics under contract DE-AC02-06CH11357. Use of the
References (29)
Nuclear Instruments and Methods
(2006)Nuclear Instruments and Methods A
(2004)Nuclear Instruments and Methods A
(2006)Nuclear Instruments and Methods A
(1995)Nuclear Instruments and Methods A
(2015)Nuclear Instruments and Methods A
(2015)Nuclear Instruments and Methods A
(2008)Nuclear Instruments and Methods A
(2004)Nuclear Instruments and Methods A
(2015)Physics Procedia
(2012)
Physics Procedia
Nuclear Instruments and Methods A
Nuclear Instruments and Methods A
Nuclear Instruments and Methods A
Cited by (14)
Characteristics of fast timing MCP-PMTs in magnetic fields
2019, Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated EquipmentCitation Excerpt :At a fixed magnetic field strength, the signal amplitude increases with increasing bias high voltage. This behavior confirms our previous measurements of MCP-PMTs in a negligible magnetic field [23]. At a fixed bias voltage of 3100 V, the signal amplitude of the MCP-PMT increases slightly as the magnetic field strength increases to 0.2 T, and then is seen to decrease as the magnetic field strength continues to increase, and eventually is reduced below 5 mV at a magnetic field strength of 0.7 T. With lower bias voltages, the signal amplitudes are seen to be reduced already at lower field strength.
First experimental demonstration of time-resolved X-ray measurements with next-generation fast-timing MCP-PMT
2019, Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated EquipmentRate capability and magnetic field tolerance measurements of fast timing microchannel plate photodetectors
2018, Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated EquipmentCharacterisation and testing of a prototype 6×6 cm<sup>2</sup> Argonne MCP-PMT
2017, Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated EquipmentCitation Excerpt :The MCP has a pulse width of approximately 5 ns @10% of the pulse maximum and a rise time of ∼0.5 ns. Ref. [4] reports a transit time spread of 0.06 ns for single photons. As can be seen in Fig. 2 (bottom) a Butterworth low bandpass filter is applied to remove noise above 700 MHz.
Coating and functionalization of high density ion track structures by atomic layer deposition
2016, Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated EquipmentCitation Excerpt :In previous work by others, ion track structures and ALD have been combined to make nanotubes after dissolving the soluble nano templates [23]. On the other hand, ALD has been used to coat with thin films MCP structures made of glass [24], lead glass [25], borosilicate glass [26,27], porous silicon [28] and anodized Al2O3 [29]. These ALD-MCP structures showed mostly similar or even better properties than conventional MCPs based on lead glass, and had increased gain and improved stability as a function of extracted charge in several cases.