Measurements of the charging-up effect in Gas Electron Multipliers

https://doi.org/10.1016/j.nima.2020.164205Get rights and content

Abstract

Gas Electron Multipliers (GEM) are widely used as amplification stages in gaseous detectors exposed to high rates. The GEM consists of a polyimide foil which is coated by two thin copper layers. Charges collected into the holes or created during the amplification process may adhere to the polyimide part inside the holes. This is often accompanied by a change of the effective gain. The effect is commonly known as the “charging-up effect”.

This work presents a systematic investigation of the effect under well-controlled and monitored conditions for a single standard GEM foil in Ar/CO2 (90/10) gas for varying rates of X-ray interactions in the detector and for different gains of the foil. In order to cover a wide range of different rates, we apply two different methods. The first one is based on a current measurement while the second one relies on the analysis of 55Fe spectra over time. Both methods give consistent results, showing an initial increase of the effective gain with time and an asymptotic saturation, which can be well described by a single-exponential function. We find that the characteristic time constants extracted from our measurements scale inversely proportional to the rate of incoming electrons for a given GEM voltage. Introducing characteristic quantities, which describe either the number of incoming electrons per hole or the total number of electrons in a hole per characteristic time, we find consistent numbers for measurements taken at the same GEM voltages. For measurements taken at different GEM voltages, however, also the characteristic total number of electrons inside the hole, which is supposed to take into account the different effective gains, is found to be higher by a factor of about 3.5 for UGEM=350V (ntot=8.8×108) compared to 400 V (ntot=2.4×108). This hints at a residual dependence of the charging-up characteristics on the GEM voltage.

Introduction

Detectors based on the Gas Electron Multiplier (GEM) [1] are widely used in particle physics experiments that require high position resolution over large areas in high-rate environments (e.g. COMPASS [2], [3], LHCb [4], TOTEM [5], [6], JLab Hall A [7] as well as ALICE [8], [9] and CMS [10], [11] after their upgrades). The GEM consists of a 50 µm thick polyimide foil which is coated on both sides with 5 µm thick copper layers. In a photolithographic process, holes are etched into this foil in a hexagonal pattern. Standard GEM foils have an inner diameter of approximately 50 µm, an outer diameter of approximately 70 µm and a pitch between two neighboring holes of 140 µm. If a suitable voltage is applied between both copper layers, strong non-uniform electric fields are created inside the holes (of the order of 50 kV/cm), sufficient for incoming electrons to start an avalanche of further ionizations. During this multiplication process, electrons and ions may diffuse to the polyimide part of the GEM and be adsorbed there as shown in Fig. 1. Due to the high resistivity of the material, the charges remain there for a rather long time. These new charges accumulate over time and dynamically change the electric field inside the hole. This is known as the “charging-up effect”. Many publications suggest that the charging-up effect is responsible for a change of the effective gain which is time-dependent and reaches asymptotically a constant value (e.g. in measurements with GEMs [3], [12], [13], [14], [15], in measurements with different micropattern gaseous detectors [16], [17], [18], as well as in simulations [19], [20], [21]). A quantitative understanding of the various effects reported (which e.g. include both an increase [3] and a decrease [15] of the effective gain), however, has not been reached in our opinion. In addition, as we had to experience ourselves, there are many other external effects which may mimic a genuine charging-up effect. Examples are changes in external conditions like temperature and pressure, initial instabilities of X-ray generators or time constants intrinsic to the high-voltage power supply system (see also Section 2). Especially for applications that require the gain to remain very stable over time, e.g. in Time Projection Chambers for dE/dx measurements [8], [22] or in photon detectors [23], a quantitative understanding of the effect is indispensable.

This work presents a systematic investigation of the charging-up effect under well-controlled and monitored conditions. By using two different methods, we are able to investigate the time-constant over a wide range of rates of X-ray interactions. The first method makes use of a conventional X-ray tube where the amplified ionization currents are sufficiently large to be measured with a picoamperemeter. For the second method, the peak position of the Kα line in an 55Fe spectrum is observed over time. Both types of measurement were conducted using a single GEM as amplification stage [24].

Section snippets

Setup

In order to measure the charging-up effect, a dedicated detector was set up with a single GEM foil as amplification stage. A sketch of the used detector is shown in Fig. 2. It consists of an aluminum vessel with a 100 µm thick Kapton®400FN0221 window on the top side to irradiate the detector with X-rays. The detector was constantly flushed with Ar/CO2 (90/10) with a flow of 3 L/h. The total gas volume is approximately 4 L.

Measurement method I – Current measurement

The first measurement method makes use of a picoamperemeter (originally developed at TU München and further improved at Bonn University [27]) which was connected to the pad plane of the detector. In its most sensitive mode, the picoamperemeter has a digital resolution of 0.5 pA and an absolute accuracy of 2 pA, given by a known temperature dependence and possible calibration uncertainties. Over a time interval of approximately 10 s, it measures 128 values and sends out the average value as well as

Measurement method I

The results from measurement method I are depicted in Fig. 3, Fig. 4, which show the currents measured at the readout anode as a function of time for different currents of the X-ray tube and for two different GEM voltages (400 V in Fig. 3 and 350 V in Fig. 4. All measured currents exhibit an initial increase with time and a saturation after longer irradiation times. To each data set, a single exponential function of the form Ireadoutt=Isat1ΔIIsatexptτwas fitted. Ireadout denotes the measured

Discussion of the results

In the presented measurements, the charging-up effect of a single GEM foil was examined at different X-ray rates and at different GEM voltages. All measurements show an effective gain which initially increases with time and asymptotically reaches a saturation value. Each single measurement can be described by an exponential function, see Eq. (2). This implies that the net accumulation of charges on the polyimide decreases proportionally to the distance from the saturation value, as it is also

Conclusion and outlook

In this work, the charging-up effect of single standard GEM foils was investigated experimentally under well-controlled conditions. In order to study the rate dependence of the effect over a wide range of X-ray interactions in the conversion volume above the GEM foil, we made use of two different measurement methods. For the first method, a conventional X-ray tube was used which provided a high rate of ionization electrons, allowing for the measurement of currents on the readout anode. The

CRediT authorship contribution statement

Philip Hauer: Conceptualization, Methodology, Investigation, Writing - original draft. Karl Flöthner: Investigation, Validation. Dimitri Schaab: Resources. Jonathan Ottnad: Resources. Viktor Ratza: Software. Markus Ball: Supervision, Writing - review & editing. Bernhard Ketzer: Conceptualization, Supervision, Funding acquisition, Writing - review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

We are indebted to Rui de Oliveira and his team at CERN-EP-DT for providing us with high-quality GEM foils and other detector components. We are also grateful to the RD51 collaboration at CERN for interesting and fruitful discussions. This work is supported by the German BMBF .

References (34)

  • BergerM.

    A large ungated TPC with GEM amplification

    Nucl. Instrum. Methods A

    (2017)
  • MeinschadT.

    GEM-Based photon detector for RICH applications

    Nucl. Instrum. Methods A

    (2004)
  • VeenhofR.

    Garfield recent developments

    Nucl. Instrum. Methods A

    (1998)
  • KillenbergM.

    Modelling and measurement of charge transfer in multiple GEM structures

    Nucl. Instrum. Methods A

    (2003)
  • BozzoM.

    Design and construction of the triple GEM detector for TOTEM

    IEEE Symp. Conf. Record Nucl. Sci.

    (2004)
  • C.Lippmann.

    A continuous read-out TPC for the ALICE upgrade

    Nucl. Instrum. Methods A

    (2016)
  • SimonF.

    Development of tracking detectors with industrially produced GEM foils

    IEEE Trans. Nucl. Sci.

    (2007)
  • Cited by (14)

    • Charging-up effects for Gas Microchannel Plate detector

      2023, Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment
    • Charging-up effect and uniformity study of a single mask triple GEM detector

      2023, Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment
    • Study of charging-up effect for a single mask triple GEM detector

      2021, Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment
    View all citing articles on Scopus
    View full text