Elsevier

Ultramicroscopy

Volume 110, Issue 1, December 2009, Pages 33-35
Ultramicroscopy

Medipix 2 detector applied to low energy electron microscopy

https://doi.org/10.1016/j.ultramic.2009.09.002Get rights and content

Abstract

Low energy electron microscopy (LEEM) and photo-emission electron microscopy (PEEM) traditionally use microchannel plates (MCPs), a phosphor screen and a CCD-camera to record images and diffraction patterns. In recent years, however, MCPs have become a limiting factor for these types of microscopy. Here, we report on a successful test series using a solid state hybrid pixel detector, Medipix 2, in LEEM and PEEM. Medipix 2 is a background-free detector with an infinite dynamic range, making it very promising for both real-space imaging and spectroscopy. We demonstrate a significant enhancement of both image contrast and resolution, as compared to MCPs. Since aging of the Medipix 2 detector is negligible for the electron energies used in LEEM/PEEM, we expect Medipix to become the detector of choice for a new generation of systems.

Introduction

Low energy electron microscopy (LEEM) [1], [2] and photo-emission electron microscopy (PEEM) [3] rely on MCPs [4] to convert electrons, backscattered or photo-emitted from a sample, into an image. Since LEEM came to prominence in 1984 [5], MCP units have been the detectors of choice. Unfortunately, the performance of an MCP detector unit is not ideal, so that it has become a limiting factor in the dynamic range of LEEM. A serious drawback is the limited signal intensity that MCPs can handle at their input. Using a 12-bit CCD camera for image recording with MCPs near saturation, a typical image exposure time of 600 ms is needed to utilize the full 12 bit resolution of the imaging system to record a 512×512 pixel image [6]. Unfortunately, frequent overexposure of the MCPs will inevitably lead to beam damage of the MCPs; an effect all too familiar to LEEM operators. Another disadvantage of MCPs are the various sources of noise, such as dark current, thermal noise and the readout noise of the CCD camera used to image the MCPs. The latter is typically the dominant source of noise in LEEM images and constitutes approximately 50 of the available 212 grey levels, depending on the particulars of the camera and its operating parameters. For MCPs the dynamic range is limited by dark counts for very long exposure times (100s). Finally, the spatial resolution of the standard LEEM/PEEM MCP detector is rather modest, with a measured resolution limit of 3.5 linepairs/mm or 140μm [7], [8]. Medipix 2 has the potential of improving on all of these issues [9], [10], [11]. It offers a much improved resolution (9 linepairs/mm or 55μm), a large dynamic range and virtually no background. Furthermore, aging effects were shown to be negligible at the electron energies used in LEEM/PEEM (10–20 keV) [12], [13], [14]. In this Letter, we demonstrate that the use of a Medipix 2 detector significantly improves the performance of LEEM and PEEM.

Medipix is a CMOS ASIC chip bump-bonded to a high-resistivity silicon sensor. The sensor consists of an array of 256×256 pixels with 55μm side lengths. Each pixel operates as a back-biased diode with the p-implanted region near the bump-bond when sufficient bias voltage is applied for full depletion of the silicon sensor. Each individual pixel on the chip contains a signal processing and counting circuit. Background rejection and signal amplification can therefore be performed independently for each pixel by setting a lower and higher threshold. Events are recorded in a 13-bit counter at a maximum rate greater than 100 kHz/pixel. An MCP detector records images at a typical CCD resolution of 512×512 pixels for a 42 mm diameter circular area, and a 12-bit CCD counter. Under practical conditions its maximum count rate is less than 1 kHz/pixel [4]. All in all, Medipix allows for very high count rates, virtually no background, and a superior dynamic range. With the capabilities of LEEM and PEEM extending more and more, particularly into the area of spectroscopic imaging, the requirement for a detector with a large dynamic range and a low background has become urgent.

Section snippets

Experimental details

In this study, a single Medipix 2 detector (14×14mm) was tested in the Elmitec LEEM III instrument at the University of Twente. Measurements were done on an Ir(111) sample with graphene flakes grown on its surface [15]. For this, an Ir(111) sample was first heated to 850C and exposed to 1×10-7mbar of O2 to remove residual carbon contamination. Graphene flakes were subsequently grown at a 5×10-8 mbar pressure of ethylene at 900C. The growth was stopped when the coverage of graphene was around

Results and discussion

In Fig. 1, we compare PEEM and LEEM images taken with both detectors. We note that in PEEM, the Ir appears dark, since its work function (5.3 eV) is higher than the photon energy (4.8 eV). In LEEM the contrast depends on the electron energy at the sample. From the images, it is immediately clear that Medipix 2 yields superior resolution and contrast. This is evidenced by the appearance of a bright line at the edges of the graphene flakes in the Medipix PEEM-image (Fig. 1(b)). For the MCPs, this

Conclusions

In summary, we have tested a Medipix 2 chip for the detection of electrons with energies up to 20 keV in LEEM/PEEM. We obtain a much better spatial resolution and contrast than for a more conventional, MCP-based detector. Moreover, through appropriate setting of the thresholds, background-free images can be recorded. We expect Medipix to become the standard detector for LEEM and PEEM, because of its many advantages. Clearly, its high resolution and contrast are very beneficial for real space

Acknowledgments

This research was funded by the Netherlands Organization for Scientific Research (NWO) via an NWO-Groot grant (‘ESCHER’) as well as by the Dutch Foundation for technical sciences (STW). We gratefully acknowledge Ruud van Egmond, Emiel Wiegers and Ewie de Kuyper for technical support. We thank Jan van Ruitenbeek and Wasi Faruqi for inspiring discussions. We are grateful to the CERN and NIKHEF Medipix groups for their advice and support.

References (22)

  • W. Telieps et al.

    Ultramicroscopy

    (1985)
  • G. Moldovan et al.

    Nucl. Instrum. Methods A

    (2008)
  • A.R. Faruqi

    Nucl. Instrum. Methods A

    (2001)
  • A.R. Faruqi et al.

    Nucl. Instrum. Methods A

    (2005)
  • G. McMullan et al.

    Ultramicroscopy

    (2007)
  • T. Holy et al.

    Nucl. Instrum. Methods A

    (2006)
  • Z. Vykydal et al.

    Nucl. Instrum. Methods A

    (2006)
  • Z. Vykydal et al.

    Nucl. Instrum. Methods A

    (2008)
  • E. Bauer

    Surf. Sci.

    (1994)
  • R.M. Tromp

    IBM J. Res. Dev.

    (2000)
  • E. Bauer

    J. Electron Spectrosc.

    (2001)
  • Cited by (42)

    • Standard deviation of microscopy images used as indicator for growth stages

      2022, Ultramicroscopy
      Citation Excerpt :

      In the case of electron microscopy, cameras based on CCD, EMCCD and sCMOS technology are widely used in combination with multichannel plates and fluorescent screens. In addition, more exotic electron detectors as MediPix 2 [5,6], EIGER [7] and just recently fiber coupled CMOS sensors [8] are employed. The various detectors differ regarding their dynamic range, signal to noise ratio, and uniformity of pixel gain and offset.

    • Principles of Electron Optics, Volume 3: Fundamental Wave Optics

      2022, Principles of Electron Optics, Volume 3: Fundamental Wave Optics
    View all citing articles on Scopus
    1

    Both authors contributed equally to this work.

    View full text