TU Berlin

AG LehmannInstrumentation and Methods

Inhalt des Dokuments

zur Navigation

Instrumentation and Methods

In our research, the development of instrumentation and methods is closely linked with each other, providing also for students a wide field of exciting research. We are currently working mainly on the following exciting topics:

Off-axis Electron Holography

Electron-optical setup for off-axis electron holography in a TEM

Conventional TEM is hampered in its possibilities by the fact that in an electron micrograph only the squared amplitude of the electron wave is recorded, while their phase information in lost. According to Gabor's idea, off-axis electron holography in a TEM overcomes the phase problem of imaging by recording an object-modulated interferogram, the so-called electron hologram. By numerical processing, the quantitative information of the electron wave is reconstructed and, in the case of atomic resolution, subsequently corrected for residual aberrations resulting in amplitude and phase of the object exit-wave. Since the object phase information is retrieved without transfer gaps, a wide unique field opens of both methodical developments and holographic applications in solid state physics, materials science, and chemistry.

Time-Resolved Electron Holography

Blue line: the signal applied to the capacitor as measured by the oscilloscope (5 periods are plotted upon each other). Light blue bars: time intervals and temporally averaged slope obtained from the individually reconstructed holograms.

Transmission electron microscopy (TEM) is a key method for the understanding of materials on the nanoscale, as it grants access to structural or atomic causes of macroscopic observations. Electron holography (EH) even deepens this microscopic insight, as it directly measures the associated electric and magnetic fields with the same spatial resolution (compare e.g. [1]). So far, EH is limited to static specimen. The extension to dynamic processes is challenging, as the time-resolution of EH is limited by the long exposure times (typically in the seconds range), which are needed to obtain data well above noise.

For reversible processes, this hurdle usually is overcome by pump-probe techniques. Pump-probe measurements of periodic processes require a temporal gating for the time-dependent signal. In TEM the required shutter/gating is commonly realized either by the use of fast direct detection cameras with readout times per frame approaching the submillisecond regime, or by blanking the beam away from the detector by transversal electrical or magnetic fields. For acceleration voltages in the 200 - 300 kV range, the transients occurring in switching the required high currents or voltages forbid time resolutions below millisecond scales.

We developed a simple, yet promising approach for temporal gating by exploiting the high sensitivity of interference techniques like electron holography to controlled instrumental instabilities [2]. In an off-axis EH setup small instabilities are easily introduced by a mutual phase shift between both partial waves [3]. As a proof-of-concept experiment, which basically only required the extra equipment of a PC sound-card as D/A converter and some batteries as constant voltage source for the electron optical biprism, we realized time-resolved electron holography in an unmodified TEM with continuous illumination allowing the measurement of periodically changing potential variations in the sample with time resolution in the microseconds regime.

[1] G. Pozzi, M. Beleggia, T. Kasama, R.E. Dunin-Borkowski, Comptes Rendus Phys. 15 (2014) 126. 

[2] T. Niermann, M. Lehmann, T.Wagner,  Ultramicroscopy 182 (2017) 54–61.

[3] T. Wagner, T. Niermann, D. Berger, M. Lehmann, Proceedings EMC 2016.

Optics for Electron Holography

Titan TEM with two biprisms.

By means of a Möllenstedt biprism, which is in fact only a minor supplement to a FEG-based TEM, reference wave and the object modulated exit-wave are brought to an overlap, where on the detector an interference pattern is formed, the so-called off-axis electron hologram. Based on the fundamental considerations by Lichte [1] and Harada [2], our specially designed FEI Titan 80-300 Holography Special Berlin TEM with two biprisms after the image Cs-corrector offers a higher experimental flexibility for adjusting width and fringe spacing of holograms [3]. In this double biprism setup using the first biprism for a shadow, in which the second biprism is located nearby its optimum position producing the overlapping of reference and image wave, atomically resolved holograms with high fringe contrast at spacings of about 35 pm are formed. In combination with improved instrumental stability, less Fresnel diffraction, and reduced vignetting effect, a full reconstruction of a GaN crystal's object exit-wave up the information limit of 75 pm of the instrument without transfer gaps has already been demonstrated [4]. [5]

[1] H. Lichte, Ultramicroscopy 64 (1996) 79.
[2] K. Harada et al., Appl. Phys. Lett. 84 (1994) 3229.
[3] F. Genz et al., Ultramicroscopy 147 (2014) 33.
[4] T. Niermann and M. Lehmann, 63 (2014) 28.
[5] The financial support through DFG INST 131/508-1 FUGG is gratefully acknowledged.

Hologram Acquisition and Statistical Evaluation

Atomically resolved electron hologram of GaN with its reconstruction in amplitude and phase as well as comparison with corresponding simulation.

By acquisition of series of electron holograms with 20 or more holograms and subsequent reconstruction considering the drifts of specimen and biprism over the series, an image wave with high signal-to-noise ratio can be retrieved beneficial for further evaluation both at medium and atomic resolution [1]. On the atomic scale by quantitative comparison of the full experimentally gained image waves with wave functions as calculated within simulations modeling the electron-object interaction and the wave propagation by the objective lens, a whole set of important specimen and imaging parameters are obtained by least square fitting: Besides specimen thickness, tilt and strain, imaging parameters like two-fold astigmatism, defocus and higher orders are extracted, which in turn are used for correction of residual aberrations. Consequently, the experimental uncertainties due to imprecise knowledge of imaging parameters are strongly reduced allowing the evaluation of the object structure with high precision. [2]

[1] T. Niermann and M. Lehmann, 63 (2014) 28.
[2] The financial support through DFG CRC 787 project A4 is gratefully acknowledged.

Potentiometry in TEM and SEM

SEM image (left) and phase image (right) of GaN p-n junction prepared as needle showing p-n junction in the middle of the needle.

Electron holography in a TEM is often used for measuring the built-in voltage of p-n junctions in doped semiconductors. The lateral distribution of p-n junctions can in general easily be imaged. However, the measured built-in voltage is always smaller than the expected one as calculated from the dopant levels. It is reported that at least for Silicon providing a conductive path to ground by coating with Carbon the discrepancy is strongly reduced. Up to now, an effect only partially considered is the influence of the electron beam illumination producing electron hole pairs so that the p-n junction acts like a solar-cell illuminated in the TEM instead of light with electrons. Additionally, the high-energy electron beam leads to the generation of secondary electrons as observed as positive charging of samples, which are badly connected to ground. Using needle-shaped GaN p-n junctions prepared by FIB, careful holographic experiments by reducing the electron dose rate over three orders of magnitude, but acquiring hologram series with an accumulated exposure time up to 1000 s enabling low dose rate electron holography, have shown that indeed electron-hole pair generation plays a significant role in explaining the discrepancy whereas the generation of secondary electrons can be neglected since they do not produce a net current over the p-n junction. Of overall importance for defined experimental conditions is a small shunt, which in the Silicon system can easily be obtained by Carbon coating, whereas in the GaN system a much larger interface resistance between bulk and conducting surface layers must exist [1]. These findings are important for further measurements assuring a good contact of the bulk sample with ground of the TEM, and its opens the window for quantitative in-situ measurements under biasing. [2]

Most recently, we have extended our research on potentiometry of nanostructured semiconductors to imaging and measurements in a scanning electron microscope (SEM). Using primary electrons with energies of around 1 keV, the potential differences can in principle be measured. A true challenge, however, is the proper surface conditions of the sample under investigation. [3]

[1] Jae Bum Park et al.,  Appl. Phys. Lett. 105 (2014) 094102.
[2] The financial support through DFG CRC 787 project A4 is gratefully acknowledged.
[3] The financial support through DFG INST 131/631-1 FUGG is gratefully acknowledged.



Schnellnavigation zur Seite über Nummerneingabe