Abstract text (incl. figure legends and references)

Most of today"s electronic devices, like solar cells or transistors, are based on electric fields at the junctions of doped semiconductors. As electronic devices keep shrinking, identifying electric fields on an increasingly smaller scale is an important part of the optimization of these devices. Two possible options to map electric fields with a high spatial resolution are electron beam induced current (EBIC) in a (S)TEM and off-axis electron holography (EH). In EBIC, the current induced by the electron beam scanning over the specimen is measured, while EH detects a shift to the electron phase by an electrostatic potential.

Although EH has been studied intensively over the past decades, it still faces several challenges, hindering quantitative determination of electrostatic potentials. Especially the changes of the properties of an effectively two-dimensional p-n junction when reducing a sample to TEM specimen still pose several problems [1]. Previous works have shown that electric stray fields outside the specimen have an influence on the measurement of the build-in potential in off-axis electron holography [2]. EBIC has, in the past, been used for the mapping of electric fields in SEM. Although, by performing EBIC in a (S)TEM (STEBIC), a higher spatial resolution is achieved [3], it is not yet a common practice and has only recently gained traction due to advancements in the development of biasing holders and STEM units. Now, STEBIC faces similar challenges that EH had to overcome, while others are still relevant for both methods.

In this work, knowledge gained through EH is used in order to deepen the understanding of collected charge carriers induced by the electron beam in such an effectively two-dimensional p-n junction. As a first step of unifying both methods, STEBIC and EH experiments are performed under the same conditions (i.e. in the same microscope). A symmetrically doped p-n junction in silicon is investigated by both EH and STEBIC (Figure 1). The negative EBIC current measured by STEBIC is proportional to the electric fields in a silicon p-n junction, while the phase shift (Φ) measured in EH is proportional to the electrostatic potential. Thus, its derivation d/dy Φ is comparable to the STEBIC signal. By setting both signals side to side, the STEBIC signal, contrary to EH, mimics the symmetry of the junction"s doping concentration (Figure 2). Under reverse bias (-2 V), the expected increase of the electric field at the junction can be observed. Based on the similar behavior that both methods exhibit under bias, we conduct further exploration of the methods with focus on the width of the recorded profiles. Through finding similarities and differences between the two methods, a pathway to quantitative investigations of electric fields in TEM seems feasible.

[1] Ankur Nipane et al., J. Appl. Phys. 122, 194501 (2017)
[2] T. Wagner, PhD-Thesis, TU Berlin (2022)
[3] Aidan P. Conlan et al., J. Appl. Phys. 129, 135701 (2021)

Fig. 1: Overview of the symmetric p-n junction in silicon. Exemplary EBIC, phase and d/dy phase images from the marked areas.

Fig. 2: Electric field in a pn-junction in terms of negative EBIC current from STEBIC and d/dy phase profiles from EH according to marked areas in Fig. 1.