S C

IE N

T IF

IC H

IG H

LI G

H T

S X

-R A

Y N

A N

O P

R O

B E

8 3 I H I G H L I G H T S 2 0 2 2

PRINCIPAL PUBLICATION AND AUTHORS

Micro-structuring to Improve the Thermal Stability of GeSn Layers; V. Bonino (a), N. Pauc (b), V. Calvo (b), M. Frauenrath (c), J.M. Hartmann (c), A. Chelnokov (c), V. Reboud (c), M. Rosenthal (a), J. Segura-Ruiz (a), ACS Appl. Mater. Interfaces 14, 19, 22270-22277 (2022); https:/doi.org/10.1021/acsami.2c01652 (a) ESRF (b) University Grenoble Alpes, CEA, Grenoble INP, IRIG, PHELIQS, Grenoble (France) (c) University Grenoble Alpes, CEA, LETI, Grenoble (France)

REFERENCES

[1] W.J. Yin et al., Phys. Rev. B - Condens. Matter Mater. Phys. 78, 161203 (2008). [2] H. Groiss et al., Sci. Rep. 7, 16114 (2017).

fabricated using out-of-equilibrium techniques are also prone to segregation at low temperatures (higher than the eutectic temperature of the Ge1-xSnx alloy, around 230°C) [2]. Therefore, the development of a technology based on this alloy requires new strategies to improve its thermal stability.

Nano X-ray fluorescence (XRF) at beamline ID16B, micro- Raman, and micro-photoluminescence (PL) spectroscopies were used to characterise blanket layers and micro- disks of Ge1-xSnx before and after annealing at different temperatures and durations. The two type of structures exhibited different behaviours after their simultaneous annealing.

Figure 72 shows the Sn distribution (Kα,β, XRF intensity) map in a blanket layer with 12.7% and 16.9% of Sn, after thermal annealing at 350°C in vacuum for 20 minutes. There is an evident Sn segregation, in contrast with the microstructures characterised, which show no Sn segregation signatures for both Sn concentrations. The high penetration, spatial resolution and sensitivity provided by the nano-XRF setup at beamline ID16B enabled to confirm that, within the spatial resolution (50 nm) and sensitivity (1.7 at.% for Sn) of the experimental setup, there was no Sn segregation in the micro-disks. Further studies performed on micro-disks annealed for one and two hours at the same temperature (350°C), showed that like the micro-disks annealed for 20 minutes there was no Sn segregation in the microstructures.

After the tolerance of micro-disks to long annealing processes at 350°C was confirmed, a new set of micro- disks with diameters ranging from 5 to 25 μm were annealed at 400°C for 20 minutes. Five micro-disks were characterised for each diameter using nano-XRF mapping. At variance with the annealing at 350°C, some micro- disks showed Sn segregation, whereas others did not. On average, larger micro-disks segregated more often than smaller ones. The Sn segregation did not depend on the shape but instead on the Sn concentration and the surface area of the microstructures. It should therefore be possible to find the maximum microstructure area needed to avoid segregation for a given Sn concentration at a certain temperature, enabling the optimisation of wafers during layer fabrication.

Figures 73a and 73b show the results of the micro-Raman and micro-PL characterisation of micro-disks P08 before and after thermal annealing at 350°C during 20 minutes and 2 hours, and at 400°C during 20 minutes. The Raman results are in good agreement with those obtained from the nano-XRF data in terms of non-segregation of Sn in the annealed micro-disks. Micro-PL results showed that the segregated micro-disks did not show any PL signal, whereas non-segregated annealed micro-disks exhibited an increased optical emission (up to 7.0x higher) compared with non-annealed ones.

In summary, this work reports on a clear improvement in the thermal stability of Ge1-xSnx alloys after the micro- structuring of thick layers with Sn concentrations of up to 1.7 at.%.

Fig. 73: a) Micro-Raman and (b) low- temperature (25 K) micro-PL spectra of Ge1-xSnx micro-disks before and after thermal annealing.