Nanoindentation of ion-implanted crystalline germanium

Most indentation studies to date on crystalline germanium (c-Ge) and related covalent semiconductors have been carried out on pristine defect-free material. This paper addresses the paucity of studies on imperfect crystalline materials by exploring the impact of defects generated by ion implantation, prior to contact damage, upon the mechanical properties of c-Ge. Implantation with Ge ions is carried out to generate a layer of highly defective but still-crystalline Ge. Under nanoindentation with a sharp diamond tip, enhanced plasticity is observed relative to pristine material. Characterization by cross-sectional transmission electron microscopy, atomic force microscopy, and load curve analysis shows softening, quasiductile extrusion, and cracking suppression taking place. These changes can be explained by the high density of defects, and dangling bonds in particular, created by ion implantation and revealed by positron-annihilation spectroscopy, and are proportional to the fraction of “missing bonds” or vacancies in the material. A thermal annealing step at $200°\text{C}$ is sufficient to restore the mechanical response of pristine material, despite incomplete recovery of the original pristine crystal structure.

Figure 1

(a) Vacancies produced as a function of depth for $3\times {10}^{13}\text{ions}\cdot {\text{cm}}^{-2}$ Ge ions implanted into Ge at 800 keV, taken from SRIM simulations. The real vacancy concentration may be lower due to dynamic annealing. (b) Bright-field (BF) XTEM of implanted layer for $1\times {10}^{13}\text{ions}\cdot {\text{cm}}^{-2}$ fluence. (c) BF XTEM of implanted layer for $3\times {10}^{13}\text{ions}\cdot {\text{cm}}^{-2}$ fluence, with selected area diffraction pattern (SADP) from implanted layer inset.

Figure 2

(Color online) Indentation hardness versus implanted fluence, for as-implanted Ge specimens and for specimens postannealed at $200°\text{C}$. Dotted line: predicted hardness based on vacancy dependence [Eq. (5)].

Figure 3

(Color online) Hardness versus postimplant annealing temperature for Ge specimens implanted to a fluence of $3\times {10}^{13}\text{ions}\cdot {\text{cm}}^{-2}$.

Figure 4

(Color online) Young’s modulus versus implanted fluence, for as-implanted specimens and specimens annealed at $200°\text{C}$. Dotted line: predicted modulus from vacancy fraction [Eq. (6)].

Figure 5

AFM micrographs of indents to 100 mN in (a) Ge implanted to $3\times {10}^{13}\text{ions}\cdot {\text{cm}}^{-2}$ (unannealed), $7\times 7\mu \text{m}$ image, and (b) unimplanted Ge, $10\times 10\mu \text{m}$ image.

Figure 6

Raman spectra collected from unimplanted Ge, as-implanted $3\times {10}^{13}\text{ions}\cdot {\text{cm}}^{-2}$ specimen, and an indent to 100 mN in the as-implanted $3\times {10}^{13}\text{ions}\cdot {\text{cm}}^{-2}$ specimen.

Figure 7

(Color online) PAS results: (a) Doppler broadening $S$ parameter as a function of incident positron energy for pristine, as-implanted, and $200°\text{C}$-annealed specimen. (b) $S$ parameter versus positron energy for $3\times {10}^{13}\text{ions}\cdot {\text{cm}}^{-2}$ implanted specimen annealed at a range of temperatures. Estimated average penetration depth is shown on the upper axis.

Figure 8

(a) BF XTEM micrograph showing indent to 100 mN in $3\times {10}^{13}\text{ions}\cdot {\text{cm}}^{-2}$ as-implanted specimen, (b) SADP from deformed layer, (c) SADP from implanted layer outside deformed region, (d) SADP from pristine crystal.

Figure 9

(a) BF XTEM of indent to 100 mN in $3\times {10}^{13}\text{ions}\cdot {\text{cm}}^{-2}$ $200°\text{C}$-annealed specimen, (b) magnified view of microcrack within deformed region.