Semiconducting Ge1-xSnx alloy offers exciting possibilities for bandgap and strain engineering in a silicon compatible technology, because of its tunable bandgap [1] and possibility of high electron and hole mobilities [2]. High hole mobilities in Ge1-xSnx based pMOSFETs has already been demonstrated [3,4]. Furthermore, due to the larger size of Sn atoms, use of GeSn as source/drain in Ge pMOSFETs has been used to provide channel compressive strain to boost hole mobility [5]. For all those applications, it is mandatory to define analytical approaches able to give accurate measurements of Sn content, since this parameter can affect both the induced strain and/or the carrier mobility. In this work, SIMS characterization of ad-hoc prepared samples will be presented in order to define analytical protocols to quantify Sn content in Ge. In particular, 1.5 µm thick Ge films epitaxially grown on Si (100) were used. Beside MBE grown Ge1-xSnx films with well controlled Sn concentration (≤ 7%), some reference samples were prepared by ion implantation. Since ion irradiation of high mass species on Ge is known to induce a characteristic “honeycomb” damage structure, impossible to anneal out with conventional thermal treatments [6, 7], ion implantation was carried out at liquid nitrogen to avoid the void formation [8] and three implant doses were tested: 5x1015 at/cm2, 1x1015 at/cm2 and 5x1014 at/cm2. Implant energy was set at 45 keV and implants were carried out through an 11 nm SiNx film. SIMS characterization was carried out in different configurations, i.e. using O2+ as primary beam and collecting positive secondary ions, Cs+ and negative secondary ions and finally Cs+ collecting MCs+ ions. Results will be compared with quantitative measurements obtained by Rutherford backscattering and issues about quantification will be addressed. [1] G. He and H.A. Atwater, Phys. Rev. Lett., 79, (2007), 1937. [2] J.D. Sau and M.L. Cohen, Phys. Rev. B, 75, (2007), 045208. [3] S. Gupta et al., IEDM 2011, [4] G. Han et al., IEDM 2011, [5] B. Vincent et al., Microelectronic Eng. 88(4), (2011), 342. [6] I.H. Wilson, J. Appl. Phys. 53(3), (1982), 1698. [7] L. Romano et al., J. Appl. Phys. 107, (2010), 084314. [8] G. Impellizzeri et al, J. Appl. Phys. 106, (2009), 013518.
Dynamic SIMS Characterization of Ge1-xSnx alloy
Secchi, Maria;Demenev, Evgeny;Giubertoni, Damiano;Bersani, Massimo
2013-01-01
Abstract
Semiconducting Ge1-xSnx alloy offers exciting possibilities for bandgap and strain engineering in a silicon compatible technology, because of its tunable bandgap [1] and possibility of high electron and hole mobilities [2]. High hole mobilities in Ge1-xSnx based pMOSFETs has already been demonstrated [3,4]. Furthermore, due to the larger size of Sn atoms, use of GeSn as source/drain in Ge pMOSFETs has been used to provide channel compressive strain to boost hole mobility [5]. For all those applications, it is mandatory to define analytical approaches able to give accurate measurements of Sn content, since this parameter can affect both the induced strain and/or the carrier mobility. In this work, SIMS characterization of ad-hoc prepared samples will be presented in order to define analytical protocols to quantify Sn content in Ge. In particular, 1.5 µm thick Ge films epitaxially grown on Si (100) were used. Beside MBE grown Ge1-xSnx films with well controlled Sn concentration (≤ 7%), some reference samples were prepared by ion implantation. Since ion irradiation of high mass species on Ge is known to induce a characteristic “honeycomb” damage structure, impossible to anneal out with conventional thermal treatments [6, 7], ion implantation was carried out at liquid nitrogen to avoid the void formation [8] and three implant doses were tested: 5x1015 at/cm2, 1x1015 at/cm2 and 5x1014 at/cm2. Implant energy was set at 45 keV and implants were carried out through an 11 nm SiNx film. SIMS characterization was carried out in different configurations, i.e. using O2+ as primary beam and collecting positive secondary ions, Cs+ and negative secondary ions and finally Cs+ collecting MCs+ ions. Results will be compared with quantitative measurements obtained by Rutherford backscattering and issues about quantification will be addressed. [1] G. He and H.A. Atwater, Phys. Rev. Lett., 79, (2007), 1937. [2] J.D. Sau and M.L. Cohen, Phys. Rev. B, 75, (2007), 045208. [3] S. Gupta et al., IEDM 2011, [4] G. Han et al., IEDM 2011, [5] B. Vincent et al., Microelectronic Eng. 88(4), (2011), 342. [6] I.H. Wilson, J. Appl. Phys. 53(3), (1982), 1698. [7] L. Romano et al., J. Appl. Phys. 107, (2010), 084314. [8] G. Impellizzeri et al, J. Appl. Phys. 106, (2009), 013518.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
