With further scaling of complementary metal-oxide semiconductor (CMOS) devices, Si3N4 has attracted much attention in microelectronic related studies due to its excellent mechanical, thermal, and electronic properties [1]. These make Si3N4 a widely used material in CMOS devices to prevent diffusion of impurities, as passivation layer, as optical layer for phase-shift mask and as a replacement for SiO2 as high K gate dielectric, especially for nonvolatile memory applications. However, the interface of the nitride with other elements in the electronic devices is not abrupt, being the nitride buffer layer amorphous and source of defects. Nonetheless, the thermally grown nitride shows the remarkable property to grow epitaxially on the Si(111) face, owing to a negligible lattice mismatch (<1.2% between the ?- Si3N4 (0001) face and the 2x2 cell of the Si (111) face) [2,3]. Although much effort has been devoted to the study of the ?- structural phase of the silicon nitride, a full understanding of the 8x8 atomic surface reconstruction [3,5] and of the chemical bonds at the interface [2, 4, 6] has not been achieved. This is of paramount importance as the presence of surface or interface dangling bonds (DBs) induces gap states across the Fermi level, creating tunnelling channels which can reduce the performance of electronic devices. None of the experimental techniques used so far have been able to suggest which of the models better describes the system. Even STM measurements didn't succeed in settling the controversy on the 8x8 reconstruction [3,5]. Here we report measurements carried out by ARPES (Angle Resolved Photo Emission Spectroscopy), in collaboration with the research group of the VUV beamline at Elettra. Patterns showing the electronic structure of the system along the high symmetry directions, as well as constant energy cuts have been recorded. Our results show the presence of a new band with a surface character likely attributed to the topmost atoms of the nitride layer. [1] V.I. Belyi et al. Silicon nitride in electronics, Vol. 34, Materials Science Monographs, Elsevier, Amsterdam (1988) [2] J. W. Kim et al PRB 67 (2003) 035304 [3] X.-S Wang et al. Surf Sci 494 (2001) 83, PRB 60 (2000) R2146 [4] Gwo et al. PRL 90 (2003) 185506; R. Flammini et al. Surf. Sci. 579, (2005) 188, J. Appl. Phys. 103, (2008) 083528, Surf. Sci. 606 (2012) 1215, K. Eguchi PRB 85 (2012) 174415 [5] H. Ahn et al. Phys. Rev. Lett. 86, (2001) 2818, C. L. Wu et al. PRB 65 (2002) 045309 [6] M. Yang et al. J. Appl. Phys. 105 (2009) 024108
Probing the electronic structure of the /Si(111)/?-Si3N4(0001)-(8x8) system by ARPES
R Flammini
2013
Abstract
With further scaling of complementary metal-oxide semiconductor (CMOS) devices, Si3N4 has attracted much attention in microelectronic related studies due to its excellent mechanical, thermal, and electronic properties [1]. These make Si3N4 a widely used material in CMOS devices to prevent diffusion of impurities, as passivation layer, as optical layer for phase-shift mask and as a replacement for SiO2 as high K gate dielectric, especially for nonvolatile memory applications. However, the interface of the nitride with other elements in the electronic devices is not abrupt, being the nitride buffer layer amorphous and source of defects. Nonetheless, the thermally grown nitride shows the remarkable property to grow epitaxially on the Si(111) face, owing to a negligible lattice mismatch (<1.2% between the ?- Si3N4 (0001) face and the 2x2 cell of the Si (111) face) [2,3]. Although much effort has been devoted to the study of the ?- structural phase of the silicon nitride, a full understanding of the 8x8 atomic surface reconstruction [3,5] and of the chemical bonds at the interface [2, 4, 6] has not been achieved. This is of paramount importance as the presence of surface or interface dangling bonds (DBs) induces gap states across the Fermi level, creating tunnelling channels which can reduce the performance of electronic devices. None of the experimental techniques used so far have been able to suggest which of the models better describes the system. Even STM measurements didn't succeed in settling the controversy on the 8x8 reconstruction [3,5]. Here we report measurements carried out by ARPES (Angle Resolved Photo Emission Spectroscopy), in collaboration with the research group of the VUV beamline at Elettra. Patterns showing the electronic structure of the system along the high symmetry directions, as well as constant energy cuts have been recorded. Our results show the presence of a new band with a surface character likely attributed to the topmost atoms of the nitride layer. [1] V.I. Belyi et al. Silicon nitride in electronics, Vol. 34, Materials Science Monographs, Elsevier, Amsterdam (1988) [2] J. W. Kim et al PRB 67 (2003) 035304 [3] X.-S Wang et al. Surf Sci 494 (2001) 83, PRB 60 (2000) R2146 [4] Gwo et al. PRL 90 (2003) 185506; R. Flammini et al. Surf. Sci. 579, (2005) 188, J. Appl. Phys. 103, (2008) 083528, Surf. Sci. 606 (2012) 1215, K. Eguchi PRB 85 (2012) 174415 [5] H. Ahn et al. Phys. Rev. Lett. 86, (2001) 2818, C. L. Wu et al. PRB 65 (2002) 045309 [6] M. Yang et al. J. Appl. Phys. 105 (2009) 024108I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
