- Email: [email protected]
3D site specific sample preparation and analysis of 3D devices (FinFETs) by atom probe tomography
Ultramicroscopy 132 (2013) 65–69
Contents lists available at ScienceDirect
Ultramicroscopy journal homepage: www.elsevier.com/locate/ultramic
3D site specific sample preparation and analysis of 3D devices (FinFETs) by atom probe tomography Ajay Kumar Kambham a,b,n, Arul Kumar a,b, Matthieu Gilbert b, Wilfried Vandervorst a,b a b
KU Leuven, Department of Physics and Astronomy, B-3001 Leuven, Belgium IMEC, Kapeldreef 75, B-3001 Leuven, Belgium
a r t i c l e i n f o
abstract
Available online 24 February 2013
With the transition from planar to three-dimensional device architectures such as Fin field-effecttransistors (FinFETs), new metrology approaches are required to meet the needs of semiconductor technology. It is important to characterize the 3D-dopant distributions precisely as their extent, positioning relative to gate edges and absolute concentration determine the device performance in great detail. At present the atom probe has shown its ability to analyze dopant distributions in semiconductor and thin insulating materials with sub-nm 3D-resolution and good dopant sensitivity. However, so far most reports have dealt with planar devices or restricted the measurements to 2D test structures which represent only limited challenges in terms of localization and site specific sample preparation. In this paper we will discuss the methodology to extract the dopant distribution from real 3D-devices such as a 3D-FinFET device, requiring the sample preparation to be carried out at a site specific location with a positioning accuracy 50 nm. & 2013 Elsevier B.V. All rights reserved.
Keywords: Atom probe tomography FinFET Dopant conformality Focused ion beam
1. Introduction The introduction of FinFETs for the (sub)-22 nm [1] node brings a major challenge for the doping profile formation and optimization. To optimize the 3D-doping profiles in FinFET-based devices, one can use ion implantation [2], vapor phase doping (VPD) [3] or plasma doping [4] which all have distinct properties in terms of doping incorporation, conformality and process integration. In each case a crucial step in optimizing these doping technologies is the ability to assess their final doping profile characteristics which thus requires the availability of adequate 3D-dopant profiling techniques which can probe directly on devices with sub-nm resolution. In this work we have developed a methodology based on the atom probe tomography (APT) [5–8] to probe the 3D-dopant distribution inside a FinFET device and extract the important parameters such as gate overlap and profile gradient. The FinFET devices used in this work originate from a standard process flow whereby the doping step is based on self regulatory plasma doping (SRPD) [4].
2. Device fabrication and sample preparation: The results were obtained on test structures with the Fin height (HFin) targeted at 60 nm, the Fin width (WFin) 40 nm n Corresponding author at: KU Leuven, Department of Physics and Astronomy, B-3001 Leuven, Belgium. Tel.: þ32 16 28 12 07; fax: þ 32 16 28 15 15. E-mail addresses: [email protected], [email protected] (A.K. Kambham).
0304-3991/$ - see front matter & 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ultramic.2012.09.013
and gate width (Wgate) 50 nm. Fig. 1 shows the SEM image of the 3D structures after gate patterning before plasma doping and spike annealing. After the doping step the structures are covered with a suitable cap layer preferably polysilicon 100 nm in order to reduce the impact of the major insulating materials around the Fins on the APT analysis and the damage of the FIB ion beam during APT sample preparation. As the field of view in APT is fairly limited, a crucial step is to position the circled location (shown in Fig. 1) in the area of APT-tip which contains the relevant information on conformality, gate overlap and 3D-dopant distribution. This positioning is aggravated by the fact that the cap layer removes most of the topographical information which was visible in Fig. 1.
2.1. Localization and sample preparation by FIB There are many reports [9–14] to solve the sample preparation issues targeting to place the area of interest (AOI) in the final sample/ tip for 1D and 2D structures using FIB system for atom probe studies. Whereas for that case the positioning is facilitated by the fact that at least one dimension is unrestricted or extends over several microns, the challenge for the present problem is the 3D confinement of the AOI to volume in the order of (50 40 60 nm3). Moreover, after depositing the polysilicon and chemical mechanical polishing (CMP) very little SEM-contrast is available from the top to localize the AOI. In order to localize and study these 3D confined areas using APT, we need to use and reference marks/cuts by FIB during the lamella preparation and annular milling. The only reference
66
A.K. Kambham et al. / Ultramicroscopy 132 (2013) 65–69
which is available after the test structure fabrication is the presence of a pad oxide ( 80 nm above the Fins) used to protect the gate during the source/drain doping [15]. We will use the
Fig. 1. SEM image of 3D test structures prior to polysilicon deposition, circled location is the area of the interest for atom probe analysis.
topography of the pad oxide to locate and identify the Fin. As a first approach we tried to deposit platinum (for a reference) at low ion beam energies 5 kV on top of the AOI in an area of 50 50 nm2. Unfortunately the concurrent energetic ion impact leads also to rapid erosion counteracting the marking through platinum deposition (Fig. 2a). The second problem is the need to mark the AOI by a FIB cut. The distance between two FIB markings is also important when we tried to place FIB marks close to the AOI ( o500 nm) the resultant marks are too large which are getting overlap, encroach and damage the AOI as shown in Fig. 2b. We solve these problems by depositing e-beam platinum on one of the Fins using the distance to the pad oxide as a reference and make FIB line cuts at a distance 4700 nm from AOI at low ion beam energies 5–8 keV. Initiating the shank preparation, Fig. 3 shows the process steps used to make the lamella with a length and width of 12 2.5 mm2 based on the lift-out method [11]. The annular milling of the shank will start by keeping the ebeam platinum mark in the center at 30 kV ion beam energies. After removing sufficient material around the AOI, we can see some contrast of the gate stack (in SEM) below the pad oxide; by using that contrast we can position the X and Y position of the shank to finalize the tip. The final cleaning and sharpening of the tip is performed at low ion beam energies (5–2 kV) [16]. The cleaning includes the removal of the pad oxide, as its presence causes tip rupture. Fig. 4 shows the tip evolution during the annular milling to get the final tip. The success rate of this sample preparation depends on the accuracy of the positioning of the ebeam platinum deposition. A small shift in the e-beam platinum position will end-up making the tip in the trench region instead of at the gate/source drain edge.
Fig. 2. SEM images of ion beam damages (a) though we ask for Pt deposition on top of the AOI, the ion beam is milled out the AOI as shown in circle. (b) Encroachment of FIB marks inside the AOI when you try to mark them at o 500 nm from AOI (beam energy is 30 kV).
Fig. 3. SEM images of process sequence used to localize the center of area of the interest for APT sample preparation: (a) image of topography presented by pad oxide above the Fins after e-beam platinum deposition on top of AOI; (b) FIB markings close to the AOI at low ion energies; and (c) final lamella with e-beam platinum dot on one of the Fins with platinum filled FIB cults.
A.K. Kambham et al. / Ultramicroscopy 132 (2013) 65–69
67
Fig. 4. SEM images of tip evolution during annular milling: (a) e-beam Pt mark is used to position center for milling; (b) the identification pad oxide at one stage of annular milling; (c) after milling at 16 kV ion beam energy by keeping AOI at the center; and (d) final cleaning of pad oxide and shaping of the tip at 5 kV and 2 kV, respectively.
Fig. 5. APT analysis of As dopants at the gate and S/D interface (orange) and gate stack composed of HfO2 (black) and TiN (pink). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 6. APT analysis of arsenic doping and diffusion from S/D region to gate (see the schematic). Silicon atoms are not shown, orange—As, black—HfO2, pink—TiN. The white dotted line (right) shows the under diffused dopant in the channel region below the gate. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
A.K. Kambham et al. / Ultramicroscopy 132 (2013) 65–69
Arsenic Concentration (cm-3)
68
Annealed Non- Annealed
1E21
Source / Drain
GATE
1E20
1E19
0
10
20
30
40
50
Depth (nm) Fig. 9. Comparison of gate overlap dopant profiles from S/D region to channel (below the gate) region for annealed and non-annealed samples.
Fig. 7. APT image of arsenic (orange) distribution with native oxide (SiO2, green) at top and side wall of the Fin in S/D region. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Atomic %
Lateral Profile Top Profile
1
5E20
0.1
5E19
0.01 0
5
10
15
20
25
30
35
Concentration (at./Cm3)
5E21
10
5E18 40
Depth (nm) Fig. 8. Comparison of lateral and vertical dopant profiles extracted at lateral and top of the Fin in source/drain region.
3. Results and discussion The APT-analysis was performed using a laser (400 fs, 515 nm) assisted wide angle atom probe (LAWATAP) from CAMECA at 80 K temperature. Tap3D data reconstruction software provided by CAMECA has used to reconstruct and analyze the data. Figs. 5–7 illustrate that we are able to probe a complete 3D-Fin transistor as highlighted by the 3D-distribution of atoms from the 3D-gate stack to S/D region. Fig. 5 shows the 2D atomic map (As, TiN, Hf) with the As-dopant distribution (orange) for a slice taken at the gate edge and S/D region. Fig. 6 shows a 3D-cross sectional view of half the Fin volume (40 20 60 nm3, see the inset) with 3D-dopant distribution of under diffused dopants from S/D to gate. Fig. 6 provides the
required information about the variations in dopant distributions, in particular the gate under lap (under diffusion from S/D region to gate) from the top surface ( 14 nm) to the bottom along the side walls ( 5 nm). Fig. 7 shows the 2D-map of the As-distribution (orange) atoms in the S/D region. Fig. 8 illustrates the comparison between the dopant profiles extracted in the center of the gate (top profile) and from one side of the Fin to the other side (lateral profile). The top dopant profile has a much higher dopant peak concentration and dose than the lateral dopant profile, which leads to the non-corformality of the dopant distribution inside the Fin. The higher doping at the top of the Fin actually induces now a larger under diffusion and thus increased the gate overlap (Fig. 6). The availability of this metrology enables us now to study the diffusion processes operative within the FinFET process. This is exemplified in Fig. 9 where we show for the unannealed and annealed sample the increase in dopant gate overlap following the anneal step as extracted from the source/drain region to the inside of the gate. Profiles in Fig. 9 are extracted at the top within the middle of the Fin width shown an increase in under diffusion (relative to the gate edge as defined by the TiN signal) from 4 nm (non-annealed) to 14 nm for annealed case.
4. Conclusion In this study, we have described a methodology to study 3D dopant distributions in 3D devices like FinFETs using the APT technique, including the site specific sample preparation. The results demonstrate the ability to provide detailed information on the 3D-dopant distributions and gate overlap throughout all regions in the Fin transistor.
Acknowledgments IMEC acknowledges the collaboration with Cameca on the LAWATAP system. References [1] Intel 22 nm 3-D Tri-Gate Transistor Technology. [2] R. Duffy, G. Curatola, B.J. Pawlak, G. Doornbos, K. van der Tak, P. Breimer, J.G.M. van Berkum, F. Roozeboom, Doping fin field-effect transistor sidewalls: impurity dose retention in silicon due to high angle incident ion implants and the impact on device performance, Journal of Vacuum Science & Technology B: Microelectronics and Nanometer Structures 26 (1) (2008) 402.
A.K. Kambham et al. / Ultramicroscopy 132 (2013) 65–69
[3] S. Takeuchi, N.D. Nguyen, F.E. Leys, R. Loo, T. Conard, W. Vandervorst, M. Caymax, Vapor phase doping with N-type dopant into silicon by atmospheric pressure chemical vapor deposition, ECS Transactions 16 (2008) 495–502. [4] Y. Sasaki, K. Okashita, B. Mizuo, M. Kubota, M. Ogura, O. Nishijima, Conformal doping mechanism for fin field effect transistors by self-regulatory plasma doping with AsH3 plasma diluted with He, Journal of Applied Physics 111 (1) (2012) 013712. [5] R.G. Forbes, Field evaporation theory: a review of basic ideas, Applied Surface Science 87–88 (1995) 1–11. [6] G.L. Kellogg, T.T. Tsong, Pulsed-laser atom-probe field-ion microscopy, Journal of Applied Physics 51 (2) (1980) 1184. [7] T.F. Kelly, M.K. Miller, Invited review article: atom probe tomography, The Review of Scientific Instruments 78 (3) (2007) 031101. [8] M.K. Miller, A. Cerezo, M.G. Hetherington, G.D.W. Smith, Atom Probe Field Ion Microscopy, Oxford University Press, 1996. [9] K. Thompson, D. Lawrence, D.J. Larson, J.D. Olson, T.F. Kelly, B. Gorman, In situ site-specific specimen preparation for atom probe tomography, Ultramicroscopy 107 (2–3) (2007) 131–139. [10] D.J. Larson, D. Lawrence, W. Lefebvre, D. Olson, T.J. Prosa, D.A. Reinhard, R.M. Ulfig, P.H. Clifton, J.H. Bunton, D. Lenz, J.D. Olson, L. Renaud, I. Martin,
[11]
[12]
[13] [14]
[15]
[16]
69
T.F. Kelly, Toward atom probe tomography of microelectronic devices, Journal of Physics: Conference Series 326 (2011) 012030. M.K. Miller, K.F. Russell, K. Thompson, R. Alvis, D.J. Larson, Review of atom probe FIB-based specimen preparation methods, Microscopy and Microanalysis: The Official Journal of Microscopy Society of America, Microbeam Analysis Society, Microscopical Society of Canada 13 (6) (2007) 36–428. S. Nagamachi, Focused ion beam direct deposition and its applications, Journal of Vacuum Science & Technology B: Microelectronics and Nanometer Structures 16 (4) (1998) 2515. A.K. Kambham, J. Mody, M. Gilbert, S. Koelling, W. Vandervorst, Atom-probe for FinFET dopant characterization, Ultramicroscopy 111 (6) (2011) 9–535. K. Thompson, B. Gorman, D. Larson, B. van Leer, L. Hong, Minimization of Ga induced FIB damage using low energy clean-up, Microscopy and Microanalysis 12 (S02) (2006) 1736. A new twin-well CMOS process using nitridized-oxide-LOCOS (NOLOCOS) isolation technology, IEEE Electron Device Letters, vol. 10(7), July 1989, pp. 307–309. D. Larson, D. Foord, A. Petford-Long, H. Liew, M. Blamire, A. Cerezo, G.D. Smith, Field-ion specimen preparation using focused ion-beam milling, Ultramicroscopy 79 (1–4) (1999) 287–293.
Contents lists available at ScienceDirect
Ultramicroscopy journal homepage: www.elsevier.com/locate/ultramic
3D site specific sample preparation and analysis of 3D devices (FinFETs) by atom probe tomography Ajay Kumar Kambham a,b,n, Arul Kumar a,b, Matthieu Gilbert b, Wilfried Vandervorst a,b a b
KU Leuven, Department of Physics and Astronomy, B-3001 Leuven, Belgium IMEC, Kapeldreef 75, B-3001 Leuven, Belgium
a r t i c l e i n f o
abstract
Available online 24 February 2013
With the transition from planar to three-dimensional device architectures such as Fin field-effecttransistors (FinFETs), new metrology approaches are required to meet the needs of semiconductor technology. It is important to characterize the 3D-dopant distributions precisely as their extent, positioning relative to gate edges and absolute concentration determine the device performance in great detail. At present the atom probe has shown its ability to analyze dopant distributions in semiconductor and thin insulating materials with sub-nm 3D-resolution and good dopant sensitivity. However, so far most reports have dealt with planar devices or restricted the measurements to 2D test structures which represent only limited challenges in terms of localization and site specific sample preparation. In this paper we will discuss the methodology to extract the dopant distribution from real 3D-devices such as a 3D-FinFET device, requiring the sample preparation to be carried out at a site specific location with a positioning accuracy 50 nm. & 2013 Elsevier B.V. All rights reserved.
Keywords: Atom probe tomography FinFET Dopant conformality Focused ion beam
1. Introduction The introduction of FinFETs for the (sub)-22 nm [1] node brings a major challenge for the doping profile formation and optimization. To optimize the 3D-doping profiles in FinFET-based devices, one can use ion implantation [2], vapor phase doping (VPD) [3] or plasma doping [4] which all have distinct properties in terms of doping incorporation, conformality and process integration. In each case a crucial step in optimizing these doping technologies is the ability to assess their final doping profile characteristics which thus requires the availability of adequate 3D-dopant profiling techniques which can probe directly on devices with sub-nm resolution. In this work we have developed a methodology based on the atom probe tomography (APT) [5–8] to probe the 3D-dopant distribution inside a FinFET device and extract the important parameters such as gate overlap and profile gradient. The FinFET devices used in this work originate from a standard process flow whereby the doping step is based on self regulatory plasma doping (SRPD) [4].
2. Device fabrication and sample preparation: The results were obtained on test structures with the Fin height (HFin) targeted at 60 nm, the Fin width (WFin) 40 nm n Corresponding author at: KU Leuven, Department of Physics and Astronomy, B-3001 Leuven, Belgium. Tel.: þ32 16 28 12 07; fax: þ 32 16 28 15 15. E-mail addresses: [email protected], [email protected] (A.K. Kambham).
0304-3991/$ - see front matter & 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ultramic.2012.09.013
and gate width (Wgate) 50 nm. Fig. 1 shows the SEM image of the 3D structures after gate patterning before plasma doping and spike annealing. After the doping step the structures are covered with a suitable cap layer preferably polysilicon 100 nm in order to reduce the impact of the major insulating materials around the Fins on the APT analysis and the damage of the FIB ion beam during APT sample preparation. As the field of view in APT is fairly limited, a crucial step is to position the circled location (shown in Fig. 1) in the area of APT-tip which contains the relevant information on conformality, gate overlap and 3D-dopant distribution. This positioning is aggravated by the fact that the cap layer removes most of the topographical information which was visible in Fig. 1.
2.1. Localization and sample preparation by FIB There are many reports [9–14] to solve the sample preparation issues targeting to place the area of interest (AOI) in the final sample/ tip for 1D and 2D structures using FIB system for atom probe studies. Whereas for that case the positioning is facilitated by the fact that at least one dimension is unrestricted or extends over several microns, the challenge for the present problem is the 3D confinement of the AOI to volume in the order of (50 40 60 nm3). Moreover, after depositing the polysilicon and chemical mechanical polishing (CMP) very little SEM-contrast is available from the top to localize the AOI. In order to localize and study these 3D confined areas using APT, we need to use and reference marks/cuts by FIB during the lamella preparation and annular milling. The only reference
66
A.K. Kambham et al. / Ultramicroscopy 132 (2013) 65–69
which is available after the test structure fabrication is the presence of a pad oxide ( 80 nm above the Fins) used to protect the gate during the source/drain doping [15]. We will use the
Fig. 1. SEM image of 3D test structures prior to polysilicon deposition, circled location is the area of the interest for atom probe analysis.
topography of the pad oxide to locate and identify the Fin. As a first approach we tried to deposit platinum (for a reference) at low ion beam energies 5 kV on top of the AOI in an area of 50 50 nm2. Unfortunately the concurrent energetic ion impact leads also to rapid erosion counteracting the marking through platinum deposition (Fig. 2a). The second problem is the need to mark the AOI by a FIB cut. The distance between two FIB markings is also important when we tried to place FIB marks close to the AOI ( o500 nm) the resultant marks are too large which are getting overlap, encroach and damage the AOI as shown in Fig. 2b. We solve these problems by depositing e-beam platinum on one of the Fins using the distance to the pad oxide as a reference and make FIB line cuts at a distance 4700 nm from AOI at low ion beam energies 5–8 keV. Initiating the shank preparation, Fig. 3 shows the process steps used to make the lamella with a length and width of 12 2.5 mm2 based on the lift-out method [11]. The annular milling of the shank will start by keeping the ebeam platinum mark in the center at 30 kV ion beam energies. After removing sufficient material around the AOI, we can see some contrast of the gate stack (in SEM) below the pad oxide; by using that contrast we can position the X and Y position of the shank to finalize the tip. The final cleaning and sharpening of the tip is performed at low ion beam energies (5–2 kV) [16]. The cleaning includes the removal of the pad oxide, as its presence causes tip rupture. Fig. 4 shows the tip evolution during the annular milling to get the final tip. The success rate of this sample preparation depends on the accuracy of the positioning of the ebeam platinum deposition. A small shift in the e-beam platinum position will end-up making the tip in the trench region instead of at the gate/source drain edge.
Fig. 2. SEM images of ion beam damages (a) though we ask for Pt deposition on top of the AOI, the ion beam is milled out the AOI as shown in circle. (b) Encroachment of FIB marks inside the AOI when you try to mark them at o 500 nm from AOI (beam energy is 30 kV).
Fig. 3. SEM images of process sequence used to localize the center of area of the interest for APT sample preparation: (a) image of topography presented by pad oxide above the Fins after e-beam platinum deposition on top of AOI; (b) FIB markings close to the AOI at low ion energies; and (c) final lamella with e-beam platinum dot on one of the Fins with platinum filled FIB cults.
A.K. Kambham et al. / Ultramicroscopy 132 (2013) 65–69
67
Fig. 4. SEM images of tip evolution during annular milling: (a) e-beam Pt mark is used to position center for milling; (b) the identification pad oxide at one stage of annular milling; (c) after milling at 16 kV ion beam energy by keeping AOI at the center; and (d) final cleaning of pad oxide and shaping of the tip at 5 kV and 2 kV, respectively.
Fig. 5. APT analysis of As dopants at the gate and S/D interface (orange) and gate stack composed of HfO2 (black) and TiN (pink). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 6. APT analysis of arsenic doping and diffusion from S/D region to gate (see the schematic). Silicon atoms are not shown, orange—As, black—HfO2, pink—TiN. The white dotted line (right) shows the under diffused dopant in the channel region below the gate. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
A.K. Kambham et al. / Ultramicroscopy 132 (2013) 65–69
Arsenic Concentration (cm-3)
68
Annealed Non- Annealed
1E21
Source / Drain
GATE
1E20
1E19
0
10
20
30
40
50
Depth (nm) Fig. 9. Comparison of gate overlap dopant profiles from S/D region to channel (below the gate) region for annealed and non-annealed samples.
Fig. 7. APT image of arsenic (orange) distribution with native oxide (SiO2, green) at top and side wall of the Fin in S/D region. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Atomic %
Lateral Profile Top Profile
1
5E20
0.1
5E19
0.01 0
5
10
15
20
25
30
35
Concentration (at./Cm3)
5E21
10
5E18 40
Depth (nm) Fig. 8. Comparison of lateral and vertical dopant profiles extracted at lateral and top of the Fin in source/drain region.
3. Results and discussion The APT-analysis was performed using a laser (400 fs, 515 nm) assisted wide angle atom probe (LAWATAP) from CAMECA at 80 K temperature. Tap3D data reconstruction software provided by CAMECA has used to reconstruct and analyze the data. Figs. 5–7 illustrate that we are able to probe a complete 3D-Fin transistor as highlighted by the 3D-distribution of atoms from the 3D-gate stack to S/D region. Fig. 5 shows the 2D atomic map (As, TiN, Hf) with the As-dopant distribution (orange) for a slice taken at the gate edge and S/D region. Fig. 6 shows a 3D-cross sectional view of half the Fin volume (40 20 60 nm3, see the inset) with 3D-dopant distribution of under diffused dopants from S/D to gate. Fig. 6 provides the
required information about the variations in dopant distributions, in particular the gate under lap (under diffusion from S/D region to gate) from the top surface ( 14 nm) to the bottom along the side walls ( 5 nm). Fig. 7 shows the 2D-map of the As-distribution (orange) atoms in the S/D region. Fig. 8 illustrates the comparison between the dopant profiles extracted in the center of the gate (top profile) and from one side of the Fin to the other side (lateral profile). The top dopant profile has a much higher dopant peak concentration and dose than the lateral dopant profile, which leads to the non-corformality of the dopant distribution inside the Fin. The higher doping at the top of the Fin actually induces now a larger under diffusion and thus increased the gate overlap (Fig. 6). The availability of this metrology enables us now to study the diffusion processes operative within the FinFET process. This is exemplified in Fig. 9 where we show for the unannealed and annealed sample the increase in dopant gate overlap following the anneal step as extracted from the source/drain region to the inside of the gate. Profiles in Fig. 9 are extracted at the top within the middle of the Fin width shown an increase in under diffusion (relative to the gate edge as defined by the TiN signal) from 4 nm (non-annealed) to 14 nm for annealed case.
4. Conclusion In this study, we have described a methodology to study 3D dopant distributions in 3D devices like FinFETs using the APT technique, including the site specific sample preparation. The results demonstrate the ability to provide detailed information on the 3D-dopant distributions and gate overlap throughout all regions in the Fin transistor.
Acknowledgments IMEC acknowledges the collaboration with Cameca on the LAWATAP system. References [1] Intel 22 nm 3-D Tri-Gate Transistor Technology. [2] R. Duffy, G. Curatola, B.J. Pawlak, G. Doornbos, K. van der Tak, P. Breimer, J.G.M. van Berkum, F. Roozeboom, Doping fin field-effect transistor sidewalls: impurity dose retention in silicon due to high angle incident ion implants and the impact on device performance, Journal of Vacuum Science & Technology B: Microelectronics and Nanometer Structures 26 (1) (2008) 402.
A.K. Kambham et al. / Ultramicroscopy 132 (2013) 65–69
[3] S. Takeuchi, N.D. Nguyen, F.E. Leys, R. Loo, T. Conard, W. Vandervorst, M. Caymax, Vapor phase doping with N-type dopant into silicon by atmospheric pressure chemical vapor deposition, ECS Transactions 16 (2008) 495–502. [4] Y. Sasaki, K. Okashita, B. Mizuo, M. Kubota, M. Ogura, O. Nishijima, Conformal doping mechanism for fin field effect transistors by self-regulatory plasma doping with AsH3 plasma diluted with He, Journal of Applied Physics 111 (1) (2012) 013712. [5] R.G. Forbes, Field evaporation theory: a review of basic ideas, Applied Surface Science 87–88 (1995) 1–11. [6] G.L. Kellogg, T.T. Tsong, Pulsed-laser atom-probe field-ion microscopy, Journal of Applied Physics 51 (2) (1980) 1184. [7] T.F. Kelly, M.K. Miller, Invited review article: atom probe tomography, The Review of Scientific Instruments 78 (3) (2007) 031101. [8] M.K. Miller, A. Cerezo, M.G. Hetherington, G.D.W. Smith, Atom Probe Field Ion Microscopy, Oxford University Press, 1996. [9] K. Thompson, D. Lawrence, D.J. Larson, J.D. Olson, T.F. Kelly, B. Gorman, In situ site-specific specimen preparation for atom probe tomography, Ultramicroscopy 107 (2–3) (2007) 131–139. [10] D.J. Larson, D. Lawrence, W. Lefebvre, D. Olson, T.J. Prosa, D.A. Reinhard, R.M. Ulfig, P.H. Clifton, J.H. Bunton, D. Lenz, J.D. Olson, L. Renaud, I. Martin,
[11]
[12]
[13] [14]
[15]
[16]
69
T.F. Kelly, Toward atom probe tomography of microelectronic devices, Journal of Physics: Conference Series 326 (2011) 012030. M.K. Miller, K.F. Russell, K. Thompson, R. Alvis, D.J. Larson, Review of atom probe FIB-based specimen preparation methods, Microscopy and Microanalysis: The Official Journal of Microscopy Society of America, Microbeam Analysis Society, Microscopical Society of Canada 13 (6) (2007) 36–428. S. Nagamachi, Focused ion beam direct deposition and its applications, Journal of Vacuum Science & Technology B: Microelectronics and Nanometer Structures 16 (4) (1998) 2515. A.K. Kambham, J. Mody, M. Gilbert, S. Koelling, W. Vandervorst, Atom-probe for FinFET dopant characterization, Ultramicroscopy 111 (6) (2011) 9–535. K. Thompson, B. Gorman, D. Larson, B. van Leer, L. Hong, Minimization of Ga induced FIB damage using low energy clean-up, Microscopy and Microanalysis 12 (S02) (2006) 1736. A new twin-well CMOS process using nitridized-oxide-LOCOS (NOLOCOS) isolation technology, IEEE Electron Device Letters, vol. 10(7), July 1989, pp. 307–309. D. Larson, D. Foord, A. Petford-Long, H. Liew, M. Blamire, A. Cerezo, G.D. Smith, Field-ion specimen preparation using focused ion-beam milling, Ultramicroscopy 79 (1–4) (1999) 287–293.