Individual Atomic Imaging of Multiple Dopant Sites in As-Doped Si

Nov 17, 2017 - The atomic scale characterization of dopant atoms in semiconductor devices to establish correlations with the electrical activation of ...
2 downloads 5 Views 1MB Size
Subscriber access provided by READING UNIV

Communication

Individual Atomic Imaging of Multiple Dopant Sites in As-doped Si Using Spectro-photoelectron Holography Kazuo Tsutsui, Tomohiro Matsushita, Kotaro Natori, Takayuki Muro, Yoshitada Morikawa, Takuya Hoshii, Kuniyuki Kakushima, Hitoshi Wakabayashi, Kouichi Hayashi, Fumihiko Matsui, and Toyohiko Kinoshita Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.7b03467 • Publication Date (Web): 17 Nov 2017 Downloaded from http://pubs.acs.org on November 19, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Nano Letters is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Individual Atomic Imaging of Multiple Dopant Sites in As-doped Si Using Spectro-photoelectron Holography Kazuo Tsutsui1*, Tomohiro Matsushita2, Kotaro Natori3, Takayuki Muro2, Yoshitada Morikawa4, Takuya Hoshii3, Kuniyuki Kakushima3, Hitoshi Wakabayashi3, Kouichi Hayashi5, Fumihiko Matsui6, and Toyohiko Kinoshita2. 1

Institute of Innovative Research, Tokyo Institute of Technology, 4259-J2-69, Nagatsuta, Midori-ku, Yokohama 227-8503, Japan

2

Japan Synchrotron Radiation Research Institute (JASRI), 1-1-1 Kouto, Sayo, Hyogo 679-5198, Japan 3

School of Engineering, Tokyo Institute of Technology, 4259-J2-69, Nagatsuta, Midori-ku, Yokohama 227-8502, Japan

4

Department of Precision Science and Technology, Osaka University, 2-1, Yamada-oka, Suita, Osaka 565-0871, Japan

5

Department of Physical Science and Engineering, Nagoya Institute of Technology, Gokiso-chō, Shōwa-ku, Nagoya, Aichi 466-8555, Japan

ACS Paragon Plus Environment

1

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

6

Page 2 of 26

Graduate School of Materials Science, Nara Institute of Science and Technology, Takayamacho, Ikoma, Nara 630-0192, Japan

KEYWORDS: photoelectron holography, impurity, dopant, cluster, arsenic, silicon

Abstract

The atomic scale characterization of dopant atoms in semiconductor devices to establish correlations with the electrical activation of these atoms is the essential to the advancement of contemporary semiconductor process technology. Spectro-photoelectron holography combined with first-principles simulations can determine the local three-dimensional atomic structures of dopant elements, which in turn affect their electronic states. In the work reported herein, this technique was used to examine arsenic (As) atoms doped into a silicon (Si) crystal. As 3d core level photoelectron spectroscopy demonstrated the presence of three types of As atoms at a total concentration of approximately 1020 cm-3, denoted as BEH, BEM, and BEL. Based on Hall effect measurements, the BEH atoms corresponded to electrically active As occupying substitutional sites and exhibiting larger thermal fluctuations than the Si atoms, while the BEM atoms corresponded to electrically inactive As embedded in the AsnV (n=2~4) type clusters. Finally, the BEL atoms were assigned to electrically inactive As in locally disordered structures.

ACS Paragon Plus Environment

2

Page 3 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Techniques for the electrical activation of impurities doped into semiconductors with high efficiency and/or at high concentrations have always been an essential aspect of semiconductor device technology. As an example, the formation of shallow pn junctions in scaled silicon (Si)CMOS devices has been actively pursued, using high concentrations of electrically active dopants in the source/drain extension regions.1 However, despite various developments in Sidevice processing, such as plasma doping,2 rapid thermal processes3 (e.g., spike-rapid thermal annealing (spike-RTA),4 flash lamp annealing,5 and laser annealing6,7), the maximum concentration of an active dopant that can be practically achieved is limited. It is known that an individual dopant atom occupying a substitutional site in a crystal matrix emits a carrier, meaning that it is electrically active. The current limitations on active dopant concentrations result from the deactivation of excess dopant atoms by the formation of various types of clusters and other defect structures. As such, an understanding of the atomic structures of dopants in semiconductors would assist in the development of new process technologies for the fabrication of high-performance devices. For this reason, atomic level dopant structures have been investigated using both theoretical and experimental approaches,6-18 although the direct observation of the three-dimensional (3D) structures of dopant arrangements has been difficult to achieve. X-ray diffraction (XRD) and electron diffraction are powerful methods of visualizing 3D atomic structures; however, these techniques are only applicable to crystalline structures and cannot be applied to the visualization of dopant structures because the dopants are not dispersed with a regular periodicity in the crystal. X-ray absorption fine structure (XAFS) can provide information regarding dopants, although only with respect to the atomic distance.7,11,15 In addition, ion scattering can be used to detect the positions of target atoms relative to crystalline

ACS Paragon Plus Environment

3

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 26

matrix channels.16-18 High resolution scanning transmission electron microscopy (STEM) has also been used to image individual dopant atoms in real space, and has provided data regarding the formation of dopant clusters.19, 20 The determination of the 3D structure of the dopant has also been attempted.21 In principle, however, direct imaging for light elements or cases involving vacancies is difficult. All these characterization methods thus provide information concerning the atomic structures, but do not improve our understanding of the chemical environment. In addition to the studies noted above, dopants such as boron (B) and arsenic (As) in Si have been examined using soft X-ray photoelectron spectroscopy (SXPES) and three different chemical bonding states with varying chemical shifts in their core level spectra (B 1s and As 3d) have been reported.22,23 In these prior works, the peaks corresponding to the lowest B 1s and highest As 3d binding energy values were assigned to electrically active dopants. However, the 3D atomic arrangements of these dopants could not be determined using solely SXPES. Photoelectron holography is a method of examining element specific local 3D atomic structures that is also applicable to non-periodic structures.24-28 Recently, the development of the SPEA-MEM/SPEA-L1 (scattering pattern extraction algorithm using the maximum entropy method/L1 regularization)29-35 reconstruction algorithms has greatly improved the quality of reconstructed atomic images. Furthermore, combining a high energy resolution electron analyzer with spectro-photoelectron holography has allowed visualization of local dopant atomic structures having different chemical bonding states that are correlated with electronic properties. The challenge when using this technique for the study of dopants in semiconductors is to obtain both high sensitivity and high energy resolution simultaneously. This is especially true because the concentrations of dopants are typically lower than 1 at%, and the chemical shifts of the core level spectra of specific dopant atoms with different structures are less than 1 eV. In the present

ACS Paragon Plus Environment

4

Page 5 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

work, we report the 3D imaging of dopant structures in semiconductors using spectrophotoelectron holography combined with first-principles simulations. This method was applied to estimations of the 3D structures of As atoms doped into a Si(001) surface as well as to assessments of different chemical bonding states. In this work, As+ ions were implanted into a p-type Si(001) wafer, employing a dose of 1.5×1015 cm-2 and an energy of 3 keV. The Si wafer was initially doped with B but at a concentration of approximately 1016 cm-3 so as not to affect the subsequent As doping. The wafer was subsequently treated using the spike-RTA method, applying a peak temperature of 1000°C for no longer than 1 s without a cap layer in a N2 atmosphere. Following this treatment, a 10 nm SiO2 cap layer was deposited by atomic layer deposition (ALD) method and a second RTA treatment was applied, consisting of heating at 1050°C for 1 min in an Ar atmosphere. Following this second treatment, the cap layer was removed with diluted HF. The spike-RTA method is widely employed to obtain shallow junctions. However, in this case, the second RTA process was incorporated to obtain a higher activation rate despite the deep diffusion of As atoms. Depth profiles of the As concentration after these processes were determined using secondary ion mass spectrometry (SIMS). Samples for spectro-photoelectron holography were prepared using a step-by-step etching process22 involving repeated oxidation of the As-doped wafer in an atmosphere containing ozone at room temperature and removal of the oxidized layer by dilute HF. In this manner, a new face that had originally been several nm or several tens of nm under the original surface was exposed. The carrier concentration in the As-doped layer on a separate sample was also evaluated based on Hall effect measurements, applying the same step-by-step etching process used for the preparation of holography specimens.

ACS Paragon Plus Environment

5

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 26

The spectro-photoelectron holography measurements were performed at the BL25SU beamline at the SPring-8 facility.36,37 Figure 1 shows the experimental arrangement. Soft X-rays generated by synchrotron radiation were incident to the sample at a 5° glancing angle relative to the surface. The principle of photoelectron holography is summarized in Figure 1(b). Essentially, the soft X-rays excite the core level electrons such that photoelectrons are emitted from various atoms. A portion of each photoelectron wave is scattered by surrounding atoms and the scattered wave interferes with the original direct photoelectron wave. The resulting interference pattern (that is, the photoelectron hologram) appears in the angular distribution of the photoelectrons. A photoelectron hologram generated by the specific core levels of the dopant atoms was captured with an electron analyzer (Scienta-Omicron DA30) capable of measuring the angular distribution of the specific kinetic energy photoelectrons with an acceptance angle of ±15° × ±10°, by sweeping the bias of the deflector lens. Entire holograms were obtained by stepwise variations in the sample angles θ and φ. The incident photon energies were 641 eV and 699 eV for the As 3d and Si 2p measurements, respectively, so as to obtain photoelectrons with kinetic energy values of approximately 600 eV. All the samples were at room temperature during the measurements. Because the photoelectron spectra acquired in this manner contained information from more than one atomic site, peak fitting was applied to all angle-resolved spectra to obtain the photoelectron hologram of individual atomic sites. This 3D atomic image reconstruction was conducted using the SPEA-L1 algorithm.29-34 This algorithm is based on a fitting process in which a holographic function created by one atom is used as a basis function and generates more accurate atomic images than can be obtained with a Fourier transform based method. This atomic image can be interpreted as the occupancy probability of atoms, and the average structure around

ACS Paragon Plus Environment

6

Page 7 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

the various dopant atoms in the crystal having the same core level binding energy can be reconstructed using the photoelectron holography technique. First-principles calculations were performed using the STATE-Senri (Simulation Tool for Atom TEchnology) program,38,39 which is based on density functional theory within a generalized

gradient

approximation

(DFT-GGA),40,41

utilizing

norm-conserving

pseudopotentials42 for both Si and As to describe the interactions between electrons and atomic ions. These calculations were based on a unit cell containing 64 Si atoms in which one or two Si atoms were substituted by As atoms. Core level shifts were calculated including both the initial and final state effects using pseudopotentials with core holes.43 First-principles molecular dynamics (MD) simulations were carried out with a time step of 2.4 fs, maintaining the system temperature at 300K using a Nose-Hoover thermostat. Figure 2(a) shows depth profiles of the As concentrations in samples following the initial spike-RTA and after the second RTA. The experimental results discussed below were obtained after the second RTA, during which the As diffused deeper into the specimen and was relatively uniformly distributed within the doped layer. A sample etched to a depth of 36 nm was used for spectro-photoelectron holography analyses, so that the As concentration at the sample surface (as indicated in Figure 2(a)) was 1.5×1020 cm-3 (0.3 at%). Figure 2(b) shows the carrier concentrations as a function of depth in the sample after the second RTA, and the corresponding As concentration profile. The carrier concentration at a depth of 36 nm from the surface, which corresponds to the surface position for photoelectron observations following the 36-nm etching described above, was determined to be 7.5×1019 cm-3, with an experimental error of ±20%. Therefore, the activation rate was approximately 50% in the region used to generate the photoelectron hologram.

ACS Paragon Plus Environment

7

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 26

Figure 3(a) presents the Si 2p photoelectron hologram, which is evidently quite clear and cross-sectional images of the reconstructed atomic sites are provided in Figure 3(b). Each Si unit cell had two equivalent atomic sites; the (0,0,0) and (1/4,1/4,1/4) sites. Therefore, the reconstructed atomic image is the sum of the images from these two atomic sites, as indicated by the circles in this figure. The square in the figure at z=0.0 nm indicates the photoelectron emitter, Si. The atomic images of the even numbered neighbors (NBs) are stronger and sharper than those of the odd numbered NBs because the associated peaks have twice the intensity due to the summation of the two atomic images from the A- and B-sites. The majority of the atomic images were reconstructed, although some are missing as a result of a loss of the scattered wave intensity. This occurs because the amplitude of the spherical photoelectron wave function decreases with the distance travelled. In addition, the intensity is also reduced by inelastic scattering because the mean free path of the 600 eV electrons is approximately 1 nm. Therefore, the intensity of the scattered waves from distant atoms was weak, and the reconstructed images were less intense. In these figures, the fifth nearest and further NB atoms are unclear, while the closer NB atoms are clearly reconstructed. Figure 4 shows the experimental As 3d spectrum and photoelectron holograms. The spectrum was found to contain three components labeled BEH, BEM, and BEL, which are attributed to three distinct atomic sites. Peak fitting was conducted for each pixel and three photoelectron holograms were obtained, as shown in Figures 4(a), (b), and (c). The chemical shifts relative to the BEL peak were 1.2 and 0.4 eV for the BEH and BEM peaks, respectively, as shown in Figure 4(d). The hologram associated with the BEH peak is very clear and its pattern quite similar to that of the Si 2p hologram. It should be noted that the forward focusing peaks corresponding to the second nearest atom directions are more intense, while the

ACS Paragon Plus Environment

8

Page 9 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

forward focusing peaks corresponding to the first nearest atom directions are less clear. Since photoelectron holography provides averaged dopant structures, the atomic images of polymorphic structures are inevitably weak. These features are discussed based on the reconstructed atomic images in the following paragraph. Although the pattern associated with the BEM peak is less intense, Kikuchi-lines are clearly evident and the forward focusing peaks are still present. In contrast, the pattern from the BEL peak is unclear and has little structure. The atomic images of these sites were subsequently reconstructed, as shown in Figures 5 (ac) and (e-g), which correspond to the BEH and BEM peaks, respectively. The atomic image for the BEH peak was similar to the Si structure and we therefore conclude that the As atoms giving this peak were located at substitutional sites in the Si lattice. However, the atomic image of the first nearest NB was quite weak compared with that for Si. One possible explanation for this result is shown in the structural image of Figure 5(d), in which the first NB Si atoms fluctuate significantly. Our first-principles MD simulations indicated that the local phonons around the As dopant atoms are more thermally exited than those in the Si lattice. In this case, the position of the first nearest NB Si atom as seen from the As atom fluctuates significantly in the angular direction, thus weakening the image of first NB atoms but not second NB atoms. The holographic amplitude for the BEM peak was quite weak, although Kikuchi lines44 can still be seen in the hologram, as shown in Figure 4(b). The reconstructed atomic image at (1/2, 1/2, 0) in Figure 5(e) (z=0) was elongated. Note that the atomic images of the first NB at (1/4,1/4,1/4) are visible. Based on these results, we conclude that As atoms occupied the substitutional sites even though the positions of surrounding atoms fluctuated. A candidate structure is shown in Figure 5(h), and this structure is discussed in more detail further on.

ACS Paragon Plus Environment

9

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 26

The photoelectron hologram associated with the BEL peak was unclear, with no atomic image in the reconstructed image. The As atoms responsible for the BEL peak are thus considered to be located in either amorphous or disordered structures, such as in zones of precipitated Si-As17 within the Si crystal. An advantage of photoelectron holography is the ability to determine the concentrations of dopants at different sites based on peak intensities in the photoelectron spectrum. Using these data, the relationship between the carrier concentration and the As structures was assessed. The relative BEH, BEM, and BEL peak intensities were determined to be 37%, 39%, and 24% based on the angle-integrated spectrum shown in Figure 4(d), while the activation rate was approximately 50% (ranging from 40% to 60% considering experimental error) as shown in Figure 2(b). The As atoms associated with the BEH peak and occupying the substitutional sites are expected to be electrically active, and their relative peak intensity of 37% is close to the activation rate. This assignment is consistent with the results of previous work.23 Based on the assignment of the BEH peak, the As atoms producing the BEM peak should be electrically inactive, because otherwise the activation rate would be too high (37%+39%=76%). The As atoms generating the BEL peak are also thought to be electrically inactive. This finding is in agreement with our previous work,23 in which As atoms with a binding energy close to that of BEL were predominant among the three chemical bonding states for a sample with a higher As concentration and lower activation rate than that used in this work. At this point, we discuss the atomic structure of the BEM atoms. The electrically inactive As atoms in the Si crystal have been thought to exist in various cluster structures,10-16 typically involving complexes of multiple As atoms associated with a Si vacancy. In these, so-called AsnV (n=1~4) structures, the As atoms generally occupy substitutional sites. Donor-pair type clusters,

ACS Paragon Plus Environment

10

Page 11 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

such as DP(2) and DP(4), have also been discussed,45 in which As atoms occupy second NB or fourth NB lattice sites. Since the reconstructed atomic images suggest significant fluctuation of the second NBs, as shown in Figure 5(e), the BEM As atoms may exist in either AsnV (n=1~4) or DP(2) clusters. In order to ascertain the validity of these structures, we performed first-principles calculations to find the binding energy shifts for As 3d core levels, and Table 1 summarizes the calculated binding energy shifts relative to the binding energy of independent substitutional As atoms. The binding energy shift of the DP(2) structure (the As pair at the second NB position without a vacancy) was determined to be approximately +0.01 eV. Since the experimental results indicated a -0.8 eV shift in binding energy, the DP(2) assignment is evidently not appropriate. In contrast, the AsnV (n=1~4) cluster structures show binding energy shifts between -1.05 and -0.68 eV, which are in good agreement with the experimental results. However, since our firstprinciples calculations and previous reports in the literature46,47 indicate that the AsnV (n=1) formation energy is significantly higher than those of the AsnV (n=2~4) structures, the AsnV (n=1) structure was excluded from the candidates. The As2V structure is shown in Figure 5(h). In addition, we carried out first-principles MD simulations for the As2V cluster in a Si lattice and found that the fluctuations of the As atom positions were slightly greater compared with that of a simple substituted As atom. Therefore, the As atoms generating the BEM peak are assigned to AsnV (n=2~4) cluster structures. According to the MD simulations, the angular fluctuations of the first NB from the As atom are also present in these AsnV cluster structures. Note that the first NB atoms are visible in the case of BEM, while they are not in the case of BEH. The present investigation brought out the importance of local vibrational behavior as well as the electronic structure and formation energy in consideration for the further pursuit of BEM structure.

ACS Paragon Plus Environment

11

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 26

In conclusion, the 3D atomic structures of As doped into a Si crystal were successfully revealed using spectro-photoelectron holography combined with measurements of electrical properties and first-principles molecular dynamics simulations. The As atoms in the Si crystal lattice exhibited three different chemical bonding states, and the local atomic arrangements of As were determined based on the holograms extracted from these photoelectron peaks. The As atoms with the highest binding energy (the BEH atoms) were identified as those substituted into the Si crystal and were found to represent the electrically active dopants. In the case of these atoms, the bonding angle to the first NB fluctuated while no such fluctuation was observed for atoms further than the second NB. The other two states, having medium and the lowest binding energies (BEM and BEL), were identified as electrically inactive. The As atoms associated with the BEM peak are believed to exist in AsnV (n=2~4) type clusters, in which the paired As atoms occupy at the second NB site. The As atoms producing the BEL peak are evidently in localized regions of As-Si precipitation or amorphous areas within the Si matrix. This work demonstrated the potential of spectro-photoelectron holography for the analysis of impurities in semiconductors by assessing the correlation of atomic arrangements with electrical properties and binding energy shifts. Such analyses are difficult to perform using conventional techniques such as XAFS, ion scattering, and high-resolution STEM. The ability to observe atomic fluctuations is also an advantage of the photoelectron holography method. This technique should therefore be useful for the investigation and development of doping techniques for various semiconductor process technologies.

AUTHOR INFORMATION Corresponding Author

ACS Paragon Plus Environment

12

Page 13 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Kazuo Tsutsui, [email protected] ORCID Kazuo Tsutsui: 0000-0002-5472-5539 Tomohiro Matsushita: 0000-0002-2547-3811 Takayuki Muro: 0000-0003-2626-0110 Yoshitada Morikawa: 0000-0003-4895-4121 Hitoshi Wakabayashi: 0000-0001-5509-521X Kouichi Hayashi: 0000-0002-8782-4293 Fumihiko Matsui: 0000-0002-0681-4270 Toyohiko Kinoshita: 0000-0002-8030-0796

Notes The authors declare no competing financial interest.

ACKNOWLEDGMENTS This work was supported by Grants-in-Aid for Scientific Research on Innovative Areas "3D Active-Site Science": Grant nos. 26105010, 2610513 and 26105014 from the Japan Society for the Promotion of Science (JSPS), and by the Photon and Quantum Basic Research Coordinated Development Program of the Ministry of Education Culture, Sports, Science and Technology (MEXT). The authors would also like to acknowledge the support of Dr. Masashi Nakatake and Dr. Yoshio Watanabe of the Aichi Synchrotron Radiation Centre and Prof. Masaru Shimomura of Shizuoka University during the preliminary experiments, and thank Prof. Emeritus Hiroshi

ACS Paragon Plus Environment

13

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 26

Iwai of Tokyo Institute of Technology for helpful discussion. The synchrotron radiation experiments were performed with the approval of the Japan Synchrotron Radiation Research Institute (JASRI, proposal nos. 2014B1819 and 2016A1261).

ACS Paragon Plus Environment

14

Page 15 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

REFERENCES 1. Taur,Y; Ning, T. H. Fundamentals of Modern VLSI Devices, 2nd Edition; Cambridge University Press: New York, 2009; pp 175-186. 2. Chang, C. Y.; Sze, S. M. ULSI Technology; MaGraw-Hill, Singapore, 1996, pp 176-192. 3. Sasaki, Y.; Jin, C.G.; Okashita, K.; Tamura, H.; Ito, H.; Mizuno, B.; Sauddin, H.; Higaki, R.; Satoh, T.; Majima, K.; Fukagawa, Y.; Takagi, K.; Aiba, I.; Ohmi, S.; Tsutsui, K.; Iwai, H. Nucl. Instrum. Methods Phys. Res., Sect. B 2005, 237, 41-45. 4. Fiory, A. T. Proc. of 8th Int. Conf. on Advanced Thermal Processing of Semiconductors, Sept. 20-22, 2000, p15. 5. Ito, T.; Iinuma, T.; Murakoshi, A.; Akutsu, H.; Suguro, K.; Arikado, T.; Okumura, K.; Yoshioka, M.; Owada, T.; Imaoka, Y.; Murayama, H.; Kusuda, T. Jpn. J. Appl. Phys., 2002, 41, 2394-2398. 6. Poon, C. H.; See, A.; Zhou, M.; Wong, C. W. IEEE Trans. Semiconductor Manufacturing 2009, 22, 175-179. 7. Giubertoni, D.; Pepponi, G.; Bersani, M.; Gennaro, S.; D’Acapito, F.; Doherty, R.; Foad, M. A. Nucl. Instrum. Methods Phys. Res., Sect. B 2006, 253, 9-12. 8. Harrison, S. A.; Edgar, T. F.; Hwang, G. S. Appl. Phys. Lett. 2004, 85, 4935-4937. 9. Solmi, S.; Nobili, D. J. Appl. Phys. 1998, 83, 2484-2490. 10. Solmi,S.; Nobili, D.; and Shao, J.; J. Appl. Phys. 2000, 87, 658-662.

ACS Paragon Plus Environment

15

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 26

11. Giubertoni, D.; Pepponi, G.; Gennaro, S.; Bersani, M.; Sahiner, M. A.; Kelty, S. P.; Doherty, R.; Foad, M. A.; Kah, M.; Kirkby, K. J.; Woicik, J. C.; Pianetta, P. J. Appl. Phys. 2008, 104, 103716. 12. Harrison, S. A.; Edgar, T. F.; Hwang, G. S. Phys. Rev. B 2005, 72, 195414. 13. Kim, Y.; Kirichenko, T. A.; Kong, N.; Henkelman, G.; Banerjee, S. K. Phys. Rev. B: Condens. Matter Mater. Phys. 2009, 79, 075201. 14. Pandey, K. C.; Erbil, A.; Cargill, G. S. III,; Boehme, R. F.; Vanderbilt, D. Phys. Rev. Lett. 1988, 61, 1282-1285. 15. Giubertoni, D.; Pepponi, G.; Sahiner, M. A.; Kelty, S. P.; Gennaro, S.; Bersani, M.; Kah, M.; Kirkby, K. J.; Doherty, R.; Foad, M. A.; Meirer, F.; Streli, C.; Woicik, J. C.; Pianetta, P. J. Vac. Sci. Technol., B Microelectron. Nanometer Struct.-Process., Meas. Phenom. 2010, 28, C1B1. 16. Kamgar, A.; Baiocchi, F. A.; Sheng, T. T. Appl. Phys. Lett. 1986, 48, 1090-1092. 17. Kobayashi, H.; Nomachi, I.; Kusanagi, S.; Nishiyama, F. Nucl. Instrum. Methods Phys. Res., Sect. B 2002, 190, 547-551. 18. Satta, A.; Albertazzi, E.; Bianconi, M.; Lulli, G.; Balboni, S.; Colombo, L. Nucl. Instrum. Methods Phys. Res., Sect. B 2005 230, 112-117. 19. Voyles, P. M.; Muller, D. A.; Grazul, J. L.; Citrin, P. H.; Gossamann, H.-J., Nature, 2002, 416, 826-829.

ACS Paragon Plus Environment

16

Page 17 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

20. Oshima, Y.; Hashimoto, Y.; Tanishiro, Y.; Takayanagi, Y.; Sawada, H.; Kaneyama, T.; Kondo, Y., Phys. Rev. B 2010, 81, 035317. 21. Ishikawa, R.; Lupini, A. R.; Findlay, S. D.; Taniguchi, T.; Pennycook S. J. Nano Lett. 2014, 14, 1903-1908. 22. Tsutsui, K.; Matsuda, T.; Watanabe, M.; Jin, C.-G.; Sasaki, Y.; Mizuno, B.; Ikenaga, E.; Kakushima, K.; Ahmet, P.; Maruizumi, T.; Nohira, H.; Hattori T.; Iwai, H. J. Appl. Phys. 2008, 104, 093709. 23. Kanehara, J.; Miyata, Y.; Nohira, H.; Izumi, Y.; Muro, T.; Kinoshita, K.; Ahmet, P.; Kakushima, K.; Tsutsui, K.; Hattori T.; Iwai, H. Ext. Abs. of 2011 Int. Conf. on Solid State Devices and Materials, Sept. 28-30, 2011, Aichi, Japan. pp 28-29. 24. Lühr, T.; Winkelmann, A.; Nolze, G.; Krull, D.; Westphal, C. Nano Lett. 2016, 16, 31953201. DOI: 10.1021/acs.nanolett.6b00524. 25. Kuznetsov, M. V.; Yashina, L. V.; Sánchez-Barriga, J.; Ogorodnikov, I. I.; Vorokh, A. S.; Volykhov, A. A.; Koch, R. J.; Neudachina, V. S.; Tamm, M. E.; Sirotina, A. P.; Varykhalov, A. Y.; Springholz, G.; Bauer, G.; Riley, J. D.; Rader: O. Phys. Rev. B 2015, 91, 085402. 26. Wider, J.; Baumberger, F.; Sambi, M.; Gotter, R.; Verdini, A.; Bruno, F.; Cvetko, D.; Morgante, A.; Greber, T.; Osterwalder, J. Phys. Rev. Lett. 2001, 86, 2337-2340. 27. Omori, S.; Nihei, Y.; Rotenberg, E.; Denlinger,J. D.; Marchesini, S.; Kevan, S. D.; Tonner, B. P.; Van Hove, M. A; Fadley, C. S. Phys. Rev. Lett. 2002, 88, 055504. 28. Roth, S.; Matsui, F.; Greber, T.; Osterwalder, J. Nano Lett., 2013, 13, 2668–2675 DOI: 10.1021/nl400815w.

ACS Paragon Plus Environment

17

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 26

29. Matsushita, T.; Guo, F. Z.; Matsui, F.; Kato Y.; Daimon, H.; Phys. Rev. B, 75, 085419 (2007). 30. T. Matsushita, F. Z. Guo, M. Suzuki, F. Matsui, H. Daimon and K. Hayashi, Phys. Rev. B: Condens. Matter Mater. Phys. 2008, 78, 144111. 31. Matsushita, T.; Matsui, M.; Daimon H.; Hayashi, K. J. Electron. Spectrosc. Relat. Phenom. 2010, 178-179, 195-220. 32. Matsushita, T.; Matsui, F.; Goto, K.; Matsumoto T.; Daimon, H. J. Phys. Soc. Jpn. 2013, 82, 114005. 33. Matsushita T.; Matsui, F. J. Electron. Spectrosc. Relat. Phenom. 2014, 195, 365-374. 34. Matsushita, T. e-J. Surf. Sci. Nanotech. 2016, 14, 158-160. 35. Matsui, F,; Eguchi, R.; Nishiyama, S.; Izumi, M.; Uesugi, E.; Goto, H.; Matsushita, T.; Sugita, K.; Daimon, H.; Hamamoto, Y.; Hamada, I.; Morikawa, Y.; Kubozono, Y. Scientific Reports 2016, 6, 36258. 36. Hara, T.; Tanaka, T.; Tanabe, T.; Marechel, X. -M.; Kumagai, K.; Kitamura, H. J. Synchrotron Rad. 1998, 5, 426-427. 37. Saitoh, Y.; Kimura, H.; Suzuki, Y.; Nakatani, T.; Matsushita, T.; Muro, T.; Miyahara, T.; Soda, K.; Ueda, S.; Harada, H.; Kotsugi, M.; Sekiyama A.; Suga, S. Rev. Sci. Instrum. 2000, 71, 3254-3259. 38. Morikawa, Y. Phys. Rev. B 1995, 51, 14802. 39. Hamamoto, Y.; Hamada, I.; Inagaki, K.; Morikawa, Y. Phys. Rev. B 2016, 93, 245440.

ACS Paragon Plus Environment

18

Page 19 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

40. Hohenberg, P.; Kohn, W. Phys. Rev. 1964, 136, B864-B871. 41. W. Kohn, W.; Sham, L. J. Phys. Rev. 1965, 140, A1133-A1138. 42. Troullier, N.; Martins, J. L. Phys. Rev. B 1991, 43, 1993-2006. 43. Pehlke, E.; Scheffler, M. Phys. Rev. Lett. 1993, 71, 2338-2341. 44. Winkelmann, A.; Schröter B.; Richter, W. Phys. Rev. B 2004, 69, 245417. 45. Chadi, D. J.; Citrin, P. H.; Park, C. H.; Adler, D. L.; Marcus, M. A.; Gossmann, H.-J., Phs. Rev. Lett. 1997, 79, 4834-4837. 46. Ramamoorthy, M.; Pantelides, S. T. Phys. Rev. Lett. 1996, 76, 4753-4756. 47. Mueller, D. C.; Alonso, E.; Fichtner, W. Pys. Rev. B 2003, 68, 045208.

ACS Paragon Plus Environment

19

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 26

Figure captions

Figure 1 (a) The experimental setup at the BL25SU beamline in the SPring-8 facility. The angle between the direction to the electron analyzer (DA30) and the incident soft X-ray beam was 90°. (b) A schematic diagram summarizing the principles of photoelectron holography.

Figure 2 (a) Depth profiles of As concentrations (as determined using SIMS) in samples after spike-RTA and after spike-RTA followed by a second RTA. Analyses reported herein were carried out using the latter sample. (b) Depth profiles of As and carrier concentrations after spike-RTA followed by a second RTA.

Figure 3 (a) A Si 2p photoelectron hologram acquired at a kinetic energy (Ek) of 600 eV, and (b) reconstructed atomic images for various cross-sections having different z values.

Figure 4 Holograms generated from the spectra labeled (a) BEH, (b) BEM, and (c) BEL, and (d) As 3d core-level photoelectron spectra with labels.

Figure 5 Atomic images reconstructed from the hologram of each atomic site. The corresponding BEH and BEM spectra are shown in Fig. 4. The BEH cross-sections are labeled as z =(a) 0, (b) 0.135 (=a/4), and (c) 0.27 nm (=a/2). The candidate structure is shown in (d). The BEM cross-

ACS Paragon Plus Environment

20

Page 21 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

sections and candidate structure are shown in (e), (f), (g), and (h) in the same manner. In the structural images, (d) and (h), emitter As atoms are red and labeled “Emitter As”, notable Si atoms and vacancies near to the emitters are blue and brown, respectively, and fluctuating atoms appear blurry.

Table title

Table 1 Dopant structures and binding energy shifts. (NB: neighbor, V: vacancy)

ACS Paragon Plus Environment

21

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 26

Figure 1 (a) The experimental setup at the BL25SU beamline in the SPring-8 facility. The angle between the direction to the electron analyzer (DA30) and the incident soft X-ray beam was 90°. (b) A schematic diagram summarizing the principles of photoelectron holography.

ACS Paragon Plus Environment

22

Page 23 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 2 (a) Depth profiles of As concentrations (as determined using SIMS) in samples after spike-RTA and after spike-RTA followed by a second RTA. Analyses reported herein were carried out using the latter sample. (b) Depth profiles of As and carrier concentrations after spike-RTA followed by a second RTA.

Figure 3 (a) A Si 2p photoelectron hologram acquired at a kinetic energy (Ek) of 600eV, and (b) reconstructed atomic images for various cross-sections having different z values.

ACS Paragon Plus Environment

23

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 26

Figure 4 Holograms generated from the spectra labeled (a) BEH, (b) BEM and (c) BEL, and (d) As 3d core-level photoelectron spectra with labels.

ACS Paragon Plus Environment

24

Page 25 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 5 Atomic images reconstructed from the hologram of each atomic site. The corresponding BEH and BEM spectra are shown in Fig. 4. The BEH cross-sections are labeled as z =(a) 0, (b) 0.135 (=a/4), and (c) 0.27 nm (=a/2). The candidate structure is shown in (d). The BEM crosssections and candidate structure are shown in (e), (f), (g) and (h) in the same manner. In the structural images, (d) and (h), emitter As atoms are red and labeled “Emitter As,” notable Si atoms and vacancies near to the emitters are blue and brown, respectively, and fluctuating atoms appear blurry.

Table 1. Dopant structures and binding energy shifts. (NB: neighbor, V: vacancy) Structure at site occupied by As

Binding energy shift / eV

Substitutional As

As

0.0

As pair at first NB w/o V

As-As

-0.59

As pair at second NB w/o V

As-Si-As

+0.01

As pair at third NB w/o V

As-Si-Si-As

-0.06

As and V pair at first NB

As-V

-1.05

As and V pair at second NB

As-Si-V

-0.48

As pair at second NB with central V

As-V-As

-0.94

As trimer with neighboring V

As3-V

-0.99

As tetramer with neighboring V

As4-V

-0.68

ACS Paragon Plus Environment

25

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 26

Graphic for TOC

ACS Paragon Plus Environment

26