Irreparable Defects Produced by the Patching of h ... - ACS Publications

Apr 10, 2017 - monolayer on Re(0001), the patching frontiers of different domains can be ... patched up seamlessly, what kinds of defects can be intro...
0 downloads 0 Views 2MB Size
Article pubs.acs.org/JACS

Irreparable Defects Produced by the Patching of h‑BN Frontiers on Strongly Interacting Re(0001) and Their Electronic Properties Yue Qi,† Zhepeng Zhang,† Bing Deng,† Xiebo Zhou,‡ Qiucheng Li,† Min Hong,‡ Yuanchang Li,*,§ Zhongfan Liu,*,† and Yanfeng Zhang*,†,‡ †

Center for Nanochemistry (CNC), Beijing National Laboratory for Molecular Sciences, College of Chemistry and Molecular Engineering, Academy for Advanced Interdisciplinary Studies, Peking University, Beijing 100871, People’s Republic of China ‡ Department of Materials Science and Engineering, College of Engineering, Peking University, Beijing 100871, People’s Republic of China § National Center for Nanoscience and Technology, Chinese Academy of Sciences, Beijing 100190, People’ s Republic of China S Supporting Information *

ABSTRACT: Clarifying the origin and the electronic properties of defects in materials is crucial since the mechanical, electronic and magnetic properties can be tuned by defects. Herein, we find that, for the growth of h-BN monolayer on Re(0001), the patching frontiers of different domains can be classified into three types, i.e., the patching of B- and N-terminated (B|Nterminated) frontiers, B|B-terminated frontiers and N|N-terminated frontiers, which introduce three types of defects, i.e., the “heart” shaped moiré-level defect, the nonbonded and bonded line defects, respectively. These defects were found to bring significant modulations to the electronic properties of hBN, by introducing band gap reductions and in-gap states, comparing with perfect h-BN on Re(0001) with a band gap of ∼3.7 eV. The intrinsic binary composition nature of h-BN and the strong h-BN-Re(0001) interaction are proposed to be cooperatively responsible for the formation of these three types of defects. The former one provides different types of h-BN frontiers for domain patching. And the later one induces multinucleation but aligned growth of h-BN domains on Re(0001), thus precluding their subsequent coalescence to some extent. This work offers a deep insight into the categories of defects introduced from the patching growth of two-dimensional layered materials, as well as their electronic property modulation through the defect engineering.



INTRODUCTION

still challenging under the traditional atmospheric pressure or low pressure CVD methods on the highly corrugated Cu foils. The growth of h-BN layers under ultrahigh vacuum (UHV) conditions on some single crystal substrates like Ni(111),27 Cu(111),28,29 Ru(0001),30,31 Ir(111),32 and Rh(111)24,33,34 provided more ideal platforms for exploring the growth mechanism, morphology and electronic property, as well as defect, with the aid of scanning tunneling microscope/ spectroscopy (STM/STS). The point defects featured with 5| 7 rings and grain boundaries (GBs) characterized with an array of dislocations were well characterized. These defects showed significant effect on the electronic, thermal and mechanical properties of h-BN, for example, enhancing the electronic scattering,35 and affecting detrimentally the device performance.8,23 In this regard, a fundamental understanding of such defects is critical for the applications of h-BN. Theoretical studies have revealed that the defects in h-BN, such as point defects,36 triangular-shaped void defects37 and

Two-dimensional (2D) monolayer hexagonal boron nitride (hBN), a structural analogue to graphene, has attracted great research attentions because of its great potential for numerous applications in far-ultraviolet light-emitting devices,1 oxidationresistant coatings,2,3 transparent electronics4 and dielectric layers,5,6 owing to its unique band structure, good thermal conductivity, outstanding chemical inertness and excellent transparency, etc.1,7−10 In addition, its combination with graphene to form in-plane11−15 or vertically stacked16−19 heterostructures have manifested great potentials in developing functional graphene-based devices.20−22 Chemical vapor deposition (CVD) is a promising approach for large-scale, high-quality growth of h-BN on metal substrates like Cu foils, thanks to the ease of operation and scalability.8,23−25 However, the quality of h-BN can be affected by many factors, such as the dosage of precursor, growth temperature, and even surface property of substrate.23,26 Hereby, in-depth explorations of the growth behavior and the resulted morphology and electronic property of h-BN layers are © 2017 American Chemical Society

Received: January 26, 2017 Published: April 10, 2017 5849

DOI: 10.1021/jacs.7b00647 J. Am. Chem. Soc. 2017, 139, 5849−5856

Article

Journal of the American Chemical Society

Figure 1. Growth of h-BN on Re(0001) under ultrahigh vacuum conditions. (a) Schematic of h-BN growth on Re(0001). (b, c) Large-area STM images of h-BN superstructure with a period of ∼3.00 nm. (d) High-resolution STM image showing a coincidence lattice of 12 × 12 B−N/11 × 11 Re(0001). The edge direction of h-BN domains (black arrow in (b)) and the zigzag orientation of h-BN atomic lattices (black arrow in (d)) are well aligned with the direction of the moiré pattern (green arrows in (b) and (d)). (d, e, f) STM images for h-BN on Re(0001) captured under different scanning conditions. Scanning conditions: (b) 1.50 V, 0.40 nA; (c) −0.01 V, 6.09 nA; (d) −0.01 V, 13.47 nA; (e) −0.01 V, 17.37 nA; (f) 0.01 V, 32.79 nA.

line defects,28,35,38 can bring some beneficial effects to the materials property. For example, the one- and three-boron defects in h-BN can induce interesting paramagnetic properties.39−41 Moreover, the defects can introduce band gap reduction of h-BN and create optical peaks in the visible light range, stimulating potential applications in h-BN electronics.42,43 Briefly, defects, if controlled in a proper manner, can be a promising strategy to engineer the magnetic and electronic properties of materials toward novel applications. However, in the patching growth of h-BN, due to the existence of both B- and N-terminated frontiers, many fundamental issues are not well understood at the patching interfaces. For instance, whether various patching frontiers, including B|N-, B|B- and N|N-terminated frontiers, can be patched up seamlessly, what kinds of defects can be introduced, and how about the structural and electronic properties? In order to solve such issues, we choose an ideal h-BN/ Re(0001) system for its capability of providing both B- and Nterminated domains concurrently, which is suitable for exploring the patching possibility of B|N-, B|B- and N|Nterminated frontiers. According to STM observations, three types of defects were introduced, i.e., the “heart” shaped moiré defect, the nonbonded and bonded line defects, respectively. And their effects on the electronic properties of h-BN were also examined with STS investigations. In this regard, this work provides the first experimental demonstration in synthesizing hBN on Re(0001), and in characterizing the atomic-scale structure, defects and various patching interfaces in a monolayer film.



chamber through opening a leak valve.23,25,44 Figure 1a shows the schematic diagram of the CVD growth of h-BN on Re(0001) under UHV conditions. The corresponding experimental setup and growth details are also addressed in Figure S1. The clean Re(0001) surface before h-BN growth is also shown in Figure S2a with its very flat STM morphology. The growth of h-BN was realized by exposing ∼650 °C heated Re(0001) substrate to borazine (BHNH)3 for ∼20 min, which was dosed under a pressure change from ∼1 × 10−10 mbar to ∼5 × 10−8 mbar of the UHV condition. As a result, hBN superstructure patterns were achieved to be mixed with irregular or even “broken” networks, as shown in the STM image in Figure S3a. In order to improve the quality of h-BN layer, a two-step growth method was applied, i.e., lowtemperature adsorption and growth (∼650 °C for ∼20 min), followed with high-temperature annealing (∼730 °C for ∼40 min), as schematically shown in Figure S2c. High-quality h-BN domains were thus achieved on Re(0001) (Figure 1b and c) showing a periodic “nanomesh-like” superstructure, with the morphology similar to that of h-BN on Rh(111)13 and Ru(0001).45 The superstructure pattern is characterized with a fixed period of ∼3.00 nm. The atomically resolved STM image is presented in Figure 1d, to show perfect h-BN lattices. Figure 1d, e and f present the STM images of h-BN moiré collected under different scanning conditions. The different STM contrasts indicate that the moiré structure displays strong electronic effects, as similarly reported for graphene on Ru(0001).46 According to the published references,13,47 the different spatial occupations of B, N atoms on single crystal surfaces (like Rh(111)) can induce disparate interface interactions, thus implying additional influence on the STM imaging of h-BN on metals. In return, the different contrasts inside a moiré for h-BN/Re(0001) should indicate different interface interactions. On the basis of the atomically resolved STM images, the h-BN superstructure/moiré on Re(0001) can

RESULTS AND DISCUSSION

h-BN was synthesized with borazine ((BHNH)3) as precursor, which was produced by heating (at ∼150 °C) the solid powder of ammonia borane (NH3−BH3) (kept in a stainless cylinder connected to the main chamber), and introduced into the UHV 5850

DOI: 10.1021/jacs.7b00647 J. Am. Chem. Soc. 2017, 139, 5849−5856

Article

Journal of the American Chemical Society

Figure 2. Strong interfacial interaction of h-BN/Re(0001) and the patching behavior of different h-BN domains. (a) STM image of h-BN on Re(0001). (b) DFT-calculated contour map showing a coincidence lattice of 12 × 12 B−N/11 × 11 Re(0001) (with vdW interactions). (c) Experimental and DFT calculated surface corrugations of h-BN along the dashed line in (a) (∼1.64 Å and ∼1.52 Å, black and red lines, respectively). (d) Calculated binding energy (Eb) for h-BN on Re(0001) (∼0.59 eV/per Re atom). (e) STM morphology of two aligned h-BN domains (marked with green arrows) growing from different nuclei. (f) Nearly full coverage monolayer h-BN on Re(0001) with defects appearing at the patching interface. (g) Schematics for the patching growth behavior. The white rectangles in the bottom schematic indicate the B|N-, B|B-, and N|Nterminated interfaces. Scanning conditions: (a) 0.08 V, 4.57 nA; (e) 2.00 V, 0.50 nA; (f) −1.03 V, 2.27 nA.

be defined to arise from the coincidence lattice of (12 × 12) B−N/(11 × 11) Re(0001). This can be illustrated from the following equation, D=

corrugation may vary with different DFT methods.48 This high surface undulation for h-BN/Re(0001), in return, confirms the strong interface interaction. Moreover, DFT calculations for the binding energy (Eb) of hBN on metals (Figure 2d) were also carried out to illustrate the interfacial interaction with the following formula,

(1 + δ)a 2(1 + δ)(1 − cos θ ) + δ 2

where D, θ, a and δ denote the moiré period of h-BN, the rotation angle between h-BN and Re(0001) lattices, the lattice constant of h-BN (∼2.50 Å) and the lattice mismatch (∼10.07%) between h-BN and Re(0001), respectively. The calculated rotation angle between h-BN and Re(0001) lattices is 0°, highly suggestive the epitaxial growth behavior. Notably, the edge direction of h-BN domain (black arrow in Figure 1b) and the zigzag orientation of h-BN atomic lattice (black arrow in Figure 1d) align well with the direction of the unit cell of the moiré pattern (green arrows in Figure 1b and d). This coincidence indicates a preferred zigzag-type edge structure of the monolayer h-BN domains. In order to achieve a deep understanding of the moiré superstructure, DFT calculation was performed. In the STM image (Figure 2a) and the DFT-calculated contour map based on the coincidence lattice of 12 × 12 B−N/11 × 11 Re(0001) (Figure 2b), the corrugation on h-BN surface is obvious. Corresponding height profile (black line in Figure 2c) captured across the moiré pores and wires in the STM image (marked by green line) reveals a corrugation value of ∼1.64 Å (scanning condition: 0.08 V, 4.57 nA). This result is consistent with the calculated value (∼1.52 Å) (red line in Figure 2c), as extracted from the line profile of the DFT-calculated contour map in Figure 2b (details for DFT calculations are shown in the Supporting Information). This calculated value is much larger than that of h-BN on Rh(111) (∼1.24 Å) calculated with the same DFT method,13 a more strongly coupled substrate comparing with that of Ir and Cu.14,28 Note that, the calculated

Eb =

[E(layer) + E(sub) − E(layer‐sub)] n

where E(layer), E(sub) and E(layer-sub) denote the total energies of the h-BN overlayer, the Re substrate, and the combined system, respectively. n represents the number of atoms in the substrate unit cell. As a result, the calculated Eb on Re(0001) is ∼0.59 eV, much higher than that of h-BN on Rh(111) (∼0.37 eV)13 and Ir(111) (∼0.39 eV),14 indicating the strong interface interaction of h-BN/Re(0001). Further STM investigations of the growth behavior manifest that, the growth of h-BN can be significantly modulated by the strong h-BN-Re(0001) interaction, featured with high nucleation density at the initial growth stage, as compared with that on Ir(111)14 at nearly the same growth condition (around 700 °C and 5 × 10−8 mbar for NH3−BH3 dosage). Later on, discrete small h-BN domains evolved, as shown in Figure 2e and Figure S4a (with coverage of ∼60%). In spite of this, the discrete h-BN domains present exactly the same orientation with each other (marked by green arrows), thus indicating the epitaxial growth feature of h-BN on Re(0001). When the coverage of h-BN approaches to a full layer (Figure 2f), the discrete h-BN domains begin to patch up with each other to form larger domains. Meanwhile, many defects occur at the patching interface, as marked in Figure 2f. This patching process is schematically shown in Figure 2g. It is proposed that the strong h-BN-Re(0001) interaction enhances the difficulty for the coalescence of different h-BN domains into 5851

DOI: 10.1021/jacs.7b00647 J. Am. Chem. Soc. 2017, 139, 5849−5856

Article

Journal of the American Chemical Society

Figure 3. Structural and electronic properties of a “heart” shaped superstructure produced by the patching of B|N-terminated frontiers in h-BN. (a) Schematic for the patching of B|N-terminated frontiers. (b) “Heart” shaped defects (marked by black arrows) survived after sample annealing at ∼730 °C for ∼50 min. The misalignment with less than one intact moiré unit exists between two or more adjacent h-BN domains, as marked by the green, red and blue dashed lines in domains “A”, “B” and “C”. (c) Simulation of the “heart” shaped pattern evolving from the overlapping and intersecting of two nanomesh “pores” with an angle of 60°. (d) Magnified STM image of the “heart” shaped superstructure. High-resolution STM image as an inset in (c) shows a 5|7 pair-like defect. (e) Schematic of the 5|7 pair-like defect. (f) STS curves in the vicinity of the “heart” shaped pattern. Curves labeled from “1−7” corresponding to positions “1−7” in (d), showing a ∼1.0 eV band gap reduction from ∼3.7 eV at positions “2” and “6” to ∼2.7 eV at position “4”. (g) Summary of the band gap in (f). Error bars of band gap in (f) (short red lines) and (g) are 0.15 eV. Scanning conditions: (b) 0.39 V, 9.20 nA; (d) −0.01 V, 6.09 nA; inset image in (d): −0.01 V, 3.03 nA; STS spectra: Vt = 1.50 V, It = 0.60 nA, Vrms = 10 mV, f = 932 Hz, T = 78 K.

For h-BN on Re(0001), in the patching growth of B|Nterminated frontiers (as schematically shown in Figure 3a), the probability for their perfect patching is still low. This is because, for h-BN on Re(0001), each nanomesh unit cell covers 11 × 11 unit cells of the underlying Re(0001) lattice, leading to 121 unique adsorption configurations. The probability for the coherent patching of two h-BN domains will be as low as 1/ 121, analogous to the analysis for h-BN on Ru(0001).30 As an example, the precise nucleation site for one h-BN domain to merge perfectly with another one is also presented in Figure S4e. In this regard, the translational misalignment between the neighboring h-BN domains is inevitable, commonly resulting in the formation of twisted or stretched moirés at the patching interfaces (marked by green arrows in Figure S5a and b). Representatively, many “heart” shaped moirés occur in the vicinity of the interface (marked by black arrows in Figure S5b). Notably, when the annealing time was extended to ∼180 min and the growth temperature is set at ∼780 °C, the twisted or stretched moiré defects could be partially healed, only with the “heart” shaped defects survived (marked by black arrows in Figure S6a). This implies, to some extent, the high stability of this “heart” shaped defect. In the vicinity of the “heart” shaped moirés, the translational misalignment featured with less than one intact moiré unit could be noticed, as marked by the green, red and blue lines in domains “A”, “B” and “C” in Figure 3b. The occurrence of the “heart” shaped moiré is proposed to be the way to release the strains induced by these misalignments. The schematic in

a perfect monolayer. And a lot of defects or defect lines could be evolved along the patching interface. In comparison with that of graphene, the patching of different h-BN domains are more complicated due to its binary composition nature, and the B- and N-terminated patching frontiers could be evolved concurrently. According to the published references, the shapes of h-BN domains could vary from regular triangles23,49 to truncated triangles,50 and even hexagons.51 For the regular triangle-shaped h-BN domains, for example for h-BN domains on Ru(0001),30 the angle between the adjacent edges is 60°, which implies the existence of only one type atom at the domain edge (i.e., either B or N), as schematically shown in Figure S4c. However, in the current system of h-BN on Re(0001), most of the angles between adjacent edges of h-BN domains are 120° (Figure 2e, 3b and Figure S4b), which address the coexistence of B- and N-terminated edges, as schematically shown in Figure S4d. This may be caused by the unique UHV-CVD growth conditions, for example, the partial pressure and annealing temperature, and all these factors affect the shape of h-BN domains.50,52−54 In this case, the patching growth of B|N-, B|Band N|N-terminated frontiers should all exist, as marked by the white rectangles in the bottom schematic in Figure 2g. Meanwhile, the translational misalignment, the orientation deviation, and the encounter of the same-atom-terminated patching frontiers should also occur, thus disturbing the perfect patching of different domains. 5852

DOI: 10.1021/jacs.7b00647 J. Am. Chem. Soc. 2017, 139, 5849−5856

Article

Journal of the American Chemical Society

Figure 4. Nonbonded line defects produced by the patching of B|B-terminated frontiers in h-BN. (a, b) Schematic and DFT calculations for the patching of B|B-terminated frontiers. A “gap” of ∼1.44 Å formed at the interface. (c) STM image of the staggered interface at moiré scale. The “pore” region of one domain faces the “ridge” region of the other one, and a half-moiré translational misalignment evolved between the neighboring h-BN domains. (d) Schematic of the staggered and nonbonded line defect. (e) High-resolution STM image of the nonbonded line defect. A “gap” with a width of 1.54 Å ∼ 4.81 Å sits between two adjacent h-BN domains marked by black lines. (f) STS curves in the vicinity of the line defect. Curves “1− 6” were recorded from positions “1−6” in (e), showing a ∼0.4 eV band gap reduction from ∼3.5 eV at positions“1” and “6” to ∼3.1 eV at positions “3” and “4”. (g) Summary of the band gap variations in (f). Error bars of band gap in (f) (short red lines) and (g) are 0.15 eV. Scanning conditions: (c): −0.01 V, 3.03 nA; (e) −0.01 V, 3.78 nA; STS spectra: Vt = 1.50 V, It = 0.60 nA, Vrms = 10 mV, f = 932 Hz, T = 78 K.

Figure 3c indicates the origin of the “heart” shaped pattern, evolving from the overlapping and intersecting of two nanomesh “pores” at an angle of 60° (marked by the dashed lines in Figure 3c and d). For more details, even high-resolution STM image of one “heart” shaped moiré was captured as an inset in Figure 3d, and a 5|7 pair-like defect can be resolved with the fitted blue and red spheres (Figure 3e), which is well consistent with the reported 5|7 pair-like defect embedded in the “heart” shaped moiré in h-BN domains on Ru(0001).45 However, further detailed elemental identity at the 5|7 pair-like defect is challenging. For the published work of h-BN on Ni(111),27 B and N sites have been assigned based on the DFT calculations of the local charge densities at the Fermi energy. And the N atoms were usually imaged by STM. As a result, for the patching of B|N-terminated frontiers in h-BN on Re(0001), the perfect patching probability is very low, due to the large periodicity of the moiré superstructure and the possible formation of translational misfit. The strong interface interaction should also enhance the difficulty for the perfect coalescence. Previous theoretical and experimental studies have pointed out that, various defects can be evolved to modify the electronic property of h-BN monolayer.28,35 For example, the grain boundary (GBs) composed of 5|7 and 4|8 defects could bring specific in-gap peaks and band gap reductions. In this work, corresponding dI/dV measurements were also performed to achieve the local electronic property of h-BN. Figure 3f shows the STS curves labeled with “1−7” captured from positions “1− 7” marked in Figure 3d, on and far away from the “heart” shaped moiré (within two moiré patterns). The STS curve of

the bare Re(0001) substrate is also displayed in Figure S7 as a reference, which is consistent with the reported STS data of the bare Re(0001).55 The “pore” and “ridge” positions in Figure 3f correspond to the white and black dots in Figure 3d, respectively. At the perfect moiré positions, two prominent features can be noticed from their STS curves, i.e., the reduced band gap and n-doping character. Specifically, the valence band maximum (VBM) and conduction band minimum (CBM) at the “ridge” positions are located at ∼ −2.6 eV and ∼1.1 eV, respectively, corresponding to a band gap of ∼3.7 eV. For the “pore” positions, the VBM and CBM are ∼ −2.4 eV and ∼1.1 eV, respectively, giving a band gap of ∼3.5 eV. These band gap reductions are considered to be mainly mediated by the electron donation from the Re substrate or the strong adlayersubstrate interaction, as similarly reported for h-BN doped by the underlying Ir(111) substrate.14,56 At the “heart” shaped moiré (position “4”), a band gap reduction of ∼1.0 eV is also observed, possibly induced by the 5|7 type defect, as compared with positions “2” and “6”, (VBM shifts from ∼ −2.6 eV upward to ∼ −1.7 eV and CBM shifts from ∼1.1 eV downward to ∼1.0 eV). Briefly, the “heart” shaped defects provides significant modulations on the local electronic properties of h-BN on Re(0001). For more details, a summary of the band gaps at different locations is shown in Figure 3g. Except for the patching of the ideal B|N-terminated frontiers, the patching of the same-atom-terminated frontiers, like B|Band N|N-terminated frontiers, could produce one-dimensional (1D), long-range, straight line-shape defects for h-BN on 5853

DOI: 10.1021/jacs.7b00647 J. Am. Chem. Soc. 2017, 139, 5849−5856

Article

Journal of the American Chemical Society

Figure 5. Bonded line defects produced by the patching of N|N-terminated frontiers in h-BN. (a, b) Schematic image and DFT calculation for the patched frontier, respectively. A continuous boundary is calculated to form at the patching interface. (c) STM image of the moiré-scale staggered boundary (marked by black and white arrows). (d) Magnified STM image corresponding to the area marked in (c). At the staggered region, the moiré “pore” of one domain faces the “ridge” of the other one. A half-moiré translational misalignment evolves between the neighboring h-BN domains. (e) High-resolution STM image showing the continuous interface between adjacent domains with the same orientation (black arrows). (f) STS curves at and far away from the line defect. Curves “0−9” were collected from positions “0−9” in (d), showing a ∼0.4 eV band gap reduction from ∼3.7 eV at positions “0” and “8” to ∼3.3 eV at position “4”. Some in-gap peaks appear in curve “5” at the exact line defect region (position “5”). (g) Summary of the band gap in (f). (h) Electron density of states map of N−N and B−B interfaces. Error bars of band gap in (f) (short red lines) and (g) are 0.15 eV. Scanning conditions: (c, d) 1.00 V, 1.33 nA; (e) −0.01 V, 3.78 nA; STS spectra: Vt = 1.50 V, It = 0.60 nA, Vrms = 10 mV, f = 932 Hz, T = 78 K.

6” corresponding to positions “1−6” in Figure 4e. Notably, a band gap reduction from ∼3.5 eV (position “1” or “6”) to ∼3.1 eV (position “3” or “4”) occurs at the domain edges. Accordingly, the VBM shifts from ∼ −2.4 eV upward to ∼ −2.1 eV and CBM shifts from ∼1.1 eV downward to ∼1.0 eV, respectively. This band gap reduction is possibly due to the appearance of strong edge states. This fact was also demonstrated in the theoretical calculation of h-BN nanoribbons, a 0.7 eV band gap reduction was predicted due to the presence of strong edge states.57 However, the STS data at the exact “gap” are not repeatable at different lateral positions, since the “gap” region is easy to be contaminated by the adsorbed impurities. Figure 4g is a summary of the band gap variations around the B|B-terminated frontiers. The last patching possibility in the growth of h-BN is the encounter of N|N-terminated frontiers, as schematically presented in Figure 5a. DFT calculations were also performed for judging the patching possibility (Figure 5b). Different from the case of B|B-terminated frontiers, the theoretical result indicates that, a continuous interface featured with N−N bonds can evolve between the neighboring h-BN domains with N|Nterminated frontiers. In the experiments, this continuous interface was indeed observed from the STM images, as featured with staggered moirés connecting neighboring h-BN domains possessing an aligned direction, as marked by black arrows in Figure 5c and d. A straight 1D line defect can be

Re(0001), endowing different structural and electronic properties around the line-shape defects. Figure 4a and b are the schematic and the DFT calculations for illustrating the patching possibility of B|B-terminated frontiers, respectively. According to the DFT calculations, the B|B-terminated frontiers cannot be patched up with each other, usually leaving a “gap” with a width of ∼1.44 Å between the component domains. As an example, STM image in Figure 4c shows the formation of such a 1D, straight line-shape defect siting between two h-BN domains, with a length of ∼25 nm. According to the STM image, the adjacent domains have a translational misalignment, leading to an offset of a half moiré along the interface. Namely, the “pore” region of one h-BN domain faces the “wire” region of the other one, as schematically shown in Figure 4d. Notably, in the magnified STM image of the staggered interface in Figure 4e, a “gap” with a width of 1.54 Å ∼ 4.81 Å can be noticed between the neighboring h-BN domains, as in agreement with the DFT calculated interface with no B−B bonds formation in Figure 4b. In this regard, it can be inferred that, the patching of B|Bterminated frontiers in h-BN could introduce long-range, nonbonded 1D line defect between the neighboring h-BN domains. This conclusion is further confirmed by the patching of N|N-terminated frontiers discussed in latter parts. The electronic properties of the B|B-terminated interfaces were also characterized with STS in Figure 4f, with curves “1− 5854

DOI: 10.1021/jacs.7b00647 J. Am. Chem. Soc. 2017, 139, 5849−5856

Journal of the American Chemical Society



noticed with a length of ∼20 nm. For more details, the STM images in Figure 5d and Supplementary Figure S8a reveal the obvious translational misalignment of the neighboring domains by less than one moiré unit. The difference between this image and Figure 4e lies in the appearance of a brighter striped contrast at the interface with regard to the neighboring region, and continuous atom contrast can be observed at the patching interface (Figure 5e and Supplementary Figure S8b). This is well consistent with the theoretical calculations in Figure 5b that, N−N bonds could be formed in the interface of h-BN domains with N|N-terminated frontiers. The electronic property variation around such kind of interface was also revealed by STM/STS measurements. Figure 5f shows the STS curves labeled with “0−9” at and far away from the line defect (within a distance of three moiré patterns), corresponding to positions “0−9” in Figure 5d, respectively. Apparently, the band gap reduction is obvious around the line defect regions. For example, at “wire” regions, the band gap changes from ∼3.7 eV at positions “0” and “8” to ∼3.3 eV at position “4”. VBM shifts from ∼ −2.6 eV upward to ∼ −2.2 eV and CBM shifts not obviously. Notably, the STS curve of position “5” (right at the line defect) displays several in-gap peaks, probably in line with the electronic states arising from the N−N interface. A summary of the band gaps at different positions is shown in Figure 5g. Notably, the different patching behaviors of N−N and B−B interfaces can also be explained from their disparate interface electron density of states, as shown in Figure 5h (green color represents the charge density isosurface = 0.9 e/Å3). Electron accumulation and depletion occur at the N−N and B−B interfaces, respectively, probably corresponding to the formation of the continuous and disconnected patching interfaces. The electron accumulation at the N−N interface should induce the enhanced contrast at the domain boundary in Figure 5e and Supplementary Figure S8b. Notably, the line defects caused by the patching of the same-atom-terminated frontiers can survive from sample annealing at ∼780 °C for ∼180 min. That means, once formed, the line-shape defects are irreparable. In summary, we have explored the three typical patching interfaces in the growth of h-BN toward a monolayer, i.e., the patching of B|N-, B|B- and N|N-terminated frontiers. The three cases introduce three types of defects, i.e., the “heart” shaped moiré defect, the nonbonded and bonded line defects, respectively. These defects were detected to imply significant modulations on the electronic properties of h-BN, by introducing band gap reductions and specific in-gap peaks. The binary composition nature of h-BN and the strong h-BNRe(0001) interaction are considered to be responsible for the formation of these defects. Briefly, this work offers a deep insight into the patching growth of h-BN domains, the resulted versatile defects and the electronic property modulations via defects engineering.



Article

AUTHOR INFORMATION

Corresponding Authors

*[email protected] *zfl[email protected] *[email protected] ORCID

Zhongfan Liu: 0000-0003-0065-7988 Yanfeng Zhang: 0000-0003-1319-3270 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The work was supported by the National Key Research and Development Program of China (2016YFA0200103), the Beijing Municipal Science and Technology Commission (No. Z161100002116020), the Open Research Fund Program of the State Key Laboratory of Low-Dimensional Quantum Physics (No. KF201601), the National Natural Science Foundation of China (Nos. 51290272, 51472008, 51432002, 50121091 and 21201012) and the National Basic Research Program of China (Nos. 2013CB932603, 2014CB921002).



REFERENCES

(1) Watanabe, K.; Taniguchi, T.; Kanda, H. Nat. Mater. 2004, 3, 404−409. (2) Liu, Z.; Gong, Y.; Zhou, W.; Ma, L.; Yu, J.; Idrobo, J. C.; Jung, J.; MacDonald, A. H.; Vajtai, R.; Lou, J.; Ajayan, P. M. Nat. Commun. 2013, 4, 2541. (3) Li, L. H.; Cervenka, J.; Watanabe, K.; Taniguchi, T.; Chen, Y. ACS Nano 2014, 8, 1457−1462. (4) Lee, G.-H.; Yu, Y.-J.; Cui, X.; Petrone, N.; Lee, C.-H.; Choi, M. S.; Lee, D.-Y.; Lee, C.; Yoo, W. J.; Watanabe, K.; Taniguchi, T.; Nuckolls, C.; Kim, P.; Hone, J. ACS Nano 2013, 7, 7931−7936. (5) Decker, R.; Wang, Y.; Brar, V. W.; Regan, W.; Tsai, H.-Z.; Wu, Q.; Gannett, W.; Zettl, A.; Crommie, M. F. Nano Lett. 2011, 11, 2291−2295. (6) Xue, J.; Sanchez-Yamagishi, J.; Bulmash, D.; Jacquod, P.; Deshpande, A.; Watanabe, K.; Taniguchi, T.; Jarillo-Herrero, P.; LeRoy, B. J. Nat. Mater. 2011, 10, 282−285. (7) Balandin, A. A.; Ghosh, S.; Bao, W.; Calizo, I.; Teweldebrhan, D.; Miao, F.; Lau, C. N. Nano Lett. 2008, 8, 902−907. (8) Song, L.; Ci, L.; Lu, H.; Sorokin, P. B.; Jin, C.; Ni, J.; Kvashnin, A. G.; Kvashnin, D. G.; Lou, J.; Yakobson, B. I.; Ajayan, P. M. Nano Lett. 2010, 10, 3209−3215. (9) Chen, Y.; Zou, J.; Campbell, S. J.; Le Caer, G. Appl. Phys. Lett. 2004, 84, 2430−2432. (10) Kubota, Y.; Watanabe, K.; Tsuda, O.; Taniguchi, T. Science 2007, 317, 932−934. (11) Ci, L.; Song, L.; Jin, C.; Jariwala, D.; Wu, D.; Li, Y.; Srivastava, A.; Wang, Z. F.; Storr, K.; Balicas, L.; Liu, F.; Ajayan, P. M. Nat. Mater. 2010, 9, 430−435. (12) Sutter, P.; Cortes, R.; Lahiri, J.; Sutter, E. Nano Lett. 2012, 12, 4869−4874. (13) Gao, Y.; Zhang, Y.; Chen, P.; Li, Y.; Liu, M.; Gao, T.; Ma, D.; Chen, Y.; Cheng, Z.; Qiu, X.; Duan, W.; Liu, Z. Nano Lett. 2013, 13, 3439−3443. (14) Liu, M.; Li, Y.; Chen, P.; Sun, J.; Ma, D.; Li, Q.; Gao, T.; Gao, Y.; Cheng, Z.; Qiu, X.; Fang, Y.; Zhang, Y.; Liu, Z. Nano Lett. 2014, 14, 6342−6347. (15) Kim, S. M.; Hsu, A.; Araujo, P. T.; Lee, Y.-H.; Palacios, T.; Dresselhaus, M.; Idrobo, J.-C.; Kim, K. K.; Kong, J. Nano Lett. 2013, 13, 933−941. (16) Yang, W.; Chen, G.; Shi, Z.; Liu, C.-C.; Zhang, L.; Xie, G.; Cheng, M.; Wang, D.; Yang, R.; Shi, D.; Watanabe, K.; Taniguchi, T.; Yao, Y.; Zhang, Y.; Zhang, G. Nat. Mater. 2013, 12, 792−797.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b00647. Detailed growth and other STM characterizations of hBN (PDF) 5855

DOI: 10.1021/jacs.7b00647 J. Am. Chem. Soc. 2017, 139, 5849−5856

Article

Journal of the American Chemical Society (17) Roth, S.; Matsui, F.; Greber, T.; Osterwalder, J. Nano Lett. 2013, 13, 2668−2675. (18) Gao, T.; Song, X.; Du, H.; Nie, Y.; Chen, Y.; Ji, Q.; Sun, J.; Yang, Y.; Zhang, Y.; Liu, Z. Nat. Commun. 2015, 6, 6835. (19) Zhang, C.; Zhao, S.; Jin, C.; Koh, A. L.; Zhou, Y.; Xu, W.; Li, Q.; Xiong, Q.; Peng, H.; Liu, Z. Nat. Commun. 2015, 6, 6519. (20) Dean, C. R.; Young, A. F.; Meric, I.; Lee, C.; Wang, L.; Sorgenfrei, S.; Watanabe, K.; Taniguchi, T.; Kim, P.; Shepard, K. L.; Hone, J. Nat. Nanotechnol. 2010, 5, 722−726. (21) Levendorf, M. P.; Kim, C.-J.; Brown, L.; Huang, P. Y.; Havener, R. W.; Muller, D. A.; Park, J. Nature 2012, 488, 627−632. (22) Britnell, L.; Gorbachev, R. V.; Jalil, R.; Belle, B. D.; Schedin, F.; Mishchenko, A.; Georgiou, T.; Katsnelson, M. I.; Eaves, L.; Morozov, S. V.; Peres, N. M. R.; Leist, J.; Geim, A. K.; Novoselov, K. S.; Ponomarenko, L. A. Science 2012, 335, 947−950. (23) Kim, K. K.; Hsu, A.; Jia, X.; Kim, S. M.; Shi, Y.; Hofmann, M.; Nezich, D.; Rodriguez-Nieva, J. F.; Dresselhaus, M.; Palacios, T.; Kong, J. Nano Lett. 2012, 12, 161−166. (24) Corso, M.; Auwärter, W.; Muntwiler, M.; Tamai, A.; Greber, T.; Osterwalder, J. Science 2004, 303, 217−220. (25) Shi, Y.; Hamsen, C.; Jia, X.; Kim, K. K.; Reina, A.; Hofmann, M.; Hsu, A. L.; Zhang, K.; Li, H.; Juang, Z.-Y.; Dresselhaus, M. S.; Li, L.-J.; Kong, J. Nano Lett. 2010, 10, 4134−4139. (26) Song, X.; Gao, J.; Nie, Y.; Gao, T.; Sun, J.; Ma, D.; Li, Q.; Chen, Y.; Jin, C.; Bachmatiuk, A.; Rümmeli, M. H.; Ding, F.; Zhang, Y.; Liu, Z. Nano Res. 2015, 8, 3164−3176. (27) Grad, G. B.; Blaha, P.; Schwarz, K.; Auwärter, W.; Greber, T. Phys. Rev. B: Condens. Matter Mater. Phys. 2003, 68, 085404. (28) Li, Q.; Zou, X.; Liu, M.; Sun, J.; Gao, Y.; Qi, Y.; Zhou, X.; Yakobson, B. I.; Zhang, Y.; Liu, Z. Nano Lett. 2015, 15, 5804−5810. (29) Preobrajenski, A. B.; Vinogradov, A. S.; Mårtensson, N. Surf. Sci. 2005, 582, 21−30. (30) Lu, J.; Yeo, P. S. E.; Zheng, Y.; Xu, H.; Gan, C. K.; Sullivan, M. B.; Castro Neto, A. H.; Loh, K. P. J. Am. Chem. Soc. 2013, 135, 2368− 2373. (31) Sutter, P.; Lahiri, J.; Albrecht, P.; Sutter, E. ACS Nano 2011, 5, 7303−7309. (32) Orlando, F.; Larciprete, R.; Lacovig, P.; Boscarato, I.; Baraldi, A.; Lizzit, S. J. Phys. Chem. C 2012, 116, 157−164. (33) Dong, G.; Fourré, E. B.; Tabak, F. C.; Frenken, J. W. M. Phys. Rev. Lett. 2010, 104, 096102. (34) Laskowski, R.; Blaha, P.; Gallauner, T.; Schwarz, K. Phys. Rev. Lett. 2007, 98, 106802. (35) Liu, Y.; Zou, X.; Yakobson, B. I. ACS Nano 2012, 6, 7053−7058. (36) Suenaga, K.; Kobayashi, H.; Koshino, M. Phys. Rev. Lett. 2012, 108, 075501. (37) Huber, S. P.; Gullikson, E.; van de Kruijs, R. W. E.; Bijkerk, F.; Prendergast, D. Phys. Rev. B: Condens. Matter Mater. Phys. 2015, 92, 245310. (38) Auwarter, W.; Muntwiler, M.; Osterwalder, J.; Greber, T. Surf. Sci. 2003, 545, L735−L740. (39) Katzir, A.; Suss, J. T.; Zunger, A.; Halperin, A. Phys. Rev. B 1975, 11, 2370−2377. (40) Zunger, A.; Katzir, A. Phys. Rev. B 1975, 11, 2378−2390. (41) Andrei, E. Y.; Katzir, A.; Suss, J. T. Phys. Rev. B 1976, 13, 2831− 2834. (42) Chen, W.; Li, Y. F.; Yu, G. T.; Zhou, Z.; Chen, Z. F. J. Chem. Theory Comput. 2009, 5, 3088−3095. (43) Yu, G. T.; Liu, D.; Chen, W.; Zhang, H.; Huang, X. R. J. Phys. Chem. C 2014, 118, 12880−12889. (44) He, J. W.; Goodman, D. W. Surf. Sci. 1990, 232, 138−148. (45) Lu, J.; Gomes, L. C.; Nunes, R. W.; Castro Neto, A. H.; Loh, K. P. Nano Lett. 2014, 14, 5133−5139. (46) Marchini, S.; Guenther, S.; Wintterlin, J. Phys. Rev. B: Condens. Matter Mater. Phys. 2007, 76, 075429. (47) Wang, B.; Caffio, M.; Bromley, C.; Früchtl, H.; Schaub, R. ACS Nano 2010, 4, 5773−5782.

(48) Iannuzzi, M.; Tran, F.; Widmer, R.; Dienel, T.; Radican, K.; Ding, Y.; Hutter, J.; Groening, O. Phys. Chem. Chem. Phys. 2014, 16, 12374−12384. (49) Zhang, Z.; Liu, Y.; Yang, Y.; Yakobson, B. I. Nano Lett. 2016, 16, 1398−1403. (50) Tay, R. Y.; Griep, M. H.; Mallick, G.; Tsang, S. H.; Singh, R. S.; Tumlin, T.; Teo, E. H. T.; Karna, S. P. Nano Lett. 2014, 14, 839−846. (51) Chatterjee, S.; Luo, Z.; Acerce, M.; Yates, D. M.; Johnson, A. T. C.; Sneddon, L. G. Chem. Mater. 2011, 23, 4414−4416. (52) Liu, Y.; Bhowmick, S.; Yakobson, B. I. Nano Lett. 2011, 11, 3113−3116. (53) Yin, J.; Yu, J.; Li, X.; Li, J.; Zhou, J.; Zhang, Z.; Guo, W. Small 2015, 11, 4497−4502. (54) Stehle, Y.; Meyer, H. M.; Unocic, R. R.; Kidder, M.; Polizos, G.; Datskos, P. G.; Jackson, R.; Smirnov, S. N.; Vlassiouk, I. V. Chem. Mater. 2015, 27, 8041−8047. (55) Ouazi, S.; Pohlmann, T.; Kubetzka, A.; von Bergmann, K.; Wiesendanger, R. Surf. Sci. 2014, 630, 280−285. (56) Schulz, F.; Drost, R.; Hämäläinen, S. K.; Demonchaux, T.; Seitsonen, A. P.; Liljeroth, P. Phys. Rev. B: Condens. Matter Mater. Phys. 2014, 89, 235429. (57) Park, C.-H.; Louie, S. G. Nano Lett. 2008, 8, 2200−2203.

5856

DOI: 10.1021/jacs.7b00647 J. Am. Chem. Soc. 2017, 139, 5849−5856