中文   |   English   |   Email Alert
Fig/Tab
Advanced Search
image

Surface Synthesis of Heteroatoms-Doped Graphene Nanoribbons

Yong Zhang , Hui Zhang , Yi Zhang , Lei Gao , Jianchen Lu , Jinming Cai

Progress in Chemistry ›› 2023, Vol. 35 ›› Issue (1) : 105 -118.

PDF
Progress in Chemistry ›› 2023, Vol. 35 ›› Issue (1) : 105-118 DOI: 10.7536/PC220622
research-article
Scan to read & Share
Surface Synthesis of Heteroatoms-Doped Graphene Nanoribbons
Author information +
History +
PDF

Abstract

Atomically precise bottom-up synthesis of graphene nanoribbons under ultra-high vacuum condition is an important tool to open band gap of graphene. Rational design of precursor molecules with heteroatoms (boron, nitrogen, oxygen, sulfur, etc.) allows the synthesis of heteroatoms-doped graphene nanoribbons. Furthermore, heteroatoms-dopant can precisely tune the electrical, magnetic, and other physicochemical properties of graphene nanoribbons. The doping effect is closely related to the type, location and density of heteroatoms. In this review, we summarize the recent research progress on the synthesis and application prospects of heteroatoms-doped graphene nanoribbons based on the molecular beam epitaxy method. The applications of doped graphene nanoribbon are also propected.

Contents

1 Introduction

2 Armchair GNRs doped with heteroatoms

2.1 Sing-heteroatom doped Armchair GNRs

2.2 Multiple-heteroatom doped Armchair GNRs

2.3 Other doped AGNRs

3 Chiral GNRs doped with heteroatoms

4 Chevron GNRs doped with heteroatoms

4.1 Sing-heteroatom doped Chevron GNRs

4.2 Multiple-heteroatom doped Chevron GNRs

4.3 Chevron GNR heterojunctions

5 Zigzag GNRs doped with heteroatoms

6 Conclusion and outlook

Graphical abstract

Keywords

graphene nanoribbon / doping / bandgap engineering / scanning tunneling microscope / bottom-up

Cite this article

Download citation ▾
Yong Zhang, Hui Zhang, Yi Zhang, Lei Gao, Jianchen Lu, Jinming Cai. Surface Synthesis of Heteroatoms-Doped Graphene Nanoribbons. Progress in Chemistry, 2023, 35(1): 105-118 DOI:10.7536/PC220622

登录浏览全文

4963

注册一个新账户 忘记密码

1 Introduction

Graphene nanoribbons (GNRs), a kind of one-dimensional graphene nanostructures, have attracted much attention in recent years due to their electrical and magnetic properties caused by quantum interference effects at the nanoscale[1⇓⇓⇓⇓~6]. Compared with semimetallic graphene, GNRs exhibit semiconductor properties with tunable band gap, and the band gap size is related to the edge geometry, bandwidth and band length, which requires the preparation of atomically accurate GNRs[1,7]. Therefore, the bottom-up synthesis method based on ultra-high vacuum scanning tunneling microscopy (UHV-STM) has emerged as the times require. Compared with the top-down cutting of graphite and tailoring of carbon nanotubes (CNTs), the bottom-up synthesis of GNRs meets the requirements of atomic-level accuracy[8][9][10]. Fig. 1A shows the basic steps of the bottom-up synthesis of GNRs. Precursor molecules containing halogen atoms are deposited on the surface of a noble metal with certain catalytic properties by thermal sublimation. At a certain annealing temperature,Molecules are dehalogenated to form intermediates with diradicals, which diffuse freely on the metal surface. During diffusion, the radicals between molecules react to form linear polymers. The solid surface is further heated to a higher temperature, and the polymer chain is cyclodehydrogenated to generate flat GNRs. Due to the diversity of precursor molecules, the synthesized GNRs are also diverse, which are mainly divided into four types according to their edge structures: Armchair GNRs, Zigzag GNRs, Chiral GNRs and V-type GNRs (Chevron GNRs), as shown in Figure 1B. Each type of GNRs has its unique properties, for example, the intrinsic Armchair-type GNRs show semiconductor properties, and the band gap shows a certain regularity with the width of GNRs.Intrinsic Zigzag GNRs have spin-polarized edge States due to their unique Zigzag edges. Chiral GNRs are composed of periodically alternating Armchair and Zigzag edges. It is shown that intrinsic Chiral GNRs have narrow-band semiconductor properties[11][2][12].
The ultimate goal of all synthesized GNRs is to apply them to semiconductor devices. To achieve this goal, the first problem is the adjustability of the band gap. The band gap of the semiconductor devices we produce is not arbitrary, and there is a more appropriate range. For example, the band gap of the commonly used silicon-based semiconductor is about 1.1 eV, which requires reasonable control of the band gap of GNRs. However, the control of the ribbon structure of GNRs by width and shape has entered a certain bottleneck period. With the growth of GNRs family, it is more and more difficult to design new precursor molecules for the growth of GNRs.It is not easy to regulate the band structure by regulating the growth length of GNRs. Controlling the growth of GNRs within a certain length range involves complex molecular dynamics theory, and only a few research groups are engaged in this work at present[3,13]. Doping is an effective way to precisely dope heterogeneous atoms into GNRs, which can not only effectively control the band gap of GNRs, but also improve the application potential of GNRs in optoelectronic, spintronic and quantum computing devices. The introduction of heteroatoms also provides a unique opportunity to prepare heterostructures of GNRs and induce them to produce other properties. In addition, the introduction of partial hetero-atoms can reduce the interaction between GNRs and the substrate, which makes it easier to probe the intrinsic electronic structure of GNRs. So far, the heterogeneous elements introduced in GNRs are boron (B), nitrogen (N), oxygen (O), sulfur (S), etc. The introduction of these heterogeneous atoms can greatly improve the electrical and magnetic properties of GNRs. For example, replacing C atoms in the lattice of GNRs with B atoms will make GNRs locally magnetic[14]. The introduction of different kinds of N atoms (pyridinic nitrogen and graphitic nitrogen) into GNRs will bring about subversive changes in the electronic structure of GNRs and produce special edge States[15]. Introducing the backbone of GNRs into N atoms to build donor-acceptor (D-A) heterostructures and so on[16,17]. In addition, doped GNRs can also form GNRs heterojunctions with intrinsic GNRs, which have been shown to have better charge transfer efficiency at the interface than traditional semiconductor heterojunctions, and have great potential applications in optoelectronic devices[18].
In this paper, four types of GNRs are taken as the starting point, and in each type, the research progress of researchers in the synthesis of doped GNRs on the surface is introduced from the aspects of single-atom doping and multi-atom doping. The structural and property changes of doped GNRs compared with undoped GNRs are emphasized. It provides ideas and references for better study of doped GNRs and their application in semiconductor devices and optoelectronic devices.

2 Armchair-type doped GNRs

Armchair-type GNRs (AGNRs) are currently the most studied class of GNRs on noble metal surfaces. Table 1 lists the AGNRs with different widths synthesized by some precursor molecules. According to the change of band structure, AGNRs with different widths are divided into three groups, namely 3p, 3p + 1 and 3p + 2 (p represents a positive integer), as shown in Figure 2. In each group, the band gap of AGNRs decreases as p increases (Fig. 2a, C). When p is a specific value, the band gaps of the three groups of AGNRs change as 3p + 1 > 3p > 3p + 2 (Fig. 2a)[1,20⇓~22]. AGNRs with different widths provide more selectivity for the research and development of semiconductor devices, and the possibility of their application can be greatly improved by adjusting the band structure of AGNRs with a certain width[19].

2.1 Single-atom doped AGNRs.

Periodically doped GNRs can be formed by doping single atoms into the precursor molecules, which can not only control the electronic structure of GNRs, but also bring many novel properties. In 2015, B-doped AGNRs were reported by Kawai et al. And Cloke et al., who used the precursor molecules as shown in Figure 3A to obtain periodic B-doped GNRs (BB-GNRs)[23⇓~25]. Due to the strong interaction of the B atom with the substrate, the B site appears darker in GNRs (Figure 3B and Figure 3D). Through the measurement of dI/dV spectra at different positions of GNRs (Figure 3C), its band gap is 2.4 eV, which is similar to that of undoped N = 7 AGNRs (2.3 eV)[26]. The main reason is that the B atom exists periodically in the center of the GNRs, and there is no obvious electron-hole asymmetry, so the complete electronic structure of the original GNRs is maintained[27]. In addition, the B-doped site shows good adsorption for NO molecules, and the free beating of NO molecules on the B site can be realized by tip manipulation.
Pedramrazi et al. And Friedrich et al. Obtained two boron-doped GNR (2B-GNR) by co-deposition, and the schematic diagram is shown in Fig. 3e[14,24]. There is a significant difference between the position of B atom doping and the original GNR (Figure 3 f green dotted box). It is found that B atom doping will introduce two new interstitial doping States with different symmetries for 2B-GNR (Fig. 3G). DFT calculation results show that compared with the band structure of undoped N = 7 AGNR, the band structure of 2B-GNR will change significantly, and two doping States will be introduced near the Fermi surface (Fig. 3H), which is consistent with the experimental results.
Theoretical studies point out that the band gap of GNRs can be affected to different degrees by introducing heterogeneous atoms into different positions of GNRs[27,28]. To this end, our research group has successfully introduced different kinds of N atoms (pyridine nitrogen and graphite nitrogen) into N = 9 GNRs, and the results are shown in Figure 4[29]. Bond-resolved STM (BR-STM) images show that the introduction of the pyridine nitrogen atom has little effect on the apparent structure of N = 9 GNRs (Fig. 4B), while a small bump appears at the position of the graphite nitrogen atom (Fig. 4C). The main reason may be that the graphite nitrogen atom disturbs the local electronic state of GNRs, which can accurately identify the position of the pyridine nitrogen atom and the graphitic nitrogen atom. The results of dI/dV spectra show that the introduction of pyridine nitrogen atoms slightly enlarges the band gap of this GNRs compared with the undoped N = 9 GNRs, while the isolated graphite nitrogen atoms introduce two metallic-like electronic States to this GNRs near the Fermi surface (Figure 4D). The DFT results show that the charge density of GNR is redistributed due to the introduction of two kinds of N atoms, with the charge depletion at the pyridine nitrogen atom and the charge accumulation at the graphite nitrogen atom. The results of the projected band structure and the corresponding projected density of States show that the pyridine nitrogen atom acts on the outside of the GNRs, which is equivalent to the hydrogen atom of a carbon atom, so it has no essential effect on the band structure of the GNRs (semiconductor properties). The graphitic nitrogen atom acts inside GNRs, and because it has one more electron than the replaced carbon atom, it has an essential effect on the band structure of GNRs (metallicity).
Nguyen et al. synthesized S-doped N = 13 AGNRs (S-AGNRs) using the precursor molecules shown in Fig. 5A, and their high-resolution STM images are shown in Fig. 5B[30]. They performed STS characterization at different sites of S-AGNRs (Fig. 5C), and found that there was an obvious peak near V = 1.23 eV, corresponding to the position of the conduction band of S-AGNRs, while the peak at V < 0 was very similar to the characteristic peak of the Au (111) substrate, and it was difficult to distinguish whether this peak was the peak of the valence band of S-ADNRs or the impurity peak affected by the substrate. To better understand the electronic properties of S-AGNRs, they did DFT calculations of the band structure for doped and undoped GNRs, respectively (Figure 5D). The band gap of S-doped GNRs is only slightly reduced (0.76 eV) compared to that of undoped N = 13 AGNRs (0.90 eV). In addition, the band structure of S-doped GNRs has a similar energy arrangement with that of undoped GNRs, and this band structure arrangement is different from that of N-doped GNRs, which is mainly attributed to the electronegativity difference of S, N and C[18]. Due to the small electronegativity difference between S and C, the band structure of GNRs with S atoms replacing C atoms has a similar energy level arrangement (fig. 5d), and the band gap is only slightly reduced. However, the electronegativity of N and C is quite different. Substituting N atom for C atom in GNR will not only affect the band gap of GNR, but also make the conduction band and valence band move rigidly[30].

2.2 Polyatomically doped AGNRs.

Polycyclic aromatic hydrocarbons (PAHs) have been a hot research topic. Doping a variety of heterogeneous atoms into GNRs can not only control their band structure, but also change the aromaticity and bond length of GNRs. In 2018, Kawai et al. Used the precursor molecules shown in Fig. 6a to synthesize a variety of heterogeneous atom (B and N) -doped AGNRs[31]. Due to the asymmetry of the precursor molecules, three different structural units are formed. The triangular defect appearing in the figure is similar to the flower-like defect in the previous study of BB-AGNRs, which easily reminds us that the triangular defect may be caused by the strong interaction between the B atom in BN-AGNRs and the substrate, but this is not the case. The calculation shows that in BN-AGNRs, the triangular defect center is occupied by N atoms, while the position occupied by B atoms appears brighter (Fig. 6e). In order to clarify the reason for the appearance of this triangular defect, they performed force spectroscopy measurements at different element points (Figure 6 f). It can be seen from the figure that the tails of the frequency curves are perfectly fused, indicating that the contribution of the long-range van der Waals force to the imaging is small, and the short-range Pauli repulsive force plays a decisive role in the imaging. In the range of short-range Pauli repulsive force, the frequency negative shift of B site is the smallest, while the frequency negative shift of N site is the largest, and the frequency positive shift of C site is in between. The frequency negative shift reflects the apparent morphology of NC-AFM image, so the position of N element will be darker. The electrical properties are shown in Fig. 6 I ~ l. The band gap of BN-AGNRs is 2.3 eV through STS characterization. Compared with the undoped AGNRs, the valence band and conduction band are shifted up by 0.4 eV, and the band gap remains unchanged. This is mainly because the B and N atoms are periodically located in the center of GNRs, and there is no obvious electron-hole asymmetry, which maintains the complete band structure of undoped AGNRs.

2.3 Other doped AGNRs

The results show that the introduction of some holes in two-dimensional materials such as graphene can not only control their band structure well, but also change their charge transport properties greatly[32,33][34,35]. Pawlak et al. Successfully synthesized N = 9 porous GNRs using the precursor molecules shown in Fig. 7A, and the periodic porous structures are shown in Fig. 7B and C[17]. Compared with the undoped N = 9 GNRs (1.4 eV), the band gap of the doped GNRs is larger, 2.7 eV (Fig. 7 d, e), which belongs to the wide band gap semiconductor[36]. It has been shown that wide band gap semiconductors have great prospects in electronics, photonics, high power and high frequency devices, solar energy conversion, photocatalysis and thermal insulation, which indicates that this GNRs has great potential in semiconductor devices[53].
In this paper, the research progress of AGNRs doped with heterogeneous atoms in recent years is reviewed. Through previous studies, the following conclusions can be drawn: 1) The position and density of heterogeneous atoms and functional groups in AGNRs can be accurately identified by the difference of electronic States; 2) The physical and chemical properties of AGNRs are significantly affected by the position of the heterogeneous atom. Doping the heterogeneous atom in the center of AGNRs has little effect on its band gap, while doping at the edge has a greater effect on its band gap. 3) Some new properties, such as magnetism and adsorption, can be introduced into AGNRs by introducing heterogeneous atoms into appropriate positions.

3 Chiraltype doped GNRs

Chiral-type GNRs (chGNRs), composed of periodically alternating Armchair and Zigzag edges, have different electrical properties from AGNRs and ZGNRs. Its original synthesis was based on an accidental experiment in which the researchers discovered the existence of N = 7 GNRs when they were deposited on Cu (111) with synthetic N = 7 GNRs (DBBA)[4,37,38]. After that, Jacobse et al. Used the precursor molecules as shown in fig. 8A and deposited them on a Cu (111) substrate to form chGNRs as shown in fig. 8B[12]. In order to characterize the electrical properties of the GNRs, they pulled the chGNRs onto NaCl by tip manipulation to obtain their intrinsic electrical properties. dI/dV spectroscopy measurements show a reduced band gap of around 1.75 eV for this type of GNRs compared to N = 7 Armchair-type GNRs (Fig. 8 C)[26].
In 2018, Wang et al. synthesized OBO unit-doped Chiral-type GNRs (OBO-chGNRs) for the first time by using the precursor molecules shown in Fig. 9a, and their STM and NC-AFM images are shown in Fig. 9b and C[39]. From the NC-AFM map, it can be seen that there is a significant darkening at the OBO doped site, which is due to the strong interaction between the B atom and the substrate, which has been introduced before. Measurement of the dI/dV spectrum shows that its conduction and valence bands are 1.77 eV and -1.56 eV, respectively (Figure 9 d), corresponding to a band gap of 3.33 eV. The corresponding DFT calculation results show that the band gap of OBO-doped GNRs is 1.5 eV, while the band gap of undoped GNRs is only 0.5 eV, as shown in Fig. 9e. The main reason for the band gap difference is the weak conjugation of the pz orbital of the doped segment with the extended vibrational system, which reduces the effective width of GNRs and thus increases the band gap.
Feng et al. Innovatively synthesized N-doped chGNRs (N-chGNRs) using the precursor molecules shown in Fig. 10a, and their STM images are shown in Fig. 10B[40]. The N element has been successfully doped into the GNRs as evidenced by XPS measurements (Fig. 10 C, d). In addition, they measured the angle-resolved ultraviolet photoelectron spectroscopy (ARUPS) of Au substrates, polymers, and N-chGNRs, respectively. Fig. 10e shows the ARUPS of the bare Au (111) surface, and the binding energy at 0.5 eV is the surface state of Au (111). Fig. 10f shows the ARUPS of polymer on Au (111) surface at high coverage. There are molecular occupied States near 2.5 eV, 4.0 eV, 6.0 eV, 7.5 eV, 11.0 eV and 13.5 eV, and there is no obvious angular dispersion of these features. The existence of molecular States indicates that the polymer is composed of independent molecular units, while the disappearance of Au (111) surface States indicates that the sample has a high coverage and a strong interaction between the molecules and the substrate. Fig. 10g and H show the ARUPS of N-chGNRs on Au (111) surface at high coverage. The non-dispersive molecular state and the π-conjugated energy band (around 8.8 eV) indicate that the all-carbon GNR has been successfully synthesized. In addition, some residual impurities were found in ARUPS, which were related to the incomplete dehydrocyclization of the free phenyl group on the outside of GNR.
In this summary, the research progress of heteroatom doped chGNRs in recent years is reviewed. The band structure of chGNRs can be significantly tuned by introducing suitable hetero-atoms, and the degree of tuning varies with different doping atoms. The introduction of periodic OBO units at the edge of chGNRs can increase the band gap of chGNRs, which is mainly due to the weak conjugation between the pz orbital of the doped segment and the extended vibrational system of the main chain, which is equivalent to reducing the effective width of chGNRs, thus increasing the band gap. Due to the difficulty in the synthesis of precursor molecules, there are few studies on the doping system of chGNRs. Through the rational design of precursor molecules, the variety of doped chGNRs can be enriched, and the influence of doping on the physical and chemical properties of chGNRs can be deeply understood.

4 Chevron-type doped GNRs.

Chevron-type GNRs are composed of alternating segments of Armchair-type GNRs with different widths. They are also called V-type GNRs (VGNRs) because they are shaped like "V"[15,41⇓~43]. In recent years, different research groups have used different precursor molecules to synthesize VGNRs with different bandwidths (see Table 2), and have made great progress by replacing C atoms in VGNRs with heterogeneous atoms or modifying different groups at the edge of VGNRs to adjust their bandgaps. In addition, due to the unique structure of the precursor molecule, people prefer to synthesize VGNR heterojunctions rather than AGNRs heterojunctions[18]. The results show that the electron mobility at the interface of the heterojunction is much higher than that of the traditional semiconductor heterojunction, which has great potential in optoelectronic devices.

4.1 Single-atom doped V-type GNRs.

Zhang et al. And Vo et al. Used two different precursor molecules to synthesize VGNRs with different nitrogen atom (N) doping concentrations, as shown in Fig. 11 a and B[44,45]. Compared with undoped VGNRs, the bands of edge-doped VGNRs tend to be parallel to each other, which is due to the weak interaction force (van der Waals force) between N atoms and H atoms[43]. Characterization of the electrical properties shows (Fig. 11 d, e) that the band gap of the edge-doped GNRs is significantly reduced compared to the undoped VGNRs (2.33 eV). The main reasons are as follows: 1) Previous studies have shown that heterogeneous atom doping has different effects on the band gap at different positions of GNRs, and doping in the middle (edge) has a smaller (larger) effect on the band gap. In this system, N atoms are doped at the edge of VGNRs, so it has a great influence on the band gap. 2) The electronegativity difference between N atom and C atom is large, and the electronegativity difference can also affect the bandwidth. After that, Zhang et al. Synthesized S-doped VGNRs. Due to the flexibility of the precursor molecule, there are many different structures in the synthesized VGNRs (shown by the arrow in Figure 11c)[46,47]. The calculation shows that the different structures have a certain influence on their electrical properties (Figure 11 f), but overall, the band gap of VGNRs is reduced by S doping.

4.2 Polyatomically doped V-type GNRs.

Durr et al. doped three heterogeneous atoms of nitrogen (N), sulfur (S), and oxygen (O) into the precursor molecule as shown in Fig. 12A, respectively, and the STM and NC-AFM images formed are shown in Fig. 12b ~ d[48]. Compared with the undoped V-type GNRs, the widths of the VGNRs formed by the doping of these hetero-atoms are slightly wider, which is mainly due to the change of the local electronic structure of the VGNRs by the doping atoms. The difference of different heteroatom doping can be distinguished from the NC-AFM images. Compared with N and S atom doping, the NC-AFM images of O atom doping sites are darker, which can be attributed to the attraction between O atoms in the tip of NC-AFM tip modified by CO molecules and O atoms in VGNRs doped by O atoms. The characterization of the electrical properties is shown in Fig. 12e, which presents the following changes compared to the undoped VGNRs: VGNR < O-VGNR < S-VGNR < N-VGNR (valence band); S-VGNR < O-VGNR < N-VGNR < VGNR (conduction band). Two unique effects dominate the changes in the band structure of edge-doped VGNRs: 1) electron transfer between VGNRs and dopant atoms, with O and S atoms as acceptors and N atoms as donors according to Lowdin charge analysis, absorbing and donating electrons to the host of VGNRs, respectively; 2) The extended delocalization of the wave function. It can be seen from Figure 12 f that the p-orbital of the lone pair electron of the dopant atom will overlap with the orbital of the extended hybrid system, resulting in different degrees of shift in the band structure. Therefore, the energy alignment close to the Fermi level is determined by the partial electron transfer between the dopant atom and the host of VGNRs and the effective conjugation of the extended system with the lone pair of electrons of the dopant atom.

4.3 V-type GNRs heterojunction

Semiconductor heterojunction is formed by combining two different materials in some way. Due to the different energy bands of the two materials, energy offset occurs at the interface, which is mainly divided into three types, as shown in Figure 13[49]. In the type-I heterojunction, the valence band of semiconductor B is lower than that of semiconductor a, and the conduction band of semiconductor B is higher than that of semiconductor a, which is also called Straddling gap heterojunction. In the type-II heterojunction, the valence band of semiconductor B is higher than that of semiconductor a (but lower than the conduction band of semiconductor a), and the conduction band of semiconductor B is also higher than that of semiconductor a, which is also called Staggered gap heterojunction. The band structure of type-Ⅲ heterojunction is similar to that of type-Ⅱ heterojunction, but the valence band of semiconductor B is higher than the conduction band of semiconductor a, which is called Broken gap heterojunction. Similar to the traditional semiconductor heterojunction, GNRs heterojunction is the combination of two different GNRs at the same interface, and its energy offset at the interface is similar to that of type-II heterojunction. Such heterojunctions are predicted to play a useful role in future nanotechnologies, including the fabrication of robust molecular diodes and transistors, among others.
In 2014, Cai et al. Synthesized p-N-VGNRs heterojunction by co-deposition using the precursor molecules in Fig. 14a[18]. It can be seen from the STM images (Fig. 14 B, C) that the doped and undoped GNRs segments show different color bands at a certain scanning bias, and the doped and undoped sites can be clearly distinguished by dI/dV maps (Fig. 14 C). The electronic structures of p-VGNRs and N-VGNRs are shown on the left and right sides of Figure 14e, respectively. Since the unpaired N atoms on the boundary of N-VGNRs are not combined with the π system in the intrinsic VGNRs, the effect on the band gap is small, only about 0.1 eV. However, the incorporation of N changes its band structure, resulting in an overall shift of its band gap by 0.5 eV. Compared with the traditional semiconductor heterojunction, the electron mobility of p-N-VGNRs heterojunction at the interface is much higher, which is 2×108V/m (Figure 14 G), which greatly demonstrates its application potential in optoelectronic devices and broadens the development of next-generation semiconductor heterojunction.
In this paper, the research progress of VGNRs doped with heterogeneous atoms in recent years is reviewed. Through previous studies, the following conclusions can be drawn: 1) The control of heteroatoms on the band structure of VGNRs is closely related to their density and doping position; 2) The effect of different heteroatoms on the band structure of VGNRs is determined by the partial electron transfer between the heteroatom and the host of VGNRs and the effective conjugation of the functional group with the extended system of VGNR; 3) By co-depositing the undoped precursor molecules with the doped precursor molecules, different GNRs heterojunctions can be obtained, which have much higher electron mobility at the interface compared with the traditional semiconductor heterojunctions, greatly demonstrating their potential applications in optoelectronic devices.

5 Zigzag-type doped GNRs

ZGNRs with spin-polarized electronic edge States are among the most promising graphene-based materials to be developed into spintronic devices. Because of the inherent characteristics of ZGNRs, the precursor molecules of this GNR are difficult to synthesize, and only one width of ZGNRs has been reported so far. In 2016, Ruffieux et al. synthesized N = 6 ZGNRs (Fig. 15b) with the precursor molecules shown in Fig. 15A. In order to characterize the intrinsic electrical properties of ZGNRs, they bridged the two ends of GNRs on NaCl islands (Fig. 15C), and the dI/dV spectrum obtained is shown in Fig. 15d[2]. The spectrum clearly shows three resonance peaks near the Fermi level, and the dI/dV mapping of the resonance peaks are clustered at the edge of the ZGNRs, and the DFT calculation results are basically consistent with the experimental results (Fig. 15e, f). Some factors can adjust the electronic structure of ZGNRs, such as strain, electric field, etc. Lu et al. Reported that the spin polarization at the edge and band gap of ZGNRs can be controlled by strain. Through theoretical calculation, they predicted that ZGNRs can be transformed into half-metallic bands under the action of in-plane electric field[50]. The unique zigzag edge structure makes ZGNRs one of the popular choices for studying carbon-based spintronics.
In 2020, Fu et al. Synthesized two kinds of NBN unit doped ZGNRs (NBN-ZGNRS-1 and NBN-ZGNRS-2) with the precursor molecules shown in Fig. 16 a and d. It can be seen from the NC-AFM images (Fig. 16 C and e) that the two kinds of NBN-ZGNRs have periodically arranged holes, and the sizes of the holes are different[51]. Experimentally, the band gaps of the two NBN-ZGNRs are 1.50 and 0.90 eV (Fig. 16d, G), which are enlarged compared with the undoped ZGNRs. An interesting phenomenon is that if the two kinds of NBN-ZGNRs are oxidized into free cations, that is, each NBN unit loses one electron (Fig. 16 J, K), their band structures become 0 gap (0 eV) and metallic (Fig. 16 l, m), respectively, which are completely different from the electronic structures of NBN-ZGNRs without oxidation. This kind of NBN-doped ZGNRs lays a foundation for the study of pure carbon-based isoelectronic structures.
Recently, Blackwell et al. successfully synthesized N = 6 N-doped ZGNRs (N-ZGNRs) using the precursor molecules shown in Fig. 17A, and their large-area STM images are shown in Fig. 17b[52]. The introduction of N atom has two effects: 1) the introduction of N atom into the precursor molecule can reduce the formation energy of the ZGNR; 2) introducing N atoms into the ZGNR can effectively reduce the interaction between the ZGNR and the metal substrate, thereby detecting the intrinsic edge state; 3) The band gap of N = 6 ZGNR is significantly tuned by the introduction of N atoms. Fig. 17c shows the dI/dV spectrum of N-ZGNRs, which shows that the conduction and valence bands of N-ZGNRs are 0.5 eV and -0.3 eV, respectively, and the band gap is 0.80 eV. The DFT calculation results show that the band gap of N-ZGNR is significantly reduced to 0.83 eV compared with that of undoped ZGNR (1.20 eV) (Fig. 17d).
In this summary, the research progress of heteroatom doped ZGNRs in recent years is reviewed. Due to the difficulty in the synthesis of precursor molecules, there are few studies on doped ZGNRs. Through doping, the interaction between ZGNR and metal substrate can be reduced, and the band gap of ZGNRs can be significantly regulated, which greatly improves the possibility of applying ZGNRs to spintronic devices.

6 Conclusion and outlook

The bottom-up method based on ultra-high vacuum scanning tunneling microscopy technology to synthesize atomically accurate GNRs is a very important research direction. In recent years, it has experienced vigorous development in both experiment and theory. By selecting different precursor molecules, GNRs with different shapes, different widths and different band gaps can be synthesized, which provides a broad selectivity for the research and development of micro semiconductor devices. By doping, the band gap of GNRs can be effectively controlled, and some other properties are introduced, which greatly improve the application prospects of GNRs in molecular switches, magnetic nano-devices and optoelectronic devices. Doped GNRs can also be co-deposited with undoped GNR to form GNR heterojunctions, which are expected to shine in next-generation semiconductor heterojunctions due to their excellent electron mobility at the interface. In this paper, the research progress of doped GNRs synthesized by bottom-up method in ultra-high vacuum environment is reviewed, which provides a reference for future research on doped GNRs.
Although a series of important results have been achieved in the preparation of atomically accurate doped GNRs by bottom-up synthesis under ultra-high vacuum conditions, there are still several problems that need to be further studied.
1) The length of the doped GNRs formed at present is still relatively short and cannot be synthesized on a large scale. It is still a big problem to synthesize doped GNRs in large scale and high quality under the condition of ensuring their excellent performance. The results show that the length of the polymer chain on the surface can be effectively increased by selecting the appropriate precursor molecules, controlling the heating rate and annealing temperature, and its feasibility needs to be further verified[36].
2) Most of the doped GNRs prepared only regulate the electrical properties, and there are few studies on other properties, such as magnetism, superconductivity, aromaticity and so on. In recent years, the study of magnetic properties of carbon-based materials has become a hot topic. People have doped some magnetic atoms or introduced some defects into two-dimensional materials such as graphene to control their magnetic properties. By designing suitable precursor molecules, GNRs with different magnetic properties are expected to be synthesized.
3) Substrate is one of the indispensable conditions for the preparation of doped GNRs by bottom-up method, and it is also the catalyst in the synthesis process of GNRs. In most of the prepared doped GNRs samples, the interaction between GNRs and the substrate is strong, which limits the application of doped GNRs in real life to some extent. Therefore, it is necessary to develop a reliable new transfer technique to transfer GNRs to insulating substrates without destroying their quality, or to find a reliable substrate so that doped GNRs can be transferred conveniently while growing on the substrate with high quality.
The bottom-up approach to synthesize atomically accurate GNRs lays a solid foundation for the application of graphene in two-dimensional functional devices. Doping can precisely control the band structure of GNR, which provides a broad selectivity for the research and development of GNR-based semiconductor devices. Recently, the excellent performance of nanographene in magnetism has begun to gain more attention and research[54,55]. Due to the existence of unsaturated electrons, the wet chemical synthesis of magnetic nanographene is more difficult, and the designed precursor molecules can be accurately transformed into magnetic nanographene by means of surface catalysis in ultra-high vacuum. It is believed that in the near future, functional carbon-based materials such as GNRs and nano-graphene will be put into practical application and write a new chapter for the development of semiconductor industry.

References

Publishing order | Descend order by publishing year | Descend order by cited within
[1]

Yang L, Park C H, Son Y W, Cohen M L, Louie S G. Phys. Rev. Lett., 2007, 99(18): 186801.
[2]

Ruffieux P, Wang S Y, Yang B, Sánchez-Sánchez C, Liu J, Dienel T, Talirz L, Shinde P, Pignedoli C A, Passerone D, Dumslaff T, Feng X L, Müllen K, Fasel R. Nature, 2016, 531(7595): 489.
[3]

Yamaguchi J, Hayashi H, Jippo H, Shiotari A, Ohtomo M, Sakakura M, Hieda N, Aratani N, Ohfuchi M, Sugimoto Y, Yamada H, Sato S. Commun. Mater., 2020, 1: 36.
[4]

Han P, Akagi K, Federici Canova F, Mutoh H, Shiraki S, Iwaya K, Weiss P S, Asao N, Hitosugi T. ACS Nano, 2014, 8(9): 9181.
[5]

Liu M Z, Liu M X, She L M, Zha Z Q, Pan J L, Li S C, Li T, He Y Y, Cai Z Y, Wang J B, Zheng Y, Qiu X H, Zhong D Y. Nat. Commun., 2017, 8: 14924.
[6]

Panighel M, Quiroga S, Brandimarte P, Moreno C, Garcia-Lekue A, Vilas-Varela M, Rey D, Sauthier G, Ceballos G, Peña D, Mugarza A. ACS Nano, 2020, 14(9): 11120.
[7]

Merino-Díez N, Garcia-Lekue A, Carbonell-Sanromà E, Li J C, Corso M, Colazzo L, Sedona F, Sánchez-Portal D, Pascual J I, de Oteyza D G. ACS Nano, 2017, 11(11): 11661.
[8]

Cai J M, Ruffieux P, Jaafar R, Bieri M, Braun T, Blankenburg S, Muoth M, Seitsonen A P, Saleh M, Feng X L, Müllen K, Fasel R. Nature, 2010, 466(7305): 470.
[9]

Jiao L Y, Zhang L, Wang X R, Diankov G, Dai H J. Nature, 2009, 458(7240): 877.
[10]

Kosynkin D V, Higginbotham A L, Sinitskii A, Lomeda J R, Dimiev A, Price B K, Tour J M. Nature, 2009, 458(7240): 872.
[11]

Houtsma R S K, de la Rie J, Stöhr M. Chem. Soc. Rev., 2021, 50(11): 6541.
[12]

Jacobse P H, Mangnus M J J, Zevenhuizen S J M, Swart I. ACS Nano, 2018, 12(7): 7048.
[13]

di Giovannantonio M, Deniz O, Urgel J I, Widmer R, Dienel T, Stolz S, Sánchez-Sánchez C, Muntwiler M, Dumslaff T, Berger R, Narita A, Feng X L, Müllen K, Ruffieux P, Fasel R. ACS Nano, 2018, 12(1): 74.
[14]

Friedrich N, Brandimarte P, Li J C, Saito S, Yamaguchi S, Pozo I, Peña D, Frederiksen T, Garcia-Lekue A, Sánchez-Portal D, Pascual J I. Phys. Rev. Lett., 2020, 125(14): 146801.
[15]

Nguyen G D, Tsai H Z, Omrani A A, Marangoni T, Wu M, Rizzo D J, Rodgers G F, Cloke R R, Durr R A, Sakai Y, Liou F, Aikawa A S, Chelikowsky J R, Louie S G, Fischer F R, Crommie M F. Nat. Nanotechnol., 2017, 12(11): 1077.
[16]

Carbonell-Sanromà E, Hieulle J, Vilas-Varela M, Brandimarte P, Iraola M, Barragán A, Li J C, Abadia M, Corso M, Sánchez-Portal D, Peña D, Pascual J I. ACS Nano, 2017, 11(7): 7355.
[17]

Pawlak R, Liu X S, Ninova S, D’Astolfo P, Drechsel C, Sangtarash S, Häner R, Decurtins S, Sadeghi H, Lambert C J, Aschauer U, Liu S X, Meyer E. J. Am. Chem. Soc., 2020, 142(29): 12568.
[18]

Cai J M, Pignedoli C A, Talirz L, Ruffieux P, Söde H, Liang L B, Meunier V, Berger R, Li R J, Feng X L, Müllen K, Fasel R. Nat. Nanotechnol., 2014, 9(11): 896.
[19]

Jacobse P H, McCurdy R D, Jiang J W, Rizzo D J, Veber G, Butler P, Zuzak R, Louie S G, Fischer F R, Crommie M F. J. Am. Chem. Soc., 2020, 142(31): 13507.
[20]

Sun K W, Ji P H, Zhang J J, Wang J X, Li X C, Xu X, Zhang H M, Chi L F. Small, 2019, 15(15): 1804526.
[21]

Chen Y C, de Oteyza D G, Pedramrazi Z, Chen C, Fischer F R, Crommie M F. ACS Nano, 2013, 7(7): 6123.
[22]

Kimouche A, Ervasti M M, Drost R, Halonen S, Harju A, Joensuu P M, Sainio J, Liljeroth P. Nat. Commun., 2015, 6: 10177.
[23]

Kawai S, Saito S, Osumi S, Yamaguchi S, Foster A S, Spijker P, Meyer E. Nat. Commun., 2015, 6: 8098.
[24]

Pedramrazi Z, Chen C, Zhao F Z, Cao T, Nguyen G D, Omrani A A, Tsai H Z, Cloke R R, Marangoni T, Rizzo D J, Joshi T, Bronner C, Choi W W, Fischer F R, Louie S G, Crommie M F. Nano Lett., 2018, 18(6): 3550.
[25]

Cloke R R, Marangoni T, Nguyen G D, Joshi T, Rizzo D J, Bronner C, Cao T, Louie S G, Crommie M F, Fischer F R. J. Am. Chem. Soc., 2015, 137(28): 8872.
[26]

Ruffieux P, Cai J M, Plumb N C, Patthey L, Prezzi D, Ferretti A, Molinari E, Feng X L, Müllen K, Pignedoli C A, Fasel R. ACS Nano, 2012, 6(8): 6930.
[27]

Biel B, Blase X, Triozon F, Roche S. Phys. Rev. Lett., 2009, 102(9): 096803.
[28]

da Costa Azevêdo A S, Saraiva-Souza A, Meunier V, Girão E C. Phys. Chem. Chem. Phys., 2021, 23(23): 13204.
[29]

Zhang Y, Lu J C, Li Y, Li B J, Ruan Z L, Zhang H, Hao Z L, Sun S J, Xiong W, Gao L, Chen L, Cai J M. Angewandte. Chemie. Int. Ed., 2022, 61(28): e202204736.
[30]

Nguyen G D, Toma F M, Cao T, Pedramrazi Z, Chen C, Rizzo D J, Joshi T, Bronner C, Chen Y C, Favaro M, Louie S G, Fischer F R, Crommie M F. J. Phys. Chem. C, 2016, 120(5): 2684.
[31]

Kawai S, Nakatsuka S, Hatakeyama T, Pawlak R, Meier T, Tracey J, Meyer E, Foster A S. Sci. Adv., 2018, 4(4): eaar7181.
[32]

Chong M C, Reecht G, Bulou H, Boeglin A, Scheurer F, Mathevet F, Schull G. Phys. Rev. Lett., 2016, 116(3): 036802.
[33]

Pedersen T G, Flindt C, Pedersen J, Mortensen N A, Jauho A P, Pedersen K. Phys. Rev. Lett., 2008, 100(18): 189905.
[34]

Tan H X, Liu T Q, Zhang X, Shan Q, Chen J, Li Z, Ihara H, Qiu H D. Anal. Chem., 2020, 92(20): 13630.
[35]

Moreno C, Vilas-Varela M, Kretz B, Garcia-Lekue A, Costache M V, Paradinas M, Panighel M, Ceballos G, Valenzuela S O, Peña D, Mugarza A. Science, 2018, 360(6385): 199.
[36]

Talirz L, Söde H, Dumslaff T, Wang S Y, Sanchez-Valencia J R, Liu J, Shinde P, Pignedoli C A, Liang L B, Meunier V, Plumb N C, Shi M, Feng X L, Narita A, Müllen K, Fasel R, Ruffieux P. ACS Nano, 2017, 11(2): 1380.
[37]

Akatyeva E, Dumitrică T. J. Chem. Phys., 2012, 137(23): 234702.
[38]

Keerthi A, Sánchez-Sánchez C, Deniz O, Ruffieux P, Schollmeyer D, Feng X L, Narita A, Fasel R, Müllen K. Chem. Asian J., 2020, 15(22): 3807.
[39]

Wang X Y, Urgel J I, Barin G B, Eimre K, di Giovannantonio M, Milani A, Tommasini M, Pignedoli C A, Ruffieux P, Feng X L, Fasel R, Müllen K, Narita A. J. Am. Chem. Soc., 2018, 140(29): 9104.
[40]

Feng Z J, Mazaheripour A, Dibble D J, Wagner P, Czap G, Kladnik G, Cossaro A, Verdini A, Floreano L, Bavdek G, Ho W, Comelli G, Cvetko D, Morgante A, Gorodetsky A A. Carbon, 2020, 170: 677.
[41]

Shekhirev M, Zahl P, Sinitskii A. ACS Nano, 2018, 12(8): 8662.
[42]

Teeter J D, Zahl P, Mehdi Pour M, Costa P S, Enders A, Sinitskii A. ChemPhysChem, 2019, 20(18): 2281.
[43]

Marangoni T, Haberer D, Rizzo D J, Cloke R R, Fischer F R. Chem. Eur. J., 2016, 22(37): 13037.
[44]

Zhang Y, Zhang Y F, Li G, Lu J C, Lin X, Du S X, Berger R, Feng X L, Müllen K, Gao H J. Appl. Phys. Lett., 2014, 105(2): 023101.
[45]

Vo T H, Perera U G E, Shekhirev M, Mehdi Pour M, Kunkel D A, Lu H D, Gruverman A, Sutter E, Cotlet M, Nykypanchuk D, Zahl P, Enders A, Sinitskii A, Sutter P. Nano Lett., 2015, 15(9): 5770.
[46]

Zhang Y F, Zhang Y, Li G, Lu J C, Que Y D, Chen H, Berger R, Feng X L, Müllen K, Lin X, Zhang Y Y, Du S X, Pantelides S T, Gao H J. Nano Res., 2017, 10(10): 3377.
[47]

Cao Y, Qi J, Zhang Y F, Huang L, Zheng Q, Lin X, Cheng Z H, Zhang Y Y, Feng X L, Du S X, Pantelides S T, Gao H J. Nano Res., 2018, 11(12): 6190.
[48]

Durr R A, Haberer D, Lee Y L, Blackwell R, Kalayjian A M, Marangoni T, Ihm J, Louie S G, Fischer F R. J. Am. Chem. Soc., 2018, 140(2): 807.
[49]

Marschall R. Adv. Funct. Mater., 2014, 24(17): 2421.
[50]

Lu Y, Guo J. Nano Res., 2010, 3(3): 189.
[51]

Fu Y B, Yang H, Gao Y X, Huang L, Berger R, Liu J Z, Lu H L, Cheng Z H, Du S X, Gao H J, Feng X L. Angew. Chem. Int. Ed., 2020, 59(23): 8873.
[52]

Blackwell R E, Zhao F Z, Brooks E, Zhu J M, Piskun I, Wang S K, Delgado A, Lee Y L, Louie S G, Fischer F R. Nature, 2021, 600(7890): 647.
[53]

Shi G Q, Zhou J F, Li Z, Sun Y, Kantorovich L N, Fang Q, Besenbacher F, Yu M. Angew. Chem. Int. Ed., 2020, 59(37): 15958.
[54]

Zhang H, Xiong W, Lu J C, Cai J M. Prog. Chem., 2021, 34(3): 557.
[55]

(张辉, 熊玮, 卢建臣, 蔡金明. 化学进展, 2021, 34(3): 557. ).
[56]

Hao Z L, Ruan Z L, Yang X T, Cai Y T, Lu J C, Cai J M. Acta. Chimica. Sinica., 2017, 66(21): 218103.
[57]

(郝振亮, 阮子林, 杨孝天, 蔡逸婷, 卢建臣, 蔡金明. 化学学报, 2017, 66(21): 218103. ).

Funding

National Natural Science Foundation of China(62271238)

National Natural Science Foundation of China(61901200)

Yunnan Fundamental Research Projects(202101AV070008)

Yunnan Fundamental Research Projects(202101AW070010)

Yunnan Fundamental Research Projects(202201AT070078)

Strategic Priority Research Program of Chinese Academy of Sciences(XDB30000000)

Dongguan Innovation Research Team Program

PDF (25342KB)

2656

Accesses

0

Citation

Detail
Sections
Recommended

/

〈 〉

Address of Editorial Office: No. 33 Beisihuan Xilu, Zhongguancun , Beijing 100190, China
Tel: +86 10 8262 7757
Fax: +86 10 8262 7757
E-mail: progchem@mail.las.ac.cn
http://www.progchem.ac.cn
    • 　
    Copyright © Editorial Office of Progress in Chemistry