Fig 1 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
a Graphical explanation of the Clar sextet as being the superposition of two resonant Kekulé configurations. b-d Each of the three equivalent Clar formulas of graphene, with its √3×√3 superstructure highlighted in cyan in b. e Superposition of all three Clar formulas. Dotted circles correspond to Clar sextets within graphene, although not all simultaneously.
Source publication
Graphene nanoribbons (GNRs) make up an extremely interesting class of materials. On the one hand GNRs share many of the superlative properties of graphene, while on the other hand they display an exceptional degree of tunability of their optoelectronic properties. The presence or absence of correlated low-dimensional magnetism, or of a widely tunab...
Contexts in source publication
Context 1
... The structure of a PAH is then the superposition of all possible Kekulé bond configurations. Within this picture, the delocalization of six - electrons in a carbon hexagon due to the resonance of two Kekulé configurations with alternating single and double bonds is called a Clar sextet, pictured as a circle within the corresponding hexagon (Fig. 1a). According to Clar´s theory [7], the most representative and stable structure of a PAH is that with the highest number of Clar sextets, which is called the "Clar formula". Here it is important to keep in mind that the bonds sticking out of a Clar sextet are formally single bonds and thereby impede two neighboring hexagons to be Clar ...
Context 2
... to keep in mind that the bonds sticking out of a Clar sextet are formally single bonds and thereby impede two neighboring hexagons to be Clar sextets simultaneously. Graphene can be represented with three equivalent Clar formulas, in each of which one out of every three hexagons is a Clar sextet, arranged in a (3×3)R30º superstructure ( Fig. 1b-d). Considering a combination of the three possible Clar formulas, all hexagons in graphene can be Clar sextets and are fully equivalent (Fig. 1e). In addition, bonds within a Clar sextet are equivalent as well. Since bond length alternation (BLA) is a measure of the aromaticity of a system and one of the main causes for the opening of ...
Context 3
... Clar sextets simultaneously. Graphene can be represented with three equivalent Clar formulas, in each of which one out of every three hexagons is a Clar sextet, arranged in a (3×3)R30º superstructure ( Fig. 1b-d). Considering a combination of the three possible Clar formulas, all hexagons in graphene can be Clar sextets and are fully equivalent (Fig. 1e). In addition, bonds within a Clar sextet are equivalent as well. Since bond length alternation (BLA) is a measure of the aromaticity of a system and one of the main causes for the opening of band gaps in conjugated organic materials [8,9], graphene is an excellent example of a perfectly aromatic system with absent BLA and zero band ...
Context 4
... coupled across the ribbon [69]. Notably, the electron-electron interactions driving the magnetization concurrently open a band gap 0 between the edge states, deterring zGNRs from truely being gapless structures. This can be observed in the calculat- ed band structure of zGNRs with and without electron-electron interactions in Fig. 10b. While tight-binding density of states has only one van Hove singularity relat-ed to the edge state´s flat bands at E = 0, including the electron-electron interac- tion U term in a mean field Hubbard model results in two pairs of density of states peaks split by 0 and 1 (Fig. 10b). In particular, 0 relates to the antiferromagnet- ...
Context 5
... of zGNRs with and without electron-electron interactions in Fig. 10b. While tight-binding density of states has only one van Hove singularity relat-ed to the edge state´s flat bands at E = 0, including the electron-electron interac- tion U term in a mean field Hubbard model results in two pairs of density of states peaks split by 0 and 1 (Fig. 10b). In particular, 0 relates to the antiferromagnet- ic correlation between the two edges, while the larger splitting 1 relates to the fer- romagnetic correlation between the spins of the most strongly localized states at k ║ = /a in the same edge. It should be mentioned here that in the nearest-neighbor approximation the ...
Context 6
... that the magnetic moment per length unit increases as the chiral angle is re- duced. Tight-binding calculations of chiral GNRs of finite width as a function of the chiral angle confirm this picture [76]. For GNRs with edge orientations close to aGNRs the edge states associated to the flat bands at the Fermi level are almost completely suppressed (Fig. 10a). Instead, in ribbons with edge orientation close to a zGNR the flat edge state band extends over the whole 1D Brillouin zone and becomes multiple degenerate due to band folding (Fig. 10a) [76]. As in zGNRs, electron-electron interactions split the degeneracy of the flat bands also in chiral GNRs (Fig. 10c,d). As discussed earlier, each ...
Context 7
... picture [76]. For GNRs with edge orientations close to aGNRs the edge states associated to the flat bands at the Fermi level are almost completely suppressed (Fig. 10a). Instead, in ribbons with edge orientation close to a zGNR the flat edge state band extends over the whole 1D Brillouin zone and becomes multiple degenerate due to band folding (Fig. 10a) [76]. As in zGNRs, electron-electron interactions split the degeneracy of the flat bands also in chiral GNRs (Fig. 10c,d). As discussed earlier, each pair of peaks arising in the density of states has a different nature, whereby 0 and 1 relate to the magnetic coupling across the nanoribbon and along its edge, respectively. As can ...
Context 8
... level are almost completely suppressed (Fig. 10a). Instead, in ribbons with edge orientation close to a zGNR the flat edge state band extends over the whole 1D Brillouin zone and becomes multiple degenerate due to band folding (Fig. 10a) [76]. As in zGNRs, electron-electron interactions split the degeneracy of the flat bands also in chiral GNRs (Fig. 10c,d). As discussed earlier, each pair of peaks arising in the density of states has a different nature, whereby 0 and 1 relate to the magnetic coupling across the nanoribbon and along its edge, respectively. As can thus be intuitively expected, for a given GNR width 0 hardly changes with the chiral angle. Instead, 1 decreases in a ...
Context 9
... with decreasing chiral angle (as the edge orientation approaches that of zGNRs) [73,77]. In other words, for a given width there is a minimum number of zigzag sites (in relation to armchair sites) along the ribbon´s edges to host a notable edge state density. The required ratio of zigzag to armchair sites be- comes higher for narrower GNRs. Fig. 10 a Band dispersion obtained from tight binding calculations of GNRs of different orienta- tion, from zigzag, through chiral, to armchair. The closer to the armchair direction, the lower the density of states of the flat bands at E=0, until a band gap opens for armchair ribbons. Similar calculations for zigzag and chiral ribbons are ...
Context 10
... calculated GW band gaps still overestimate the experimental ones [34]. Best results have been obtained when combining the GW approximation with a semiclassical image-charge model to account for sub- strate screening [80]. The screening-derived band gap renormalization varies with the substrate and the GNR, being in the 1 eV range for aGNRs on Au (Fig. 11) and slightly lower for aGNRs on a NaCl bilayer on Au [80]. Its combination with the calculated GW band gaps ultimately provides a remarkably good agreement with currently available experimental values for 5 [43], 7 [50], 9 [45] and 13- aGNRS [46] (Fig. ...
Context 11
... varies with the substrate and the GNR, being in the 1 eV range for aGNRs on Au (Fig. 11) and slightly lower for aGNRs on a NaCl bilayer on Au [80]. Its combination with the calculated GW band gaps ultimately provides a remarkably good agreement with currently available experimental values for 5 [43], 7 [50], 9 [45] and 13- aGNRS [46] (Fig. ...
Context 12
... of 1.44 Å) [78]. In the following we provide Table 1 comparing the exper-imentally reported effective masses to the values predicted by either of the models mentioned above. It can be seen that there is an excellent agreement between pre- dictions and experiments, best of all when taking the experimental values obtained from ARPES measurements. Fig. 11 Calculated bandgap energies for isolated and Au-supported aGNRs of different width for each of the three sub-families (3p, 3p+1, 3p+2). Calculations are with the GW method, further adding substrate screening for the supported ribbons. Superimposed we find the currently availa- ble experimental values of Au-supported aGNRs obtained by ...
Context 13
... of GNRs does not only af- fect armchair oriented ribbons, but also ribbons with other orientations. As de- scribed in the previous section, the edge state magnetization in zigzag and chiral GNRs drives the opening of a band gap 0 and an additionally split resonance 1 related to the inter-edge and intra-edge magnetic coupling, respectively (Fig. 10) [69, 74,76]. Thus, as expected from their respective nature, 1 hardly varies with the ribbon´s width. Instead, 0 decreases with growing width, although in the zGNR case in a monotonic way and faster than for aGNRs [74][75][76]. This qualita- tive behavior of 0 and 1 is at least what results from calculations on free stand- ing ...
Context 14
... its saturation value at widths around N z = 10 [69,74]. However, it is important to consider also the relative weight of those magnetic edge states on the total density of states of the nanorib- bon. Initially they increase with increasing N z , but after a peak around N z = 7 it ends up decreasing at a rate approximately proportional to 1/N z (Fig. 12a) [73]. This effect shows how the significance of the special edge state disappears as the ribbon grows infinitely wide and becomes 2D graphene ( Fig. ...
Context 15
... total density of states of the nanorib- bon. Initially they increase with increasing N z , but after a peak around N z = 7 it ends up decreasing at a rate approximately proportional to 1/N z (Fig. 12a) [73]. This effect shows how the significance of the special edge state disappears as the ribbon grows infinitely wide and becomes 2D graphene ( Fig. ...
Context 16
... chiral ribbons than in aGNRs. Describing the band gap dependence as E g 1/N increases from for aGNRs, to larger values as the chiral angle is re-duced (e.g. for º; for º; or for º) [77]. In contrast, the overimposed band gap modulation becomes less pronounced for lower chiral angles [77,85]. Fig. 12 a Flatness index ηa, defined as the ratio between the number of states with E ≈ 0 and the total number of states in the ribbon, calculated as a function of zGNR width Nz. Density of states calculated for zGNR widths of Nz = 6 (b), Nz = 11 (c) and Nz = 51 (d), evidencing the decreasing significance of the edge states in the overall ...
Context 17
... eV per N atom per precursor molecule unit, thus making it a stronger electron acceptor (n-doping of the ribbon) [55,89,90]. The decrease of the position of the valence band by in- creasing the number of N dopants in the precursor molecule has been also charac- terized with ARPES, STS and DFT for 3-aGNRs grown on stepped Au(788) as shown in Fig. 13 a-c [36]. The electron lone pair of those nitrogen atoms being not in conjugation with the GNR π-system, it does not further affect the density of states of the frontier bands, leaving critical parameters like the band gap or the band´s effective mass ...
Context 18
... in the 7-aGNRs' back- bone, allowing a full conjugation of empty B p z orbitals with the ribbon's π- system [48,91]. According to calculations, B-doping induces the formation of a new acceptor band lying only 0.8 eV above the valence band. As a result, the band gap is strongly decreased with respect to that of pristine 7-aGNR (displayed in Fig. 13 d,e). Experimentally, the acceptor band and its distribution along the B- doped 7-aGNR backbone has been measured by STS, showing good agreement with calculations ( Fig. 13 f,g). However, the position of the valence band (and consequently the experimentally determined band gap) has not been reported yet. Fig. 13 a ARPES spectra of pristine ...
Context 19
... of a new acceptor band lying only 0.8 eV above the valence band. As a result, the band gap is strongly decreased with respect to that of pristine 7-aGNR (displayed in Fig. 13 d,e). Experimentally, the acceptor band and its distribution along the B- doped 7-aGNR backbone has been measured by STS, showing good agreement with calculations ( Fig. 13 f,g). However, the position of the valence band (and consequently the experimentally determined band gap) has not been reported yet. Fig. 13 a ARPES spectra of pristine and nitrogen doped poly-para-phenylene (PPP) polymers taken at k close to the valence band maximum. Red lines correspond to parabolic fits of the GNR´s bands revealing an ...
Context 20
... of pristine 7-aGNR (displayed in Fig. 13 d,e). Experimentally, the acceptor band and its distribution along the B- doped 7-aGNR backbone has been measured by STS, showing good agreement with calculations ( Fig. 13 f,g). However, the position of the valence band (and consequently the experimentally determined band gap) has not been reported yet. Fig. 13 a ARPES spectra of pristine and nitrogen doped poly-para-phenylene (PPP) polymers taken at k close to the valence band maximum. Red lines correspond to parabolic fits of the GNR´s bands revealing an unchanged effective mass of 0.19 m0. b Differential conductance spectra performed on PPP (black) and N2-PPP (blue), displaying the rigid ...
Context 21
... materials offers the possibility to control their optical and elec- tronic properties [92]. Thus another route to tune the band-gap of GNRs is to ex- ploit strain. To understand the effect of strain on GNRs, it is necessary first to re- vise what occurs on strained graphene. Two types of strain are usually considered: uniaxial and shear (Fig. 14a,b). When graphene is stretched in one direction, it will shrink in the perpendicular one [93]. Tight binding and first principle calcula- tions demonstrated that the effect of a uniform strain applied to the atomic lattice of graphene is to drive in the reciprocal space the position of the Dirac cone cross- ing (E D ) away from the K (K') ...
Context 22
... aGNR shear strain modifies only slightly the band structure but uniaxial strain has a relevant effect. The one dimensional Brillouin zone of aGNRs is de- fined by electronic states with allowed k values that lie on parallel lines (k=r/N+1 with r=1, 2, …,N+1) (Fig. 14c). The three families of aGNRs charac- terized by their different width (N=3p, 3p+1, 3p+2; p being an integer) have dif- ferent conditions for the crossing of their allowed k-lines with the K (K') points of the Brillouin zone (Fig. 14d). When uniaxial strain is applied E D moves away from K (K') along the strained Brillouin zone in a ...
Context 23
... is de- fined by electronic states with allowed k values that lie on parallel lines (k=r/N+1 with r=1, 2, …,N+1) (Fig. 14c). The three families of aGNRs charac- terized by their different width (N=3p, 3p+1, 3p+2; p being an integer) have dif- ferent conditions for the crossing of their allowed k-lines with the K (K') points of the Brillouin zone (Fig. 14d). When uniaxial strain is applied E D moves away from K (K') along the strained Brillouin zone in a direction perpendicular to the k-lines (Fig. 14c). In this way when E D arrives in the middle of two lines, the energy gap will be maximum; when E D coincides with a k-line, then the gap will close. This finding explains why the energy ...
Context 24
... charac- terized by their different width (N=3p, 3p+1, 3p+2; p being an integer) have dif- ferent conditions for the crossing of their allowed k-lines with the K (K') points of the Brillouin zone (Fig. 14d). When uniaxial strain is applied E D moves away from K (K') along the strained Brillouin zone in a direction perpendicular to the k-lines (Fig. 14c). In this way when E D arrives in the middle of two lines, the energy gap will be maximum; when E D coincides with a k-line, then the gap will close. This finding explains why the energy gap of aGNR is modified in a periodic way with a zigzag pattern which is different for each family (Fig. 14b). Moreover, within each family, the wider ...
Context 25
... zone in a direction perpendicular to the k-lines (Fig. 14c). In this way when E D arrives in the middle of two lines, the energy gap will be maximum; when E D coincides with a k-line, then the gap will close. This finding explains why the energy gap of aGNR is modified in a periodic way with a zigzag pattern which is different for each family (Fig. 14b). Moreover, within each family, the wider the ribbon is, the closer the k-lines are. Thus, the maximum en- ergy gap value will ...
Context 26
... for which one may combine segments with different prop- erties like those described above (edge orientation, width, doping, strain), or also other parameters like edge terminations or edge structures. Experimentally, some examples have been readily demonstrated, creating atomically sharp and precise junctions by combination of different reactants (Fig. 15) [89,98]. An alternative way to create randomly distributed GNR heterostructures is by the use of only one type of reactant that can, however, react in two different ways [99]. Fig. 15 a Schematic representation of the precursors leading to differently wide GNRs and an as- sociated heterostructure, whose staggered bandgap can be ...
Context 27
... structures. Experimentally, some examples have been readily demonstrated, creating atomically sharp and precise junctions by combination of different reactants (Fig. 15) [89,98]. An alternative way to create randomly distributed GNR heterostructures is by the use of only one type of reactant that can, however, react in two different ways [99]. Fig. 15 a Schematic representation of the precursors leading to differently wide GNRs and an as- sociated heterostructure, whose staggered bandgap can be classified as a type I heterojunction. Below, constant current STM images display the experimental realization of those heterostruc- tures. b Schematic representation of the heterostructure ...
Context 28
... of states of the heterojunction displaying the roughly unchanged bandgap across the in- terface and the type II heterojunction energy alignment. (a) Reprinted with permission from [98]. First to be realized was a type II heterojunction in which pure hydrocarbon re- actants were combined with doped reactants that included nitrogen heteroatoms (Fig. 15b) [89]. As described in Sect. 4.3 and graphically displayed in Fig. 15b, the substitution along the GNR edges of C-H by N has little impact on the elec- tronic band gap. In turn, nitrogen´s more attractive core potential lowers the onset energies of valence and conduction band and turns the N-doped GNRs into n-type semiconductors. On the ...
Context 29
... across the in- terface and the type II heterojunction energy alignment. (a) Reprinted with permission from [98]. First to be realized was a type II heterojunction in which pure hydrocarbon re- actants were combined with doped reactants that included nitrogen heteroatoms (Fig. 15b) [89]. As described in Sect. 4.3 and graphically displayed in Fig. 15b, the substitution along the GNR edges of C-H by N has little impact on the elec- tronic band gap. In turn, nitrogen´s more attractive core potential lowers the onset energies of valence and conduction band and turns the N-doped GNRs into n-type semiconductors. On the other hand, due to the particular GNR and substrate com- bination, ...
Context 30
... On the other hand, due to the particular GNR and substrate com- bination, undoped GNRs on Au are p-doped, displaying their valence band close to the Fermi level. Combination of pristine and N-doped segments into the same nanoribbon thus creates sharp p-n heterojunctions, with a band offset of around 0.5 eV for the heterojunction displayed in Fig. 15b. Because the band bending oc- curs over a distance in the order of 2 nm, the resulting electric field at the interface is extremely high (2 × 10 8 V/m), making these heterostructures highly promising for electronic device based on p-n junctions ...
Context 31
... I heterojunctions have been synthesized combining precursors that lead to differently wide GNRs [98]. The resulting structure is thus a width-modulated GNR as shown in Fig. 15a, with segments of 7-aGNRs and of 13-aGNRs. The lat- ter has a larger band gap than the former, resulting in the straddling gap evolution that characterizes type I heterojunctions. Because the 7-aGNR segments serve as energy barrier for charge carriers localized along the 13-aGNR segment, the for- mation of quantum well states can be ...
Context 32
... alternative way for the creation of quantum dots embedded inside 7-aGNRs has been demonstrated by mixing pristine 7-aGNR with a small amount of boron doped 7-aGNRs precursors (Fig. 16a) [100]. The pristine regions in these hybrid ribbons preserve the electronic structure of 7-aGNR, while the borylated segments lack an energy level aligned with the pristine VB. As a result, the VB electrons on the pristine segments become confined by the boron atom pairs, since the VB ends abruptly over the borylated regions (Fig. ...
Context 33
... precursors (Fig. 16a) [100]. The pristine regions in these hybrid ribbons preserve the electronic structure of 7-aGNR, while the borylated segments lack an energy level aligned with the pristine VB. As a result, the VB electrons on the pristine segments become confined by the boron atom pairs, since the VB ends abruptly over the borylated regions (Fig. ...
Context 34
... calculated transmission function (Fig. 16c) for these free-standing hybrid 7-aGNR, that is, the transmission of electrons from one side of the ribbon to the other, shows that the boron pairs are very efficient reflectors and that the VB-1 acts as a transmission channel even though the VB is confined (Fig. 16d). This ac- tually implies that boron atoms selectively confine VB ...
Context 35
... calculated transmission function (Fig. 16c) for these free-standing hybrid 7-aGNR, that is, the transmission of electrons from one side of the ribbon to the other, shows that the boron pairs are very efficient reflectors and that the VB-1 acts as a transmission channel even though the VB is confined (Fig. 16d). This ac- tually implies that boron atoms selectively confine VB electrons while leaving the VB-1 unaffected. The reason behind this selectivity stems from the symmetry be- tween the boron induced states and the 7-aGNR bands [100]. These results high- light that the use of substitutional heteroatoms as dopants, and their related ...
Context 36
... properties it is in electronic and optoelectronic devices applications where they arouse most attention. As a re- sult, research efforts have pushed towards the integration of GNRs as active com- ponents in electronic devices like sensors [109,110], photodetectors [111][112][113] and field effect transistors (FETs) [114][115][116][117][118][119]. Fig. 16 a Schematic representation of the hybrid 7-aGNR, consisting in extended pristine regions and disperse borylated segments. b dI/dV maps of a pristine region (dashed blue square) en- closed between two borylated segments. The maps show an increasing number of modes in the conductance with increasing negative bias, fingerprint of the VB ...
Context 37
... have been addressed locally in ultra-high-vacuum by STM with single ribbon precision [121]. Transport occurs in a tunneling regime and the tun- neling decay length through a 7-aGNR measured at different bias voltages reveals the dependence of conductance with the ribbon electronic states, that is, energy level alignment and band gap value (Fig. 17a-b). Graphene nanoribbons technology is still in its infancy and a few problems need still to be solved to speed up the use of GNRs in current technology [4]. On- surface chemistry, as outlined in the previous pages, could overcome the problems of high quality fabrication and functionalization of GNRs. In fact, it allows pro- ducing ...
Context 38
... last problem of GNR technology is the final fabrication and processing of the device. A few prototypes of field effect transistors have been realized exploit- ing on-surface synthesized GNRs (Fig. 17 c-d) [118,119]. The interest on FETs stems on the extreme thinness of GNRs which is expected to enable fabricating FETs with very short channels. This should significantly increase the device speed, while avoiding the unfavorable short-channel effects of standard CMOS technologies [123]. In GNR-FETs the driving current is limited by the ...
Similar publications
Graphene is a type of two-dimensional material with special properties and complex mechanical behavior. In the process of growth or processing, graphene inevitably has various defects, which greatly influence the mechanical properties of graphene. In this paper, the mechanical properties of ideal monolayer graphene nanoribbons and monolayer graphen...
We investigate the optical response of graphene nanoribbons (GNRs) using the broadband nonlinear generation and detection capabilities of nanoscale junctions created at the LaAlO$_3$/SrTiO$_3$ interface. GNR nanoclusters measured to be as small as 1-2 GNRs in size are deposited on the LaAlO$_3$ surface with an atomic force microscope tip. Time-reso...
In this article, we report directly patterned graphene nanoribbons by laser irradiation. The transformation of polymer into graphene nanostructure by breaking oxygen bonding and followed by graphitization of carbon atoms has been employed by using the photochemical technique. The scalable and fast growth of graphene nanoribbon was carried out by ir...
High quality graphene nanoribbons epitaxially grown on the sidewalls of silicon carbide (SiC) mesa structures stand as key building blocks for graphene-based nanoelectronics. Such ribbons display 1D single-channel ballistic transport at room temperature with exceptionally long mean free paths. Here, using spatially-resolved two-point probe (2PP) me...
In this work we investigate the ballistic transport of electrons through three-terminal graphene-based devices. The system consists of a Y-shaped junction formed by three armchair-edged graphene nanoribbons with a rectangular gate potential applied to one of the output branches, whereby current control can be established by the controlling of the r...
Citations
... The interest in graphene nanoribbons (GNRs) has surged since graphene was first isolated, largely due to the potential for fine-tuning their physicochemical properties by manipulating their size and shape 1,2 . The tailored fabrication of these one-dimensional systems, employing atomically precise control over their width and edge structure, enables the synthesis of GNRs with variable electronic band gaps [3][4][5][6][7] , specific topological quantum phases 8,9 or even spin-polarized edge states 10 in a single material class. ...
Graphene nanoribbons (GNRs) exhibit a broad range of physicochemical properties that critically depend on their width and edge topology. While the chemically stable GNRs with armchair edges (AGNRs) are semiconductors with width-tunable band gap, GNRs with zigzag edges (ZGNRs) host spin-polarized edge states, which renders them interesting for applications in spintronic and quantum technologies. However, these states significantly increase their reactivity. For GNRs fabricated via on-surface synthesis under ultrahigh vacuum conditions on metal substrates, the expected reactivity of zigzag edges is a serious concern in view of substrate transfer and device integration under ambient conditions, but corresponding investigations are scarce. Using 10-bromo-9,9':10',9''-teranthracene as a precursor, we have thus synthesized hexanthene (HA) and teranthene (TA) as model compounds for ultrashort GNRs with mixed armchair and zigzag edges, characterized their chemical and electronic structure by means of scanning probe methods, and studied their chemical reactivity upon air exposure by Raman spectroscopy. We present a detailed identification of molecular orbitals and vibrational modes, assign their origin to armchair or zigzag edges, and discuss the chemical reactivity of these edges based on characteristic Raman spectral features.
... Atomically precise graphene nanoribbons (GNRs) are increasingly being hailed as the potential future of molecular electronics due to the tunability of their electronic properties, which can be altered via adjusting their length, width, edge structure, and the addition of heteroatoms or other doping groups. 1,2 Indeed, preliminary devices based on GNRs of various widths and armchair-oriented edges (aGNRs) have already been demonstrated, such as field-effect transistors 3,4 and sensors. 5 The advantage of aGNRs is that their edges are relatively unreactive, allowing for their synthesis directly in solution 6,7 or, if synthesized in an inert vacuum environment, surviving transfer through ambient conditions at room temperature (and the rest of the processes associated with the device implementation) while still maintaining their electronic properties. ...
Nanostructured graphene has been widely studied in recent years due to the tunability of its electronic properties and its associated interest for a variety of fields, such as nanoelectronics and spintronics. However, many of the graphene nanostructures of technological interest are synthesized under ultrahigh vacuum, and their limited stability as they are brought out of such an inert environment may compromise their applicability. In this study, a combination of bond-resolving scanning probe microscopy (BR-SPM), along with theoretical calculations, has been employed to study (3,1)-chiral graphene nanoribbons [(3,1)-chGNRs] that were synthesized on a Au(111) surface and then exposed to oxidizing environments. Exposure to the ambient atmosphere, along with the required annealing treatment to desorb a sufficiently large fraction of contaminants to allow for its postexposure analysis by BR-SPM, revealed a significant oxidation of the ribbons, with a dramatically disruptive effect on their electronic properties. More controlled experiments avoiding high temperatures and exposing the ribbons only to low pressures of pure oxygen show that also under these more gentle conditions the ribbons are oxidized. From these results, we obtain additional insights into the preferential reaction sites and the nature of the main defects that are caused by oxygen. We conclude that graphene nanoribbons with zigzag edge segments require forms of protection before they can be used in or transferred through ambient conditions.
... The reason for this is that while samples prepared from 2,2´-DBBA reveal no lateral fusion of the chiral ribbons after annealing to temperatures of 590 K (Fig. S5a), starting from TBBA at a similar coverage we observe wider, fused GNRs readily at 530 K (Fig. S5b). The additional halogenation thus appears to be an efficient way to increase the chiral ribbons´ widths (with its associated impact on their electronic properties [23] ) with only mild annealing treatments. ...
Within the collection of surface-supported reactions currently accessible for the production of extended molecular nanostructures under ultra-high vacuum, Ullmann coupling has been the most successful in the controlled formation of covalent single C-C bonds. Particularly advanced control of this synthetic tool has been obtained by means of hierarchical reactivity, commonly achieved by the use of different halogen atoms that consequently display distinct activation temperatures. Here we report on the site-selective reactivity of certain carbon-halogen bonds. We use precursor molecules halogenated with bromine atoms at two non-equivalent carbon atoms and found that the Ullmann coupling occurs on Au(111) with a remarkable predilection for one of the positions. Experimental evidence is provided by means of scanning tunneling microscopy and a rationalized understanding of the observed preference is obtained from density functional theory calculations.
... The most successful realization of on-surface synthesis is the creation of graphene nanoribbons (GNR) with perfectly defined atomic structure. [5][6][7] While such one-dimensional (1D) approach remains rather easily accessible and produces polymeric wires up to the micrometer range, [8][9][10][11][12] the quest for producing two-dimensional (2D) covalent structures, or 2D polymers, [13] is highly challenging. The formation of surface-supported 2D covalent structures can be seen as an extension of graphene growth, whereby the extraordinary properties of this material [14] are improved further by controlling the atomic scale structure and the introduction of heteroatoms. ...
Covalent networks formed by on‐surface synthesis usually suffer from the presence of a large number of defects. We report on a methodology to characterize such two‐dimensional networks from their experimental images obtained by scanning probe microscopy. The computation is based on a persistent homology approach and provides a quantitative score indicative of the network homogeneity. We compare our scoring method with results previously obtained using minimal spanning tree analyses and we apply it to some molecular systems appearing in the existing literature.
... Within each family, the bandgap scales inversely with GNR width [6]. Recent advances in the onsurface synthesis of GNRs have allowed to reach the required selectivity and atomic control over width and edge structure [7,8]. Scanning probe microscopy and spectroscopy studies have confirmed the intimate structure-property relationship by providing morphological and electronic information at the atomicscale [1,4,9,10]. ...
The on-surface synthesis of graphene nanoribbons (GNRs) allows for the fabrication of atomically precise narrow GNRs. Despite their exceptional properties which can be tuned by ribbon width and edge structure, significant challenges remain for GNR processing and characterization. In this contribution, we use Raman spectroscopy to characterize different types of GNRs on their growth substrate and to track their quality upon substrate transfer. We present a Raman-optimized (RO) device substrate and an optimized mapping approach that allows for acquisition of high-resolution Raman spectra, achieving enhancement factors as high as 120 with respect to signals measured on standard SiO2/Si substrates. We show that this approach is well-suited to routinely monitor the geometry-dependent low-frequency modes of GNRs. In particular, we track the radial breathing-like mode (RBLM) and the shear-like mode (SLM) for 5-, 7- and 9-atom wide armchair GNRs (AGNRs) and compare their frequencies with first-principles calculations.
... The three coinage metal surfaces also exhibit distinct efficiency for some reactions involving two or more reactive groups. It is known that Au (111) and Ag (111) have been often employed as the surfaces to synthesize the classical straight armchair 7-GNRs, by using 10,10'-dibromo-9,9'-bianthryl (DBBA) as the precursor (Fig. 2a) [58,135]. In the first step, Ag (111), respectively. ...
On-surface synthesis has been one of the hottest research fields in surface science in the last decade, owing to its great potential for bottom-up synthesis of functional molecules and covalent nanomaterials. Compared to classical in-solution chemistry, all of the on-surface reactions are done without solvent, thus very minimal byproducts and no limitation of solubility are involved. However, because of its typically required ultra-high vacuum conditions, where only limited catalysts can be used, a key challenge for on-surface synthesis is the precise control of the reaction pathway. Countless efforts have been made for controllable synthesis of target chemical structures on surfaces by distinct strategies. These strategies can be summarized under following aspects: 1) rational choice of surfaces; 2) template effects based on two-dimensional (2D) environments; 3) on-surface thermodynamic and kinetic controls; 4) the participation of chemisorbed nonmetal adatoms on surfaces. This report reviews the recent progress toward the control of on-surface synthesis and raises a series of questions at the end, which deserve further explorations in the future.
... However, its relevance should be explicitly acknowledged, being the basis for many of the milestones along the development of "on-surface synthesis" in a broader perspective. The following are offered by way of examples: it allowed in one of the early works in 2008 the selective synthesis of fullerenes from aromatic precursor molecules, 67 it is a key step in the synthesis of graphene nanoribbons 68 or nanographenes, 69 and it has allowed the study of highly reactive species like arynes 28,29,70 as well as the synthesis of elusive molecules like triangulene 71 or higher acenes. 19 ...
... 335 One of the sources for such great versatility is the enormous variability they display in their optoelectronic properties as a function of their precise atomic structure. 9,68,336 On the other hand, the marked dependence between structure and functionality underlines the strict requirement of atomic precision in the synthesis of the GNRs. To date, this has only been achieved from the bottom-up approach. ...
... Although in this respect there have also been important advances in wet-chemistry, 337−341 the first successful proof of atomic precision in GNRs was obtained by on-surface synthesis, which still remains the approach whereby the largest pool of GNRs have been grown. 9,68,151 Thus, it is the great potential for applications, coupled with on-surface synthesis as a viable route to fully exploit it, which has acted as the main driving force behind the fascinating developments in GNR synthesis. Sharing many of the superlative properties of its parent material graphene, GNRs owe the additional tunability of their electronic properties to the boundary conditions imposed by the ribbon's edges. ...
On-surface synthesis is appearing as an extremely promising research field aimed at creating new organic materials. A large number of chemical reactions have been successfully demonstrated to take place directly on surfaces through unusual reaction mechanisms. In some cases the reaction conditions can be properly tuned to steer the formation of the reaction products. It is thus possible to control the initiation step of the reaction and its degree of advancement (the kinetics, the reaction yield); the nature of the reaction products (selectivity control, particularly in the case of competing processes); as well as the structure, position, and orientation of the covalent compounds, or the quality of the as-formed networks in terms of order and extension. The aim of our review is thus to provide an extensive description of all tools and strategies reported to date and to put them into perspective. We specifically define the different approaches available and group them into a few general categories. In the last part, we demonstrate the effective maturation of the on-surface synthesis field by reporting systems that are getting closer to application-relevant levels thanks to the use of advanced control strategies.
Photon-energy dependence of photoemission from seven-atoms-wide armchair graphene nanoribbons is studied experimentally and theoretically up to hν = 95 eV. A strong hv dependence of the normal emission from the valence band maximum (VB 1 ) is observed, sharply peaked at hv = 12 eV. The detailed analysis of the light-polarization dependence of the photoemission from VB 1 unambiguously
characterizes the symmetry of the state. The experimental observations are analyzed based on ab initio one-step theory of photoemission. Off-normal emission is studied in detail and its relation to the standing-wave character of the valence band states is discussed. Excellent agreement with the earlier experiment [Senkovskiy et al. 2018 2D Materials 5 035007] is obtained. Rapid variations of the intensity with the ribbon transverse photoelectron momentum are predicted from the ab initio theory, which are at variance with the prediction of the popular tight-binding rigid-wall model. These findings are instrumental for the study of the electronic structure of nanoribbons with angle-resolved photoemission. Moreover, the strong enhancement of the photoyield could trigger the GNR application as narrow-band photodetectors and contribute to the design of novel photocathodes for vacuum ultraviolet photodetection.
Graphene nanoribbons (GNRs) are lengthened one-dimensional monolayer strips of graphene and have a hexagonal honeycomb lattice structure. The captivating properties like electrical conductivity, emerging band gap, optical property, thermal conductivity, high mechanical strength, and ultrahigh surface area make them a better candidate for biomedical applications. The properties can be significantly reformed and controlled by altering the edge functionalities and geometry. The exhibition of a wide potential window coupled with an ultra-high surface area to host sensing element makes GNR an excellent biosensing platform. Consequently, biosensing is one of the most explored applications of GNR. This review presents an overview of the characteristics, methods of synthesis, and biosensing applications of GNR. Overall, GNR is considered a promising platform for efficient signal transduction compared to conventional biosensing platforms. Further, it offers high electrical conductivity, large surface area, high adsorption, synergistic effects with combined materials, fast response, sensitivity, and selectivity.
We demonstrate a nanoscale photoconductive photodetector with seven-atom wide armchair-edge graphene nanoribbons as the active material. The detector responsivity is 0.04 mAW ⁻¹ with a dark current below 30 pA under a bias voltage of 1.5 V.
