Graphene has attracted enormous attention due to its extraordinary combination of electronic, mechanical, thermal and optical properties . In recent years, graphene research has been unprecedentedly active, and the progress in fabrication and characterization of graphene has been very significant. As a result, a decade after the first experimental studies of graphene, there is an active discussion of how graphene’s properties will be implemented in a broad spectrum of applications in the near future [1a, 2]. One of the most highly praised properties of graphene, its exceptionally high charge-carrier mobilities, makes it promising for electronics applications [1a, 2, 3]. However, since graphene is a semimetal, it does not have an energy bandgap, which prevents its use in logic devices [1b, 4]. Numerous studies have focused on different ways to open a bandgap in graphene. One approach for tuning electronic properties of graphene relies on making few-nanometer-wide strips of graphene that are typically referred to as graphene nanoribbons (GNRs).
Interestingly, GNRs have been a subject of theoretical studies long before the first experiments on isolated graphene samples by Novoselov, Geim et al. in 2004 . A comprehensive theoretical analysis of GNRs has been reported by Nakada et al. in 1996 . In general, GNRs can have two principle edge structures – armchair (Figure 1(a)) and zigzag (Figure 1(b)) – in addition to an infinite number of mixed, chiral and disordered edges. Figure 1(a) and (b) also shows the convention on how the carbon atoms are counted across zigzag and armchair GNRs to distinguish ribbons with different widths. For example, for the armchair GNR shown in Figure 1(a), N = 10, so this GNR is typically referred to as 10-AGNR, whereas for the zigzag GNR shown in Figure 1(b), N = 5, so this is 5-ZGNR.
It was shown that the electronic properties of armchair and zigzag GNRs are very different. Narrow armchair GNRs can exhibit substantial energy bandgaps that were predicted to be very sensitive to the ribbon’s width. Figure 1(c) shows the results of tight-binding calculations by Nakada et al. for armchair GNRs with N = 4, 5 and 6 . According to the results of these calculations, 4-AGNR is a semiconductor. However, if the width of this GNR is increased by just one carbon atom, the calculated energy bandgap decreases to zero, and the resulting 5-AGNR becomes metallic. If one more carbon atom is added to this ribbon, the resulting 6-AGNR becomes a semiconductor again (Figure 1(c)). While nanoscience knows many examples of nanomaterials, such as semiconductor quantum dots  and plasmonic metallic nanoparticles , which have size-dependent physical properties, this particular example, where the change in one dimension of a nanostructure by just one atom changes physical properties in such a dramatic way, is among the most impressive ones.
Figure 1(d) further illustrates the dependence of the bandgap of N-AGNRs on their width according to the results of tight-binding calculations . Depending on the N value, an N-AGNR may belong to one of three different families with N = 3p, N = 3p + 1 or N = 3p + 2, where p is a positive integer (accordingly, N-AGNRs in Figure 1(c) provide examples of nanoribbons from all these three families). For N-AGNRs with N = 3p or N= 3p + 1 the bandgap is inversely proportional to the ribbon’s width, while N-AGNRs with N = 3p + 2 are metallic (Figure 1(d)) . The results may change if a different computational model is used [7, 10]. Figure 1(e) shows energy bandgaps for the same armchair GNRs that were obtained using first-principle calculations . In this model, there are no metallic AGNRs, although N-AGNRs from the N = 3p + 2 family still have much lower bandgaps compared to the N-AGNRs from N = 3p and N = 3p + 1 families (Figure 1(e)). These three families of AGNRs are now well separated, and within each family the bandgap is inversely proportional to the ribbon’s width.
Depending on the calculation method used, zigzag GNRs are predicted to be either metallic or narrow bandgap semiconductors with much less-pronounced width dependence than for armchair GNRs  (Figure 1(f) and (g)). However, unlike armchair GNRs, they possess nonbonding edge states that were predicted to exhibit a peculiar type of magnetic ordering [6, 7, 11]. The magnetically ordered state of zigzag GNRs can be controlled by means of an external electric field , which may find use in spintronics, an emerging area of electronics that uses electron’s spin instead of charge for data transmission and storage .
Computational study of GNRs is an active area of research. Nanoribbons with different widths, geometries and edge structures have been investigated using various calculation approaches; a discussion of advances in this area can be found in the recent review papers . While we do not intend to provide an in-depth review of these studies in this chapter, there are several practically important points that have to be made. On the one hand, even a brief glimpse at the results of computational studies reveals a wide breadth of properties that could be potentially realized in properly designed GNRs, including a largely tunable energy bandgap, half metallicity and nanomagnetism. On the other hand, these computational results can be used to formulate requirements to experimental approaches for the fabrication of GNRs with desired properties. For example, the following requirements should be met to achieve semiconductor characteristics in armchair GNRs (Figure 1):
In order to achieve practical energy bandgaps of >0.5 eV, AGNRs should be less than 2 nm wide (Figure 1(d)).
Not only should the AGNRs be narrow, but their width should be the same along their length with single atom precision. As we have already discussed, addition or subtraction of a single carbon atom from the width of a GNR can make a dramatic difference to the ribbon’s energy gap (Figure 1(c)).
The ribbons should have atomically precise edges. The discussed calculations show a clear difference between GNRs with armchair and zigzag edges. Other computational studies have specifically focused on the edge disorder in graphene ribbons and showed that even a weak edge roughness could have a strong effect on the charge carrier transport in GNRs .
In practice, these requirements are very difficult to meet. While with the state-of-the-art electron-beam lithography it is possible to achieve a ~10 nm resolution, this is not sufficient to carve a sub-2-nm wide GNR from a graphene sheet, not to mention to achieve an atomically precise width and edge structure of a ribbon. In addition to nanofabrication , other top-down approaches that rely on fabrication of GNRs from larger blocks of sp2 carbon materials, such as graphite or carbon nanotubes, have been developed and tested over the last decade. These methods include sonochemical method , nanowire lithography , nanoscale cutting of graphene using nickel nanoparticles  or a diamond knife , meniscus-mask lithography , chemical etching of prefabricated graphene nanostructures  and unzipping of carbon nanotubes  among others. Top-down approaches typically yield GNRs with width >10 nm and have limited control over their edge structure. Although several groups demonstrated that such GNRs could exhibit an insulating state in electrical measurements, it was later argued that the observed transport bandgaps of up to ~200–400 meV [16b, 17, 18] are likely to be caused by strong localization effects due to edge disorder, rather than a true gap between valence and conduction bands [15c, 24].
Despite the great progress in the development of top-down fabrication approaches for GNRs over the last decade, none of the currently existing methods meet the requirements (1–3). This is why the 2010 paper from Müllen and Fasel groups, which demonstrated the potential of the bottom-up chemical approaches for the synthesis of narrow atomically precise GNRs, was a game changer in the field . In this paper, ribbons that are only a few benzene rings wide and have atomically smooth armchair edges were synthesized on a surface of either Au(111) or Ag(111) single crystal by coupling molecular precursors into linear polyphenylenes followed by cyclodehydrogenation. This paper demonstrated that bottom-up techniques could yield narrow atomically engineered GNRs that meet the requirements (1–3) and are currently unachievable by any existing top-down approach.
Since 2010, the surface-assisted synthesis has become a popular approach for the preparation of atomically precise GNRs . While the surface-assisted synthesis produces very high-quality GNRs that can serve as ideal test objects for the fundamental studies of nanoscale graphene-based materials, this approach has several important practical limitations. First of all, the surface-assisted synthesis cannot be scaled up to produce bulk quantities of atomically precise GNRs for large-scale applications. Also, GNRs prepared on a conductive Au(111) or Ag(111) single crystal cannot be directly used for device fabrication and electrical testing and thus should be somehow transferred to a dielectric substrate. Because of these limitations, the original report on the on-surface synthesis of GNRs  stimulated interest in the development of alternative solution-based bottom-up approaches for atomically precise GNRs, which will be the topic of this chapter.
Chemistry-wise, synthesis of a GNR from molecular precursors is the synthesis of a conjugated polymer. Examples of successful fabrications of ladder-type polymers were well known for years  and Müllen’s group in particular has been notably active in this area, attempting the bottom-up synthesis of “graphite ribbons” in 2000 and 2003 , several years before the start of the general interest in graphene in 2004 . However, the ladder-type polymers are too narrow and do not show the desired properties, such as high carrier mobilities, while the early attempts to synthesize wider “graphite ribbons” showed a lack of structural uniformity. It should also be noted that carrier mobilities depend on the width of the ribbons ; therefore relatively wide and atomically precise GNRs are of considerable interest. Even though a number of different approaches, strategies and designs of GNR structures have been proposed and implemented since then, the field of bottom-up synthesized GNRs is still in its infancy and many fascinating discoveries and developments are yet to be made.
2 Structure of Solution-Synthesized GNRs
As previously mentioned, the edge type and the width determine the electronic properties of GNRs. Thus, one of the most attractive advantages of the bottom-up synthetic approaches is that they provide a complete control over the edge structure. Figure 2 shows some of the GNR core aromatic structures synthesized in solution. In addition to straight armchair GNRs (Figure 2(a)) , several types of nanoribbons, such as chevron-type GNRs (Figure 2(b)) , cove-type GNRs (Figure 2(c)) , “kinked” GNRs (Figure 2(d)) , necklace-like GNRs (Figure 2(e))  and “chiral” or mixed GNRs (Figure 2(f)) , have been designed and synthesized. One particular type of bottom-up GNRs, poly-peri-naphthalene or 5-AGNR, which belongs to the N = 3p + 2 family and therefore, depending on the computational method, is predicted to either be metallic  or have a small bandgap , will not be discussed in this chapter, as the extensive studies of this and related GNRs have been recently reviewed in great detail . Notably, the variety of possible structures is almost infinite, and tuning the precursor in order to achieve the optimum GNR structure for a specific application is another important advantage of the bottom-up approach. For example, chevron-type GNRs 3 were predicted to have larger exciton binding energies than straight armchair GNRs, which makes them promising for nano-optoelectronic applications .
Same GNR structure can often be synthesized using different routes and strategies with their own advantages and limitations. At this point, rational design of the synthesis and choice of the precursors play a very important role for achieving high-quality nanoribbons. For example, Figure 3 (a)–(e) shows five different routes to synthesize GNRs with the core structure 1, where all methods have certain disadvantages [28b, 30a]. For example, because of the high steric hindrance in routes (a) and (b), polymerizations of the monomer 11 or the monomers 13 and 14 do not provide the corresponding polymers 12 and 17. In case of less-hindered monomers 18 and 14 (route (d)), only oligomers were found after the polymerization, which do not form the ribbon oligomer 1 upon the following cyclodehydrogenation step. When a successful polymerization is performed, such as the polymerization of monomers 20 and 21 in route (e), the formation of structural isomers during the Diels–Alder reaction results in a mixture of irregular GNRs, but not exclusively the GNR 1. Finally, Suzuki coupling of precursors 15 and 16 to form polymer 17, which was followed by the cyclodehydrogenation into GNR 1, was successfully performed (route (c)) [30a]; however, the resulting GNRs were only 12 nm long, which may be too short for device fabrication. It was argued that the synthesis of GNR 1 was not successful due to high steric hindrance, as well as due to rigid backbone of the polymers, resulting in limited solubility of the polymeric products. Alternatively, a new GNR 6 was designed and synthesized from less-hindered precursors 23 and 24 that produced a “kinked” and more soluble polymer 25. The resulting GNRs 6 were 25 nm long in contrast to 12-nm-long GNRs 1 (Figure 3(f)) [30a, 33a]. The examples in Figure 3 show that the bottom-up solution synthesis of GNRs, despite the large number of powerful tools in organic chemistry, still requires development of new synthetic approaches and routes toward high-quality nanoribbons of appreciable length.
3 Synthetic Approaches Toward Atomically Precise GNRs
Several successful attempts to synthesize atomically precise GNRs in solution have been reported to date. In general, the synthesis consists of two steps: polymerization of rationally designed monomers and planarization of the resulting polymer structure to yield polyaromatic graphene molecules. The majority of the syntheses utilize Suzuki, Yamamoto or Diels–Alder polymerization reactions, and oxidative cyclodehydrogenation using FeCl3 for the planarization step.
3.1 Suzuki Coupling
The synthesis of GNR 1 (Figure 3(c)), which was discussed earlier, was one of the first successful syntheses of atomically precise GNRs. It employed Suzuki cross-coupling as a polymerization reaction [30a]. Similarly, the synthesis of “kinked” GNRs 6 with improved solubility and polymerization efficiency was also performed through Suzuki polymerization (Figure 3(f)) [33a]. Later, the synthesis of “kinked” structures was extended to wider GNRs using naphthalene diboronic ester 26 or anthracene diboronic ester 28, as shown in Figure 4(a) [33a]. Despite the successful cyclization step for the narrow version of GNRs 6, the cyclodehydrogenation reaction for naphthalene and anthracene units to form GNRs 7 and 8 was shown to be less effective with degrees of cyclization of 78% and 75%, respectively. Notably, the “kinked” GNRs were used to fabricate thin-film field-effect transistors (FETs), and wider GNRs 8, as expected, exhibited higher charge carrier mobilities.
Suzuki coupling was used to synthesize not only “kinked” GNRs but also necklace-like GNRs 9 (Figure 4(b)) . Reactive sites in the precursor 30 are less hindered than in the precursor 16, but the large size of the laterally extended structure 31 and the rigidity of the poly(p-phenylene) backbone are responsible for the relatively short length (13 nm) of GNRs 9 obtained in this synthesis .
3.2 Yamamoto Coupling
One of the problems with Suzuki cross-coupling polymerization reactions presented in Figure 4 is the need for two different monomers. This is why Yamamoto homocoupling reaction, which requires only one dihalogenated monomer, became an attractive alternative to prepare long GNR precursors.
Several synthetic procedures for GNRs that involve Yamamoto coupling were recently reported (Figure 5). Historically, the first synthesis of GNRs employing Yamamoto coupling was reported in 2012 . The monomer 32 was polymerized to form polymer 33 (Figure 5(a)) in a notably more efficient way in comparison to previously discussed Suzuki-coupled polymers, and the polymer 33 was then converted to the chiral GNR 10 . Later, an even wider GNR 2 with an estimated width of ~2.1 nm was attempted, starting with the monomer 34 (Figure 5(b)), but the resulting polymers were rather short and could be considered as large nanographene molecules [30a]. The nanographenes were sublimed onto a substrate in order to fabricate FET devices, which showed better results in comparison to liquid-phase deposited molecules .
In 2014, our group reported a gram-scale synthesis procedure for chevron-type GNRs 3 and demonstrated their high structural quality . Notably, the solubilizing alkyl groups that are usually employed to improve solubility of GNRs were not attached to the aromatic core of the GNRs 3. This significantly facilitated characterization of solution-synthesized GNRs 3 using scanning tunneling microscopy (STM) to visualize their atomically precise aromatic structure. Even though the absence of the solubilizing groups limited the liquid-phase processability of the nanoribbons, the Yamamoto coupling as well as the following oxidative cyclodehydrogenation were shown to be very effective .
3.3 Diels–Alder Reaction
In attempts to achieve more efficient polymerization, excellent results were shown for GNRs synthesized via Diels–Alder polymerization reaction [32a]. Cove-type GNR precursor 39 was synthesized using a relatively simple procedure of refluxing monomer 38 in diphenyl ether or by heating it in a melt (Figure 6 (a)). This elegant approach, which does not require any reagents other than the monomer itself to perform the polymerization, not only yielded GNRs with novel “cove-type” edges, but also provided polymer precursors 39 and GNRs 4 with lengths of over 600 nm. These GNRs are not only significantly longer than nanoribbons produced using other coupling approaches, but also long enough to serve as channels of electronic devices prepared by the standard electron-beam lithography . Another attractive feature of this GNR design is the possibility of the lateral extension of nanoribbons without changing the overall chemical approach. The wider cove-type GNR 5 was synthesized using the same Diels–Alder polymerization reaction from monomer 40, see Figure 6 (b). Because of their larger width, GNRs 5 have a smaller optical bandgap of ~1.2 eV compared to ~1.9 eV for the original GNR 4 [32a]. Noncontact ultrafast terahertz photoconductivity measurements revealed high intrinsic mobility of GNRs 4 and 5 [32, 40], proving the effectiveness of the cyclodehydrogenation reaction and the high quality of both cove-type nanoribbons.
Recently, another type of Diels–Alder-like cycloaddition reaction, aza-Diels–Alder or Povarov reaction, was proposed to synthesize atomically precise GNRs in solution (Figure 7) . It is based on the synthesis of polybenzoquinoline precursor 43 from alkyne- and aldimine-modified units 42. However, the synthesis of GNR 44 is not reported yet.
3.4 Planarization of Synthesized Polymers
The last step in the solution synthesis of GNRs is the graphitization of a 3D polymer fabricated at the polymerization step. The most common procedure is the Scholl reaction, an oxidative cyclodehydrogenation using iron (III) chloride as an oxidant and a Lewis acid. This easy procedure has shown excellent to good yields for a number of polyaromatic systems . However, in some cases, it results in a mixture of partially cyclized or chlorinated products preventing its use for the synthesis of atomically precise aromatic structures . A number of other cyclodehydrogenation techniques and reagent systems have been developed for the synthesis of polyaromatic hydrocarbons (PAHs), which could be adapted for the synthesis of GNRs if for some reason the FeCl3-based reaction does not provide a desired product at a good yield. Other possible reagents include AlCl3/NaCl , MoCl5 , dichlorodicyanobenzoquinone (DDQ)/CH3SO3H , CuCl2 or Cu(OTf)2/AlCl3 , bis(trifluoroacetoxy)iodobenzene (PIFA)/BF3·Et2O  and others. Notably, the choice of the reagent is important not only to achieve a product of cyclodehydrogenation at a good yield (see the example in Figure 8(a) [43a, 49]), but also to control the direction of the reaction (see the example in Figure 8(b)) , which could potentially be used in syntheses of new types of polyaromatic structures. More details on the Scholl reaction can be found in the recent review paper by Butenschön, Gryko et al. [43a].
3.5 Alternative Approaches
As alternatives to the polymerization reactions described above, other polymerization techniques could also be employed, especially if the reactions fail or do not yield the desired product. For example, polymerization of monomer 51 using a Yamamoto coupling reaction was not successful, probably due to the high steric hindrance (Figure 9) [43a]. The oligomer products 52 were obtained via Ullmann coupling and purified by column chromatography. Furthermore, the following cyclodehydrogenation with FeCl3 yielded mainly chlorinated products, and a DDQ/CF3SO3H mixture was employed to achieve cove-type GNR oligomers 53.
In addition to the concept of alkyne cyclization from Swager and coworkers , interesting approaches have been recently proposed for the synthesis of polycyclic aromatic hydrocarbons (PAHs) and GNRs from polyynes (Figure 10). The first approach is based on the controllable radical cascade reactions to form central bonds of nanoribbons, while the bonds on the periphery could be formed via oxidative dehydrogenation (Figure 10(a)) . Two reactions, namely “all-exo” and “all-endo”, have been developed and tested for short oligomers (Figure 10(b) and (c)); however, long GNRs have not been synthesized yet. The second approach utilizes Cu(OTf)2-catalyzed benzannulation reaction of poly(phenylene ethynylene) precursor 56 with 2-(phenylethynyl)benzaldehyde (Figure 10(d)) to yield polymer 57 , which may eventually be converted to a GNR with the core structure 58 (Figure 10(e)).
4 Chemical Modification
As discussed in the previous sections of this chapter, bottom-up synthetic approaches allow a precise control over structural parameters of GNRs, such as their width and edge structure, which define GNRs’ physical properties. Chemical modification is another important tool that can be used to alter GNRs’ properties.
Two chemical modification strategies, edge functionalization and precise substitution of carbon atoms with heteroatoms, have been realized in solution-synthesized atomically precise GNRs. These approaches showed promise not only for the modification of electronic properties of GNRs, but also for improving solubility/processability and controlled self-assembly of nanoribbons.
4.1 Nitrogen Doping
For conventional semiconductors, such as silicon, substitutional doping is one of the key approaches to fine-tune electronic properties necessary for functional devices. Random substitution of carbons with heteroatoms, such as boron or nitrogen, has also been realized in graphene . In contrast, the bottom-up synthetic approaches allow to not only dope a GNR with a certain amount of heteroatoms, but also incorporate them in specific positions in the ribbon’s structure. Nitrogen-doped GNRs, which are also referred to as N-GNRs, have been considered in several theoretical studies , but only two types of solution-synthesized GNRs with precise nitrogen doping have been reported to date (Figure 11).
In 2013, Jo et al. reported modification of the GNRs 7, employing 2,3-dibromo-5,6-dialkoxypyrazine molecule 59 during the Suzuki polymerization step as a nitrogen source . In the resulting N-GNRs the level of doping could be controlled by changing the ratio of dibromopyrazine and dibromobenzene derivatives 59 and 23 during the synthesis (Figure 11(a)). It was shown that the nitrogen doping can tune the electronic transport behavior of the GNRs changing it from ambipolar for undoped GNR 7 to n-type upon nitrogen substitution. In the device measurements, N-doped GNRs with the highest level of nitrogen substitution (GNR 61) showed highest electron mobility values .
In 2014 and 2015, our group reported nitrogen doping of previously synthesized chevron-type GNRs (Figure 11(b)) . Unlike in the case of “kinked” GNRs (Figure 11(a)), here the level of doping was varied by changing the amount of nitrogen substitution in the monomer, not by changing the ratio of N-doped and undoped precursors. In principle, copolymerization of monomers 36 and 62/64 should also result in random chevron-type N-GNRs, in which the average nitrogen concentration could be tuned by the ratio of N-doped and undoped precursors, but this possibility is not yet explored experimentally for solution reactions. The solution-synthesized chevron-type N-GNRs 63 and 65 were visualized by STM and characterized by a number of spectroscopic techniques. The optical properties N-GNRs 63 and 65 were similar to those of undoped chevron-type GNRs 3 .
Because of the solubilizing alkyl chains that effectively screen edge nitrogen atoms from the environment, incorporation of nitrogens in the “kinked” GNRs 60 and 61, while changing their electronic properties, did not significantly affect interactions between the nanoribbons. However, in the case of chevron-type N-GNRs 63 and 65, where no side alkyl chains are present, the situation is very different. In addition to slight changes in electrical properties of the nanoribbons, the nitrogen substitution triggers the hierarchical self-assembly of N-GNRs into ordered structures. The phenomenon was first observed for GNR 63 [56a] and then studied in detail using surface-assisted and solution approaches to compare the behaviors of undoped GNRs 3 and N-doped GNRs 65 (Figure 12) [56a]. On a surface of Au(111) substrate, after the thermal polymerization of the monomer 64 and cyclodehydrogenation of the resulting polymer, STM analysis reveals the formation of hydrogen-bonded assemblies of N-GNRs 65, as shown in Figure 12(a–c). In addition to the same hydrogen bonding, solution-synthesized GNRs have van der Waals and π–π interactions between the basal planes of GNRs (Figure 12(d)). These interactions ultimately result in the formation of macroscopic crystalline structures of N-GNRs, which is schematically shown in Figure 12(d). These results demonstrate that substitutional doping of GNRs could be a powerful tool to promote self-assembly of nanoribbons and eventually form new carbon nanostructures [56a].
4.2 Edge Chlorination
Edge chlorination is another method of precise chemical modification of GNRs. An important advantage of this method is that it allows a post-synthetic change of the GNR structure. Edge chlorination was originally demonstrated for PAH molecules but it also works well for bottom-up solution-synthesized GNRs . A scheme of the edge chlorination of a GNR 4 is shown in Figure 13. The chlorination affects electronic properties of the GNRs decreasing the optical bandgap from 1.9 eV for the pristine GNR 4 to 1.7 eV for the chlorinated GNR 66. Additionally, the edge chlorination promotes solubility of the GNRs as the edge Cl atoms induce deformation of the structure and force the GNRs to adopt non-planar geometry.
5 Characterization Techniques for Solution-Synthesized GNRs
Materials characterization is one of the most important and still one of the most challenging aspects of the studies of atomically precise GNRs. Low solubility of GNRs in many cases precludes their complete structural characterization and a combination of different techniques is required to gain insights into the structure of the nanoribbons.
Several techniques can be utilized to determine the average length of solution-synthesized GNRs. Matrix-assisted laser desorption/ionization time of flight mass spectrometry (MALDI-TOF MS), which is usually applied to determine masses of PAHs, has limitations for high molecular weight polymers with a broad molecular weight distribution . As a result, the largest mass detected with MS does not necessarily correspond to the largest mass of a polymer. In order to estimate the molecular weight and the length of GNRs, size exclusion chromatography (SEC) of polymer precursors can be used. Typically, polymers with non-rigid core structure are analyzed against polystyrene standards, but poly(p-phenylene) standard was also suggested as a second estimate of molecular weight . Although not giving absolute molecular weights, SEC measurements provide approximate values for comparison with different polyphenylene precursors. Preparative SEC or re-precipitation may also help to remove small oligomers and improve polydispersity index . In order to determine absolute molecular weights of polymers, dynamic light scattering measurements could be employed .
When possible, liquid-phase nuclear magnetic resonance (NMR) spectroscopy provides important structural information about polymeric precursors, GNRs and their transformations. For example, a comparison of NMR spectra of polymer 29 and GNR 8 was used to estimate the degree of the polymer cyclization (Figure 14(a)) [33a]. However, GNRs are often insoluble and solid-state NMR techniques could be employed, like in case of GNR 5 [32a]. Typically, solid-state NMR spectroscopy suffers from peak broadening and overall lower quality of spectra, providing less information than liquid-state NMR (Figure 14(b)). In many cases, synthesis of representative model PAH molecules is very helpful as they are usually suitable for conclusive characterization via conventional techniques, such as liquid-phase NMR or mass spectroscopy. For example, polyaromatic compound 10a was synthesized as a model for GNR 10 (Figure 14(c)) . Since no partially fused or chlorinated products were detected after the last cyclodehydrogenation step, the same protocol was used for the synthesis of GNR 10 avoiding potential problems with the Scholl reaction, such as undesirable ring formation or chlorination.
STM becomes an increasingly important characterization method, as it provides a direct way to visualize the atomically precise structure of a GNR. STM is particularly helpful in case of insoluble GNRs, for which other options of characterization methods may be limited. Figure 15 shows the STM characterization of solution-synthesized chevron-type GNRs 3 . Notably, STM images show a clear difference between the non-planar polymer structure and the flat GNR (Figure 15(a)). Although STM has enormous potential for the GNR research, its use for the visualization of solution-synthesized atomically precise nanoribbons is rather rare. A cost of a high-resolution STM instrument is one limiting factor, but even when the access to such piece of equipment is available, STM characterization of solution-synthesized GNRs is very challenging. This is due to the fact that a typical sample preparation involves drying a droplet of a GNR dispersion in an appropriate solvent on a conductive substrate, such as Au(111). If GNRs have poor solubility, the resulting sample will mostly contain various aggregates of nanoribbons, and finding a rare individual GNR on a substrate for STM imaging could be very difficult. Also, since the sample preparation is done in air, it inevitably results in the presence of atmospheric adsorbates and other forms of contamination on a substrate in addition to the residual solvent molecules, which further complicates imaging. Sample contamination can be seen in Figure 15(b). When GNRs are functionalized with solubilizing alkyl chains, this, on the one hand, improves dispersibility of nanoribbons, but on the other hand makes GNR structures more difficult to resolve.
Despite the difficulties with high-resolution imaging of nanoribbons, STM and other microscopy techniques, such as atomic force microscopy (AFM), scanning electron microscopy (SEM) and transmission electron microscopy (TEM), are used in the studies focused on the aggregation behavior of GNRs (Figure 16). Supramolecular assembly of large PAH molecules with solubilizing groups have been reported and investigated as an important premise for organic electronic devices . Similarly, the assembly of extended PAHs – GNRs – also attracted a considerable attention. Depending on the structure of GNRs and alkyl chains attached to the aromatic core, GNRs show different self-assembly patterns on surfaces. For example, PAH 67 adopts columnar structure when deposited by drop-casting on a surface of highly oriented pyrolytic graphite (HOPG) (Figure 16(a)) [42d]. In the observed columnar arrangement the molecular planes are oriented perpendicular to the HOPG surface and parallel to each other, which facilitates electronic transport between PAH molecules 67 and makes these assemblies promising for electronic applications. In contrast, long cove-type GNRs 4 with bulky branched alkyl chains installed in the peripheral positions show highly ordered self-assembled monolayers on a HOPG surface (Figure 16(b)) [32a]. The branched alkyl chains prevent π-stacking between the co-adsorbed nanoribbons, which however was observed for the same GNRs with linear dodecyl chains.
In the case of chevron GNRs 3, which were synthesized without solubilizing alkyl chains at the edges, microscopy studies suggested formation of “nanobelts” – side-by-side arrangements of GNRs on substrates (Figure 16(c)) . The “nanobelts” are one graphene layer thick, according to AFM measurements (Figure 16(c-i)), but wider than one GNR, which is shown by SEM (Figure 16(c-ii)). STM images show side-by-side arrangement of chevron GNRs (Figure 16(c-iii)), where the protrusions of one ribbon perfectly fit into the grooves of another one. With this side-by-side assembly, the GNRs form wide and long nanostructures that can be visualized by AFM and SEM; a possible structure of these “nanobelts” of GNRs 3 is shown in Figure 16(c-v).
As discussed in Section 4.1, nitrogen substitution in GNRs triggers additional mechanism of aggregation, the hydrogen bonding. In this case, TEM images reveal flake-like structures of bulk material with interlayer distance similar to the interplanar distance in graphite, 0.34 nm (Figure 16(d-i,ii)) [56a]. The flakes consist of N-GNRs 65 that are arranged in a side-by-side fashion (see STM image in Figure 16(d-iii)), but in this case the convex elbows of adjacent GNRs project toward each other as shown in Figure 16(d-v).
Since solution methods allow synthesis of atomically precise nanoribbons in appreciable quantities, GNRs can often be characterized by various spectroscopic techniques, such as X-ray photoelectron spectroscopy (XPS), Fourier transform infrared spectroscopy (FTIR), UV-vis absorption spectroscopy and Raman spectroscopy. These methods are often helpful to confirm the efficiency of the synthetic procedure and assess the structural quality of GNRs.
XPS is widely used in the studies of graphene, graphene oxide and related materials . For example, in case of graphene oxide, i.e. oxidized graphene sheets with hydroxyl and epoxy groups at the basal planes and carboxyl and carbonyl groups at the edges [61a, 62], XPS can be used to determine elemental composition and coordination of carbons and oxygens and thus estimate the abundance of different functional groups. Similarly, XPS can be helpful for the characterization of heteroatom-substituted GNRs. As an example, Figure 17(a) shows XPS spectra for pristine chevron-type GNRs 3 and nitrogen-doped chevron N-GNRs 65. The GNR 3 exhibits only one component in the C1s spectrum, which corresponds to the sp2 carbons, and, as expected, nitrogen doping introduces additional peaks in the N1s and C1s regions of the XPS spectrum of N-GNRs 65 (Figure 17(a)).
FTIR spectroscopy is also a very informative tool to investigate the efficiency of the last cyclodehydrogenation step of the synthesis. For example, the successful transformation of polymer 41 into GNR 5 was confirmed through the comparison of their FTIR spectra (Figure 17(b)) [32a]. The strong attenuation of the signal triad from aromatic C–H stretching vibrations, the disappearance of out-of-plane C–H deformation bands, as well as the band appearing at 863 cm–1 (the band from the aromatic C–H moiety at the cove) all support the successful conversion of the polymer 41 to atomically precise GNR 5.
Raman spectroscopy is one of the most important characterization techniques in graphene research , as it could be used to study the number and orientation of layers, the edge type, disorder, doping and chemical functionalization. While the Raman spectra of graphene samples are typically dominated by D, G and 2D bands, the most intense lines around 1,300, 1,600 and 2,700 cm–1, respectively, the spectra of low-dimensional graphene materials, PAH molecules and GNRs, are usually much more complex and exhibit fine structures of smaller peaks. The positions and intensities of these peaks can often be accurately predicted using density functional theory (DFT) simulations . However, the fine structures could only be observed for high-quality GNRs, as with increasing concentration of defects, nanoribbons become qualitatively similar to other disordered carbon nanomaterials, such as carbon black, amorphous carbon and graphene oxide, which have only very broad D and G bands in their Raman spectra [61a, 65]. An example of a Raman spectrum of a solution-synthesized GNR 3 is shown in Figure 17(c); smaller bands around most intense D and G bands are shown by the arrows in the inset. Additional bands were also experimentally observed for other types of GNRs, such as cove-type GNR 4 [32a, 66]. The number and the positions of the side bands are characteristic for a particular nanoribbon structure, and the experimental observation of the fine structure in a Raman spectrum could be considered as an evidence for the high structural quality of synthesized GNRs. As shown in Figure 17(c), for GNR 3 the number and the positions of the side bands agree well with the spectrum obtained by DFT simulations. On the other hand, for a number of solution-synthesized GNRs Raman spectra exhibit only broad D and G bands; an example of such spectrum is shown in Figure 17(c) for a “kinked” GNR 8 [33a].
Spectroscopic techniques can be used to probe not only the structure of GNRs, but also their electronic properties. UV-vis absorption spectroscopy is often used to estimate optical bandgaps of solution-synthesized GNRs. However, the shape of an absorption spectrum may strongly depend on the measurement conditions, such as the solvent, concentration of GNRs and the temperature, as the aggregation of GNRs gives rise to excitonic effects . Therefore, in order to assess electronic properties of GNRs it may be useful to combine UV-vis spectroscopy with other techniques, such as scanning tunneling spectroscopy (STS) or photoluminescence (PL) spectroscopy. For example, PL spectroscopy can also be used to estimate a bandgap of nanoribbons, as shown in Figure 17(d) for GNRs 3 and 65 [56a]. Very interesting results can also be obtained for atomically precise GNRs using ultrafast THz photoconductivity measurements. Despite the progress in the device fabrication and several reports on GNR-based FETs [33a, 55, 68], study of the intrinsic electronic properties of GNRs via transport measurements is still challenging. The THz spectroscopy provides a straightforward and noninvasive approach to evaluate the photoconductivity within individual GNRs on ultrashort timescales [32, 40].
6 Challenges in Solution-Based Synthesis of GNRs
The field of atomically precise GNRs is still in its infancy and there is still a long way before it will reach the point when a bottom-up synthesis of all-carbon electronic circuits will become a reality. However, there are a number of immediate problems to solve, including limited solubility of solution-synthesized GNRs and the development of more effective chemical routes toward high-quality GNRs. On the other hand, these synthetic challenges might be considered as an opportunity for new and fascinating developments.
High lateral extension and length of nanoribbons are two very important structural parameters of solution-synthesized GNRs for practical applications. However, increasing aromatic core of GNRs results in stronger π–π interactions, which in practice leads to insolubility of a material. Reduced solubility of GNRs complicates their synthesis and nullifies one of the advantages of the solution-based methods, a liquid-phase processing of nanoribbons for fabrication of functional devices.
One of the most common ways to improve solubility and liquid-phase processability of GNRs is based on the attachment of alkyl chains to the periphery of molecules. This concept was successfully applied to solubilize large PAHs, such as hexabenzocoronene [60, 69], and was also used in many designs of solution-synthesized GNRs . For example, in cove-type GNRs 4 the installed alkyl chains on the periphery significantly enhanced dispersability of nanoribbons, which made possible further deposition of the GNRs on different substrates, materials characterization and fabrication of FETs and chemical sensors [68, 71].
Despite the efficiency of alkyl chains, it was argued that the solubilizing groups may affect packing of the molecules, reduce electrical contacts between nanoribbons in bulk and result in inferior properties of GNR-based materials . Thus, development of other approaches to improve solubility of GNRs is very important.
An interesting approach was developed to reduce the aggregation of the large PAH molecule 68, which employs tert-butyl groups introduced in the indentations of the “cove-type” edge structure (Figure 18(a)) . These groups introduce a distortion into the planar structure of the PAH, suppressing aggregation and improving solubility. The same concept was partially adopted for the synthesis of elongated homologues of PAH 68 [42d], as well as for the edge-chlorinated GNRs 65 (Figure 18(b)). With further development, synthesis of non-planar GNRs with improved solubility might be another advantage of solution bottom-up syntheses over the surface-assisted techniques.
A number of techniques to improve solubility of bottom-up synthesized GNRs could also be learned from graphene research. For example, some methods of dispersing graphenes in strong acids , appropriate organic solvents , surfactant-stabilized aqueous solutions  and other media  could be potentially adapted for processing and further investigation of solution-synthesized GNRs.
Overall, the synthesis of GNRs with improved solubility, as well as the development of controlled self-assembly strategies for GNRs to form 1D, 2D and 3D functional materials, appear to be promising areas of research in polyaromatic materials in general and GNRs in particular .
Clearly, isomerization is highly undesirable during the GNR synthesis. As it leads to a number of different final structures, isomerization essentially breaks the idea of atomically precise GNRs. Thus, the rational design of monomers and the use of appropriate chemistries to avoid the formation of isomers are very important.
We have already discussed several examples of isomerization problems. In one example, isomerization during the synthesis of the GNR 1 by Diels–Alder polymerization reaction from precursors 20 and 21 resulted in a random mixture of three regioisomeric repeating units (Figure 3(e)). Even though the following cyclodehydrogenation provided a graphitized polymer, the atomically precise synthesis was not possible with this design of monomer structures. In another example, previously discussed “kinked” GNRs 7 and 8 are also subject to isomerization. Rotation along single carbon–carbon bonds could result in rather irregular aromatic structures 69 and 70 after the cyclodehydrogenation step in addition to the proposed GNRs 7 or 8 (Figure 19(b) and (c)). Notably, the free rotation in the case of phenyl-based GNR 6 is not a problem as the cyclization is possible only for one position (Figure 19(a)). This isomerization could explain the reduced degree of cyclodehydrogenation for wider GNRs in comparison to GNR 6 [33a].
The isomerization could be avoided with proper designs of the GNR precursors. One way to solve the problem is to use substituents in order to direct the desired bond formation and block the unwanted ones . For example, alkyl chains on the periphery of the polymer precursor 39, in addition to improved solubility, also serve as blocking groups to prevent isomerization during the cyclodehydrogenation step (Figure 19(d)).
In addition to design problems, isomerization could also happen during the last cyclodehydrogenation step. Rearrangements in the Scholl reaction were reported for some polyphenylenes , making synthesis of some structures even more challenging. Related to the problem, complete cyclization and planarization of some structures are not possible, as the reaction yields a mixture of partially cyclized or chlorinated products [43a].
One of the ways to overcome the problems of the Scholl reaction could be the use of a surface-assisted cyclodehydrogenation step. Combination of solution-based and surface-assisted methods has a number of advantages, utilizing the best of both techniques. On the one hand, the solution synthesis is more versatile and flexible, and it is possible to synthesize structures that cannot be synthesized on surface. Also, some solution-based polymerization techniques are more effective than on-surface polymerizations, providing access to longer polymers. Yet, the use of a surface-assisted thermal cyclization may help to avoid some solution-based cyclization problems. After the cyclization, the resulting GNRs could be transferred to other substrates using transfer techniques well established in graphene research .
7 Summary and Future Outlook
Over the last few years there has been a great progress in the field of atomically precise GNRs. Several synthetic strategies that we have reviewed in this chapter made a considerable number of different types of GNRs available for materials studies. Because of their small size and proneness to aggregation, solution-synthesized GNRs are difficult to characterize. Yet, there has been a significant progress in the materials characterization of GNRs, which resulted in better understanding of their properties.
As the field of solution-synthesis of atomically precise GNRs will grow, future studies will likely focus on the following research topics. While we reviewed a number of examples of solution syntheses of atomically precise GNRs with various edge types (Figure 2), none of these examples showed a solution synthesis of GNRs with pure zigzag edges, which are predicted to show interesting electronic and magnetic properties [11d, 12]. Therefore, at the present, the solution synthesis of zigzag GNRs appears to be one of the most important challenges in the field. Despite the significant progress in the studies of acenes, which could be considered as the narrowest type of ZGNRs , and recent reports on the surface-assisted synthesis of certain ZGNRs , the development of a versatile solution-based approach for ZGNRs remains a formidable challenge.
In addition to ZGNRs, GNRs with non-six-membered rings, such as five- or seven-membered rings, integrated into the structure in atomically precise fashion, could also be very interesting. In recent years, a significant progress in the synthesis of PAHs with seven- and eight- membered rings has been shown , and some of these approaches may be potentially adapted for the synthesis of GNRs with distorted structures.
Doping of GNRs with heteroatoms has also shown promise for tuning electronic properties of GNRs, as well as triggering their self-assembly [56a]. As we discuss in this chapter, so far the reported examples of solution-synthesized nanoribbons are limited to nitrogen-doped GNRs. Therefore, the development of solution-based approaches for GNRs doped with other elements, such as boron, could be an interesting research direction. Further studies focused on varying concentration of heteroatoms and their positions in a GNR structure are very important for understanding the effect of substitutional doping on GNR properties.
Also of great interest are the optimization of the existing polymerization techniques and the development of new approaches to achieve longer polymers. So far, Diels–Alder polymerization has shown very promising results for the synthesis of longitudinally extended GNRs without any additional catalyst [32a]. From this prospective, other cycloaddition reactions and radical polymerization techniques could be found practical for facile and effective synthesis of longer GNRs with novel structures. Furthermore, post-synthetic treatment of GNRs and their edge-modification is a promising area of research for fine-tuning GNRs’ optoelectronic properties and self-assembly behavior.
Finally, while GNRs are expected to be relevant for a number of applications, such as organic electronics and photovoltaics, there is still a very limited number of reports where solution-synthesized GNRs are used in device studies. As the number of available GNRs grows and their structural quality improves, more examples of GNR-based devices will be reported. A significant progress in the device fabrication will be impossible without deeper understanding of the aggregation behavior of solution-synthesized GNRs, as well as the development of effective approaches for GNR self-assembly, which means that these research topics are very important for this growing field.
This article is also available in: Muellen, Feng, Chemistry of Carbon Nanostructures. De Gruyter (2016), isbn 978-3-11-028450-8.
(a) Ferrari AC, Bonaccorso F, Fal’ko V, Novoselov KS, Roche S, Boggild P, et al. Science and technology roadmap for graphene, related two-dimensional crystals, and hybrid systems. Nanoscale 2015;7(11):4598–810; (b) Geim AK, Novoselov KS, The rise of graphene. Nat Mater 2007;6(3):183–91; (c) Castro Neto AH, Guinea F, Peres NM, Novoselov KS, Geim AK. The electronic properties of graphene. Rev Mod Phys 2009;81(1):109–62. PubMedCrossrefGoogle ScholarGoogle ScholarGoogle Scholar
Kim JY, Voznyy O, Zhitomirsky D, Sargent EH. 25th anniversary article: colloidal quantum dot materials and devices: a quarter-century of advances. Adv Mater 2013;25(36):4986–5010. PubMedCrossrefGoogle Scholar
(a) Fujita M, Wakabayashi K, Nakada K, Kusakabe K, Peculiar localized state at zigzag graphite edge. J Phys Soc Jpn 1996;65(7):1920–3; (b) Ezawa M, Peculiar width dependence of the electronic properties of carbon nanoribbons. Phys Rev B 2006;73(4):045432; (c) Yazyev OV, Katsnelson MI, Magnetic correlations at graphene edges: basis for novel spintronics devices. Phys Rev Lett 2008;100(4):047209; (d) Yazyev, OV, Emergence of magnetism in graphene materials and nanostructures. Rep Prog Phys 2010;73(5):056501. CrossrefGoogle ScholarGoogle ScholarGoogle ScholarGoogle Scholar
(a) Palacios JJ, Fernández-Rossier J, Brey L, Fertig HA, Electronic and magnetic structure of graphene nanoribbons. Semicon Sci Technol 2010;25(3):033003; (b) Yazyev OV, A guide to the design of electronic properties of graphene nanoribbons. Acc Chem Res 2013;46(10):2319–28. CrossrefGoogle ScholarGoogle Scholar
(a) Areshkin DA, Gunlycke D, White CT. Ballistic transport in graphene nanostrips in the presence of disorder: importance of edge effects. Nano Lett 2006;7(1):204–10; (b) Gunlycke D, Areshkin DA, White CT, Semiconducting graphene nanostrips with edge disorder. Appl Phys Lett 2007;90(14):142104; (c) Mucciolo ER, Castro Neto AH, Lewenkopf CH, Conductance quantization and transport gaps in disordered graphene nanoribbons. Phys Rev B 2009;79(7):075407. Google ScholarGoogle ScholarGoogle Scholar
(a) Chen ZH, Lin YM, Rooks MJ, Avouris P, Graphene nano-ribbon electronics. Phys E 2007;40(2):228–32; (b) Han MY, Özyilmaz B, Zhang YB, Kim P, Energy band-gap engineering of graphene nanoribbons. Phys Rev Lett 2007;98(20):206805. CrossrefGoogle ScholarGoogle Scholar
(a) Bai JW, Duan XF, Huang Y, Rational fabrication of graphene nanoribbons using a nanowire etch mask. Nano Lett 2009;9(5):2083–7; (b) Sinitskii A, Tour JM, Patterning graphene nanoribbons using copper oxide nanowires. Appl Phys Lett 2012;100(10):103106. CrossrefPubMedGoogle ScholarGoogle Scholar
(a) Ci LJ, Xu ZP, Wang LL, Gao W, Ding F, Kelly KF, et al. Controlled nanocutting of graphene. Nano Res 2008;1(2):116–22; (b) Campos LC, Manfrinato VR, Sanchez-Yamagishi JD, Kong J, Jarillo-Herrero P, Anisotropic etching and nanoribbon formation in single-layer graphene. Nano Lett 2009;9(7): 2600–4. CrossrefGoogle ScholarGoogle Scholar
Mohanty N, Moore D, Xu ZP, Sreeprasad TS, Nagaraja A, Rodriguez AA, et al. Nanotomy-based production of transferable and dispersible graphene nanostructures of controlled shape and size. Nat. Commun 2012;3:844. CrossrefPubMedGoogle Scholar
(a) Wang X, Dai H, Etching and narrowing of graphene from the edges. Nat Chem 2010;2(8):661–5; (b) Yang R, Shi Z, Zhang L, Shi D, Zhang G, Observation of Raman g-peak split for graphene nanoribbons with hydrogen-terminated zigzag edges. Nano Lett 2011;11(10):4083–8. PubMedCrossrefGoogle ScholarGoogle Scholar
(a) Kosynkin DV, Higginbotham AL, Sinitskii A, Lomeda JR, Dimiev A, Price BK, et al. Longitudinal unzipping of carbon nanotubes to form graphene nanoribbons. Nature 2009;458(7240):872–6; (b) Jiao LY, Zhang L, Wang XR, Diankov G, Dai HJ, Narrow graphene nanoribbons from carbon nanotubes. Nature 2009;458(7240):877–80; (c) Cano-Marquez AG, Rodriguez-Macias FJ, Campos-Delgado J, Espinosa-Gonzalez CG, Tristan-Lopez F, Ramirez-Gonzalez D, et al. Ex-MWNTs: graphene sheets and ribbons produced by lithium intercalation and exfoliation of carbon nanotubes. Nano Lett 2009;9(4):1527–33; (d) Sinitskii A, Fursina AA, Kosynkin DV, Higginbotham AL, Natelson D, Tour JM, Electronic transport in monolayer graphene nanoribbons produced by chemical unzipping of carbon nanotubes. Appl Phys Lett 2009;95(25):253108; (e) Jiao LY, Wang XR, Diankov G, Wang HL, Dai HJ, Facile synthesis of high-quality graphene nanoribbons. Nat Nanotechnol 2010;5(5):321–5; (f) Kosynkin DV, Lu W, Sinitskii A, Pera G, Sun ZZ, Tour JM. Highly conductive graphene nanoribbons by longitudinal splitting of carbon nanotubes using potassium vapor. ACS Nano 2011;5(2):968–74; (g) Sinitskii A, Dimiev A, Kosynkin DV, Tour JM, Graphene nanoribbon devices produced by oxidative unzipping of carbon nanotubes. ACS Nano 2010;4(9):5405–13. PubMedCrossrefGoogle ScholarGoogle ScholarGoogle ScholarGoogle ScholarGoogle ScholarGoogle ScholarGoogle Scholar
(a) Han MY, Brant JC, Kim P. Electron transport in disordered graphene nanoribbons. Phys Rev Lett 2010;104(5):056801; (b) Oostinga JB, Sacepe B, Craciun MF, Morpurgo AF. Magnetotransport through graphene nanoribbons. Phys Rev B 2010;81(19):193408. PubMedCrossrefGoogle ScholarGoogle Scholar
(a) Shifrina ZB, Averina MS, Rusanov AL, Wagner M, Müllen K. Branched polyphenylenes by repetitive Diels–Alder cycloaddition. Macromolecules 2000;33(10):3525–9; (b) Wu JS, Gherghel L, Watson MD, Li JX, Wang ZH, Simpson CD. et al. From branched polyphenylenes to graphite ribbons. Macromolecules 2003;36(19):7082–7089. CrossrefGoogle ScholarGoogle Scholar
(a) Obradovic B, Kotlyar R, Heinz F, Matagne P, Rakshit T, Giles MD, et al. Analysis of graphene nanoribbons as a channel material for field-effect transistors. Appl Phys Lett 2006;88(14):142102; (b) Wang J, Zhao R, Yang M, Liu Z, Liu Z. Inverse relationship between carrier mobility and bandgap in graphene. J Chem Phys 2013;138(8):084701. CrossrefGoogle ScholarGoogle Scholar
(a) El Gemayel M, Narita A, Dössel LF, Sundaram RS, Kiersnowski A, Pisula W, et al. Graphene nanoribbon blends with P3HT for organic electronics. Nanoscale 2014;6(12):6301–14; (b) Yang X, Dou X, Rouhanipour A, Zhi L, Räder HJ, Müllen K. Two-dimensional graphene nanoribbons. J Am Chem Soc 2008;130(13):4216–7. CrossrefPubMedGoogle ScholarGoogle Scholar
(a) Narita A, Feng X, Hernandez Y, Jensen SA, Bonn M, Yang H. et al. Synthesis of structurally well-defined and liquid-phase-processable graphene nanoribbons. Nat Chem 2014;6(2):126–32; (b) Narita A, Verzhbitskiy IA, Frederickx W, Mali KS, Jensen SA, Hansen MR, Bonn M, De Feyter S, Casiraghi C, Feng X, Müllen K. Bottom-up synthesis of liquid-phase-processable graphene nanoribbons with near-infrared absorption. ACS Nano 2014;8(11):11622–30. CrossrefPubMedGoogle ScholarGoogle Scholar
(a) Dössel L, Gherghel L, Feng X, Müllen K. Graphene nanoribbons by chemists: nanometer-sized, soluble, and defect-free. Angew Chem Int Ed Engl 2011;50(11):2540–3; (b) Kim KT, Jung JW, Jo WH. Synthesis of graphene nanoribbons with various widths and its application to thin-film transistor. Carbon 2013;63:202–209. CrossrefPubMedGoogle ScholarGoogle Scholar
Schwab MG, Narita A, Hernandez Y, Balandina T, Mali KS, De Feyter S, et al. Structurally defined graphene nanoribbons with high lateral extension. J Am Chem Soc 2012;134(44):18169–72. CrossrefPubMedGoogle Scholar
Abbas AN, Liu B, Narita A, Dössel LF, Yang B, Zhang W, et al. Vapor-phase transport deposition, characterization, and applications of large nanographenes. J Am Chem Soc 2015;137(13):4453–9. CrossrefPubMedGoogle Scholar
Jensen SA, Ulbricht R, Narita A, Feng X, Müllen K, Hertel T, et al. Ultrafast photoconductivity of graphene nanoribbons and carbon nanotubes. Nano Lett 2013;13(12):5925–30. PubMedCrossrefGoogle Scholar
Dibble DJ, Park YS, Mazaheripour A, Umerani MJ, Ziller JW, Gorodetsky AA. Synthesis of polybenzoquinolines as precursors for nitrogen-doped graphene nanoribbons. Angew Chem Int Edit 2015;54(20):5883–7. CrossrefGoogle Scholar
(a) Bohme T, Simpson CD, Müllen K, Rabe JP. Current-voltage characteristics of a homologous series of polycyclic aromatic hydrocarbons. Chemistry 2007;13(26):7349–57; (b) Ito S, Wehmeier M, Brand JD, Kubel C, Epsch R, Rabe JP et al. Synthesis and self-assembly of functionalized hexa-peri-hexabenzocoronenes. Chemistry 2000;6(23):4327–42; (c) Wu J, Pisula W, Müllen K. Graphenes as potential material for electronics. Chem Rev 2007;107(3):718–47; (d) Fogel Y, Zhi LJ, Rouhanipour A, Andrienko D, Räder HJ, Müllen K. Graphitic nanoribbons with dibenzo[e,l]pyrene repeat units: synthesis and self-assembly. Macromolecules 2009;42(18):6878–84. CrossrefPubMedGoogle ScholarGoogle ScholarGoogle ScholarGoogle Scholar
(a) Liu J, Li BW, Tan YZ, Giannakopoulos A, Sanchez-Sanchez C, Beljonne D, et al. Toward cove-edged low bandgap graphene nanoribbons. J Am Chem Soc 2015;137(18):6097–103; (b) Grzybowski M, Skonieczny K, Butenschön H, Gryko DT. Comparison of oxidative aromatic coupling and the Scholl reaction. Angew Chem Int Ed Engl 2013;52(38):9900–30; (c) Waldvogel SR, Trosien S. Oxidative transformation of aryls using molybdenum pentachloride. Chem Commun (Camb) 2012;48(73):9109–19. CrossrefGoogle ScholarGoogle ScholarGoogle Scholar
Morgenroth F, Kübel C, Müller M, Wiesler UM, Berresheim AJ, Wagner M, et al. From three-dimensional polyphenylene dendrimers to large graphite subunits. Carbon 1998;36(5–6):833–7. CrossrefGoogle Scholar
Takada T, Arisawa M, Gyoten M, Hamada R, Tohma H, Kita Y. Oxidative biaryl coupling reaction of phenol ether derivatives using a hypervalent iodine(III) reagent. J Org Chem 1998;63(22):7698–706. CrossrefGoogle Scholar
Goldfinger MB, Swager TM. Fused polycyclic aromatics via electrophile-induced cyclization reactions – application to the synthesis of graphite ribbons. J Am Chem Soc 1994;116(17):7895–6. CrossrefGoogle Scholar
(a) Byers PM, Alabugin IV. Polyaromatic ribbons from oligo-alkynes via selective radical cascade: stitching aromatic rings with polyacetylene bridges. J Am Chem Soc 2012;134(23):9609–14; (b) Byers PM, Rashid JI, Mohamed RK, Alabugin IV. Polyaromatic ribbon/benzofuran fusion via consecutive endo cyclizations of enediynes. Org Lett 2012;14(23):6032–5; (c) Mohamed RK, Mondal S, Gold B, Evoniuk CJ, Banerjee T, Hanson K. Alkenes as alkyne equivalents in radical cascades terminated by fragmentations: overcoming stereoelectronic restrictions on ring expansions for the preparation of expanded polyaromatics. J Am Chem Soc 2015;137(19):6335–49. CrossrefPubMedGoogle ScholarGoogle ScholarGoogle Scholar
(a) Jin Z, Yao J, Kittrell C, Tour JM. Large-scale growth and characterizations of nitrogen-doped monolayer graphene sheets. ACS Nano 2011;5(5):4112–17; (b) Lv R, Li Q, Botello-Mendez AR, Hayashi T, Wang B, Berkdemir A. Nitrogen-doped graphene: beyond single substitution and enhanced molecular sensing. Sci Rep 2012;2:586; (c) Wang HB, Maiyalagan T, Wang X. Review on recent progress in nitrogen-doped graphene: synthesis, characterization, and its potential applications. ACS Catal 2012;2(5):781–94. CrossrefPubMedGoogle ScholarGoogle ScholarGoogle Scholar
(a) Yu SS, Zheng WT, Wen QB, Jiang Q. First principle calculations of the electronic properties of nitrogen-doped carbon nanoribbons with zigzag edges. Carbon 2008;46(3):537–43; (b) Li Y, Zhou Z, Shen P, Chen Z. Spin gapless semiconductor−metal−half-metal properties in nitrogen-doped zigzag graphene nanoribbons. ACS Nano 2009;3(7):1952–8; (c) Wei X-L, Fang H, Wang R-Z, Chen Y-P, Zhong J-X. Energy gaps in nitrogen delta-doping graphene: A first-principle study. Appl Phys Lett 2011;99(1):012107; (d) Kim H, Lee K, Woo SI, Jung Y. On the mechanism of enhanced oxygen reduction reaction in nitrogen-doped graphene nanoribbons. Phys Chem Chem Phys 2011;13(39):17505–10; (e) Cruz-Silva E, Barnett ZM, Sumpter BG, Meunier V. Structural, magnetic, and transport properties of substitutionally doped graphene nanoribbons from first principles. Phys Rev B 2011;83(15):155445. CrossrefGoogle ScholarGoogle ScholarGoogle ScholarGoogle ScholarGoogle Scholar
(a) Vo TH, Shekhirev M, Kunkel DA, Orange F, Guinel MJ, Enders A, Sinitskii A. Bottom-up solution synthesis of narrow nitrogen-doped graphene nanoribbons. Chem Commun (Camb) 2014;50(32):4172–4; (b) Vo TH, Perera UG, Shekhirev M, Mehdi Pour M, Kunkel, DA, Lu H, et al. Nitrogen-doping induced self-assembly of graphene nanoribbon-based two-dimensional and three-dimensional metamaterials. Nano Lett 2015;15(9):5770–7. CrossrefPubMedGoogle ScholarGoogle Scholar
Tan YZ, Yang B, Parvez K, Narita A, Osella S, Beljonne D, et al. Atomically precise edge chlorination of nanographenes and its application in graphene nanoribbons. Nat Commun 2013;4:2646. PubMedCrossrefGoogle Scholar
Martin K, Spickermann J, Räder HJ, Müllen K. Why does matrix-assisted laser desorption/ionization time-of-flight mass spectrometry give incorrect results for broad polymer distributions? Rapid Commun Mass Spect 1996;10(12):1471–4. CrossrefGoogle Scholar
Kroeger A, Deimede V, Belack J, Lieberwirth I, Fytas G, Wegner G. Equilibrium length and shape of rod-like polyelectrolyte micelles in dilute aqueous solutions. Macromolecules 2007;40(1):105–15. CrossrefGoogle Scholar
(a) Stankovich S, Dikin DA, Piner RD, Kohlhaas KA, Kleinhammes A, Jia Y, et al. Synthesis of graphene-based nanosheets via chemical reduction of exfoliated graphite oxide. Carbon 2007;45(7):1558–65; (b) Sinitskii A, Tour JM. Chemical approaches to produce graphene oxide and related materials. In: Murali R, editors. Graphene Nanoelectronics. Springer US: 2012:205–234. CrossrefGoogle ScholarGoogle Scholar
(a) Negri F, Castiglioni C, Tommasini M, Zerbi G. A computational study of the Raman spectra of large polycyclic aromatic hydrocarbons: toward molecularly defined subunits of graphite. J Phys Chem A 2002;106(14):3306–17; (b) Castiglioni C, Mapelli C, Negri F, Zerbi G. Origin of the D line in the Raman spectrum of graphite: a study based on Raman frequencies and intensities of polycyclic aromatic hydrocarbon molecules. J Chem Phys 2001;114(2):963–74. CrossrefGoogle ScholarGoogle Scholar
(a) Ferrari AC, Robertson J, Interpretation of Raman spectra of disordered and amorphous carbon. Phys Rev B 2000;61(20):14095–107; (b) Wilson PM, Orange F, Guinel MJ, Shekhirev M, Gao Y, Colon Santana JA, et al. Oxidative peeling of carbon black nanoparticles. RSC Adv 2015;5(112):92539–44. CrossrefGoogle ScholarGoogle Scholar
Hughes JM, Hernandez Y, Aherne D, Dössel L, Müllen K, Moreton B, et al. High quality dispersions of hexabenzocoronene in organic solvents. J Am Chem Soc 2012;134(29):12168–79. PubMedCrossrefGoogle Scholar
(a) Abbas AN, Liu G, Narita A, Orosco M, Feng X, Müllen K, et al. Deposition, characterization, and thin-film-based chemical sensing of ultra-long chemically synthesized graphene nanoribbons. J Am Chem Soc 2014;136(21):7555–8; (b) Zschieschang U, Klauk H, Mueller IB, Strudwick AJ, Hintermann T, Schwab MG, Narita A, Feng XL, Müllen K, Weitz RT, Electrical characteristics of field-effect transistors based on chemically synthesized graphene nanoribbons. Adv Electron Mater 2015;1(3):1400010. CrossrefPubMedGoogle ScholarGoogle Scholar
Kastler M, Pisula W, Wasserfallen D, Pakula T, Müllen K. Influence of alkyl substituents on the solution- and surface-organization of hexa-peri-hexabenzocoronenes. J Am Chem Soc 2005;127(12):4286–96. PubMedCrossrefGoogle Scholar
Konnerth R, Cervetti C, Narita A, Feng X, Müllen K, Hoyer A, et al. Tuning the deposition of molecular graphene nanoribbons by surface functionalization. Nanoscale 2015;7(30):12807–11. PubMedCrossrefGoogle Scholar
Behabtu N, Lomeda JR, Green MJ, Higginbotham AL, Sinitskii A, Kosynkin DV, et al. Spontaneous high-concentration dispersions and liquid crystals of graphene. Nat Nanotechnol 2010;5(6):406–11. CrossrefPubMedGoogle Scholar
(a) Khan U, O‘Neill A, Lotya M, De S, Coleman JN, High-concentration solvent exfoliation of graphene. Small 2010;6(7):864–71; (b) Hamilton CE, Lomeda JR, Sun Z, Tour JM, Barron AR, High-yield organic dispersions of unfunctionalized graphene. Nano Lett 2009;9(10):3460–2. PubMedCrossrefGoogle ScholarGoogle Scholar
(a) Lotya M, Hernandez Y, King PJ, Smith RJ, Nicolosi V, Karlsson LS, et al. Liquid phase production of graphene by exfoliation of graphite in surfactant/water solutions. J Am Chem Soc 2009;131(10):3611–20; (b) Lotya M, King PJ, Khan U, De S, Coleman JN, High-concentration, surfactant-stabilized graphene dispersions. ACS Nano 2010;4(6):3155–62. CrossrefPubMedGoogle ScholarGoogle Scholar
(a) Tönshoff C, Bettinger HF, Beyond pentacenes: synthesis and properties of higher acenes. In: Siegel SJ, Wu Y-T, editors. Polyarenes I. Springer, Berlin, Heidelberg: Berlin, Heidelberg, 2014: 1–30; (b) Anthony JE, The larger acenes: versatile organic semiconductors. Angew Chem Int Ed Engl 2008;47(3):452–83. Google ScholarGoogle Scholar