# Fileset

[20230413 Angew Chem Int Ed - 2023 - Xu - Magnetism in Nonplanar Zigzag Edge Termini of Graphene Nanoribbons.pdf](https://mdr.nims.go.jp/filesets/54bb0891-4efe-41b8-8fe2-98e53edc3820/download)

## Creator

[Xiushang Xu](https://orcid.org/0000-0002-3225-0392), [Kewei Sun](https://orcid.org/0000-0002-1835-243X), [Atsushi Ishikawa](https://orcid.org/0000-0001-6908-831X), [Akimitsu Narita](https://orcid.org/0000-0002-3625-522X), [Shigeki Kawai](https://orcid.org/0000-0003-2128-0120)

## Rights

[Creative Commons BY Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/)

## Other metadata

[Magnetism in Nonplanar Zigzag Edge Termini of Graphene Nanoribbons](https://mdr.nims.go.jp/datasets/e0315e7b-79a5-418f-9195-7a46628e4158)

## Fulltext

Magnetism in Nonplanar Zigzag Edge Termini of Graphene NanoribbonsGraphene Nanoribbons Hot PaperMagnetism in Nonplanar Zigzag Edge Termini of GrapheneNanoribbonsXiushang Xu+, Kewei Sun+, Atsushi Ishikawa,* Akimitsu Narita,* and Shigeki Kawai*Abstract: Graphene nanoribbons (GNRs) and nanogra-phenes synthesized by on-surface reactions using tailor-made molecular precursors offer an ideal playground for astudy of magnetism towards nano-spintronics. Althoughthe zigzag edge of GNRs has been known to hostmagnetism, the underlying metal substrates usually veil theedge-induced Kondo effect. Here, we report the on-surfacesynthesis of unprecedented, π-extended 7-armchair GNRsusing 7-bromo-12-(10-bromoanthracen-9-yl)tetraphene asthe precursor. Characterization by scanning tunnelingmicroscopy/spectroscopy revealed unique rearrangementreactions leading to pentagon- or pentagon/heptagon-incorporated, nonplanar zigzag termini, which demon-strated Kondo resonances even on bare Au(111). Densityfunctional theory calculations indicate that the nonplanarstructure significantly reduces the interaction between thezigzag terminus and the Au(111) surface, leading to arecovery of the spin localization of the zigzag edge. Such adistortion of planar GNR structures offers a degree offreedom to control the magnetism on metal substrates.IntroductionQuasi-one-dimensional graphene nanoribbons (GNRs) andzero-dimensional nanographenes (NGs) have attracted tre-mendous attention from researchers in the past decade sincethey are expected to offer key elements in forthcomingnano-electronics due to their tunable electronic propertiesthat are dependent on the sizes and edge structures as wellas substituted heteroatoms.[1] To date, various GNRs[2] andNGs[3] with atomically defined structures have been synthe-sized on surfaces and directly visualized by high-resolutionscanning probe microscopy.[4] In the synthesis processes,designer precursor molecules are deposited on a surface andsubsequently planarized by annealing, namely by surface-assisted cyclodehydrogenation. Direct planarization of pre-cursor molecules leads to the formation of NGs, whileGNRs can be obtained by functionalizing them withhalogens to enable polymerization before planarization.Besides the electronic band gap engineering, the magneticproperties can be induced in GNRs and NGs[1e,5] as the spinpolarization arises from the unpaired electrons[6] and the zigzagterminal structures.[7] From the early studies of the N=7armchair GNR (7-AGNR), it has been known that the zigzagterminus hosts a net spin.[2b,8] However, such magneticproperty is usually suppressed by the strong interactionbetween the zigzag terminus and the metal substrate.[9] When7-AGNR is successfully transferred onto an insulating surface,such as a NaCl bilayer film on Au(111) by the tip-inducedmanipulation, the spin splitting at the zigzag terminus can bedetected by scanning tunneling spectroscopy (STS) at lowtemperature.[2c,10] This spin splitting of the zigzag terminus wasinterpreted as Coulomb gap arising from the electron-electroninteraction.[6a,d] In contrast, Kondo effect relates to themagnetic moment screened by conduction electrons of a metalsubstrate. So far, Kondo resonances have been investigated ina few NGs and their dimers, demonstrating significantstructure-dependence of their magnetic properties.[1d,6c,e,7a,11]Substituted boron atoms in 7-GNRs induce the apparentzigzag termini, leading to Kondo resonances if the interactionto the substrate is reduced by the tip lifting[6c] and the AuSixdecoupling layer.[12] However, the detection of Kondo peakfrom GNRs on bare metal substrates still remains challenging,[*] X. Xu,+ A. NaritaOrganic and Carbon Nanomaterials Unit,Okinawa Institute of Science and Technology Graduate University1919-1 Tancha, Onna-son, Kunigami-gun,Okinawa, 904-0495 (Japan)E-mail: akimitsu.narita@oist.jpK. Sun,+ S. KawaiResearch Center for Advanced Measurement and Characterization,National Institute for Materials Science (NIMS)1-2-1 Sengen, Tsukuba, Ibaraki, 305-0044 (Japan)E-mail: KAWAI.Shigeki@nims.go.jpA. IshikawaCenter for Green Research on Energy and Environmental Materials(GREEN), National Institute for Materials Science (NIMS)Namiki 1-1, Tsukuba, Ibaraki, 305-0044 (Japan)andPresent address: Department of Transdisciplinary Science andEngineering, School of Environment and Society, Tokyo Institute ofTechnology,2-12-1 Ookayama, Meguro-ku, Tokyo, 152-8552 (Japan)E-mail: ishikawa.a.ai@m.titech.ac.jpS. KawaiGraduate School of Pure and Applied Sciences,University of TsukubaTsukuba, 305-8571 (Japan)[+] These authors contributed equally to this work.© 2023 The Authors. Angewandte Chemie International Editionpublished by Wiley-VCH GmbH. This is an open access article underthe terms of the Creative Commons Attribution License, whichpermits use, distribution and reproduction in any medium, providedthe original work is properly cited.AngewandteChemieResearch Articleswww.angewandte.orgHow to cite: Angew. Chem. Int. Ed. 2023, 62, e202302534doi.org/10.1002/anie.202302534Angew. Chem. Int. Ed. 2023, 62, e202302534 (1 of 6) © 2023 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbHhttp://orcid.org/0000-0002-3225-0392http://orcid.org/0000-0002-1835-243Xhttp://orcid.org/0000-0001-6908-831Xhttp://orcid.org/0000-0002-3625-522Xhttp://orcid.org/0000-0003-2128-0120https://doi.org/10.1002/anie.202302534http://crossmark.crossref.org/dialog/?doi=10.1002%2Fanie.202302534&domain=pdf&date_stamp=2023-04-13and thus far only achieved with chiral (3,1)-GNR upon fusionto form junctions with rearranged structures, hosting localizedspins,[7d] or after a rather random introduction of carbonylgroups on zigzag edges, e.g. through exposure to oxygen gas.[13]Herein, we report the on-surface synthesis of π-extended7-AGNRs with nonplanar zigzag termini, which exhibits theKondo effect on Au(111), using 7-bromo-12-(10-bromo-anthracen-9-yl)tetraphene (DBATP) as the precursor. Afterthe Ullmann-type coupling reaction on Au(111), a three-dimensional (3D) helical segment was observed as anintermediate by controlling the annealing temperature.Subsequently, the cyclodehydrogenation reaction yieldedbenzo- and indeno-fused 7-AGNR (BI-7-AGNR), involvingskeletal rearrangement. In the planarization process, thecleavage and formation of the C� C bond at the armchairedge were systematically induced. Remarkably, we found astrong zero-bias peak at the terminus with a nonplanarstructure in the vicinity of the zigzag edge. Moreover,density functional theory (DFT) calculations revealed thatstructural nonplanarity reduces the interaction between thezigzag edge and the meal substrate, leading to a recovery ofthe spin polarization.Results and DiscussionOur original motivation was the fabrication of helical 3D-GNR[14] on a metal surface, and DBATP 1 was designed asthe precursor based on 10,10’-dibromo-9,9’-bianthryl(DBBA), which has been commonly used for the synthesisof 7-AGNRs.[2b] DBATP 1 features an additional benzenering fused to DBBA, and was expected to provide 3D-GNRs with helical segments by Ullmann-type coupling andcyclodehydrogenation on surface (Scheme 1). For the syn-thesis of DBATP 1, 2-(naphthalen-2-ylmethyl)benzaldehyde(4) was initially prepared by palladium-catalyzed C(sp3)-Harylation of 2-methylbenzaldehyde (2) with 2-iodonaphtha-lene (3) in 25% yield (Scheme 1).[15] Subsequently, 4 wasreacted with anthracen-9-yllithium that was generated bylithiation of 9-bromoanthracene, to give a secondary alcoholintermediate, followed by cyclization with BF3·OEt2 andthen oxidation with 2,3-dichloro-5,6-dicyano-1,4-benzoqui-none (DDQ) to provide 12-(anthrancen-9-yl)tetraphene (5)in 48% yield. Finally, DBATP 1 was obtained by bromina-tion of 5 with N-bromosuccinimide (NBS) in 64% yield andcharacterized by 1H and 13C nuclear magnetic resonance(NMR) spectroscopy and high-resolution mass spectrometry(see Supporting Information, Figures S1–S6).Toward the on-surface synthesis of 3D-GNRs, DBATP 1was deposited on Au(111) kept at room temperature, andsubsequently annealed at 180 °C to induce the debromina-tion and the C� C coupling. Corrugated oligomers wereobserved on the surface as indicated by arrows in Figure 1a,which appeared almost identical to the intermediate productin the synthesis of 7-AGNR.[2b] The successful covalent bondformation was confirmed by the tip-induced manipulation(Figure S7), which displayed the bending of the linearoligomer 6. Higher temperature annealing at 300 °C led tothe formation of planar ribbons, which has a small numberof corrugated moieties with a bright feature as indicated byan arrow in Figure 1b. Since the contrast of the bright dotdiffers from that of 7-AGNR synthesized by imperfectcyclodehydrogenation,[16] it can be readily assigned to anextra benzene ring originally existing in the structure ofDBATP (Figure S8). Thus, the 3D-helical segment 7 wassuccessfully incorporated in the edges of GNRs, which hasnever been reported before to the best of our knowledge.However, most of the parts in the ribbon became flat,suggesting the occurrence of strain-induced skeletalrearrangement.[17] After heating the sample to 370 °C,completely planar GNRs without bright dots appeared onsurface as indicated by arrows in Figure 1c. Figure 1d showsthe close view of the GNR, in which both sides in thelongitudinal axis have irregular nodes as marked by arrows.To investigate the structures, the tip apex was terminatedwith a CO molecule[18] and was scanned at a constant heightover the GNR while recoding the differential conductancedI/dV (Figure 1e). The contrast at the terminus is muchbrighter than those of other parts of the GNR, which hasalready indicated the presence of the spin-polarized state atthe terminus. We first focus on the edges of the longitudinalaxis as indicated by a rectangle in Figure 1e. The innerstructure of the GNR segment was clearly observed in thebond-resolved dI/dV map (Figure 1f), in which the extrabenzene rings from the DBATP precursor were not visible.We instead found a systematic rearrangement of carbonatoms, consequently forming two types of structures; (i) onehexagonal ring (Type 1) and (ii) a set of hexagonal andpentagonal rings (Type 2) fused to the armchair edges(Figure S26), leading to benzo- and indeno-fused 7-AGNR(BI-7-AGNR). The chemical structures of Type 1 andType 2 have been defined in Scheme 1. For a betterobservation, the close-up bond-resolved image of Type 1and Type 2 was recorded as shown in Figure S9. AdditionalScheme 1. The synthetic route to DBATP 1 in solution and on-surfacesynthesis of BI-7-AGNR.AngewandteChemieResearch ArticlesAngew. Chem. Int. Ed. 2023, 62, e202302534 (2 of 6) © 2023 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH 15213773, 2023, 24, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.202302534 by Cochrane Japan, Wiley Online Library on [24/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licenseimages of other GNR segments are displayed in Figure S10.Type 1 has two carbon atoms less compared with theprecursor, which might be removed as ethylene or ethane.We also found that this ethenylene can occasionally fuse tothe adjacent armchair edge of GNR (Figure S10). Weassume that the reaction mechanism leading to Types 1 and2 might be similar to the reaction pathway that waspreviously proposed for a related system,[17] but considerthat in-depth theoretical studies would be necessary forfurther discussion.[19,20] Another edge site (indicated by lightblue arrow in Figure 1f) appeared different from Type 1 andType 2, and we propose the corresponding chemical struc-ture as shown in Figure S11.Given the fact that the BI-7-AGNR lack periodic edges,the electronic structures measured by STS were rathercomplicated. A series of the dI/dV curves were taken alongthe longitudinal axes of GNR and showed a strongmodification of electronic structures by the nodes. Suchelectronic structures were also seen in dI/dV maps taken atdifferent bias voltages (Figure S12).In the vicinity of the zigzag termini of the GNRs, weobserved three different structures. Figure 2a shows a close-up view of the first type of the terminus, in which a smallnode is indicated by an arrow. The constant height dI/dVmap (Figure 2b) and the corresponding Laplace filteredimage (Figure 2c) show the distinct inner structure, in whichthe formation of the additional hexagonal ring can beobserved (Figure 2d). Hereafter, we name this terminus asTerminal 1. Figure 2e shows the second type of the terminuswith a larger node in the STM topography as indicated byan arrow. The corresponding chemical structure was identi-fied from the constant height dI/dV maps (Figure 2f, Fig-ure S13) and the Laplace filtered image (Figure 2g), namedas Terminal 2 (Figure 2h). The large node in the STMtopography is related to the apparent feature of thepentagonal ring and the hexagonal ring. Unlike the bare 7-AGNR directly adsorbed on Au(111), strong contrasts weredetected at the zigzag edge. The structure of Terminal 2appeared similar to that of Type 2, but the positions of thepentagonal and hexagonal rings were switched each other(see also Figure S26). Consequently, the additional hexago-nal ring at the zigzag edge is slightly pulled towards theGNR main body by the small pentagonal ring. Figure 2ishows the last terminus, having two small nodes as markedby arrows, named as Terminal 3. We found the strong signalin the constant height dI/dV map around the nodes,indicating the corrugated structure (Figure 2j, Figure S14).According to the corresponding Laplace filtered image(Figure 2k), in which one heptagonal ring and two adjacentpentagonal rings are visible, we assign Terminal 3 to thechemical structure as shown in Figure 2l. We also observed asimilar structure within a GNR segment (called Type 3 inFigure S10d), supporting this structural assignment. Interest-ingly, for Terminals 2 and 3, the constant height dI/dV mapstaken at close to the Fermi level (Figure 2f and g) showsignificant modulations of the differential conductance. Thisphenomenon suggests the presence of spin-polarized states,Figure 1. On-surface synthesis of BI-7-AGNR on Au(111). a) STM topography of DBATP on Au(111) after annealing to 180 °C for 10 min. b) STMtopography of the surface after heating to 300 °C for 10 min. The inset shows the structural model of a 3D-GNR segment. c) STM overview of thesurface after heating to 370 °C for 10 min. d) Close-up view of BI-7-AGNR on Au(111) and e) the corresponding constant height dI/dV map.f) Constant height dI/dV map of the area indicated by a rectangle in (e). Measurement parameters: Sample bias voltage V=200 mV and tunnelingcurrent I=5 pA in (a,b, c). V=200 mV and I=3 pA in (d). For constant height dI/dV map: V=1 mV, Vac=10 mV in (e, f).AngewandteChemieResearch ArticlesAngew. Chem. Int. Ed. 2023, 62, e202302534 (3 of 6) © 2023 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH 15213773, 2023, 24, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.202302534 by Cochrane Japan, Wiley Online Library on [24/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensewhich is usually quenched by the interaction between thezigzag terminus of GNR and the underlying Au(111).To investigate the origin of the bright contrast at themodified termini of the GNRs on Au(111), the electronicstructures with large energy ranges (Figure S15) and near theFermi level (Figure 3) were measured with STS. Figure 3ashows a bond-resolved image of Terminal 1, superimposedwith the structural model and dots, which indicates the tippositions for the STS measurements. We observed the absenceof the zero-bias peak in the dI/dV curves (Figure 3b), meaningthat no net spin exists in Terminal 1. This result is consistentwith that of non-decorated 7-AGNR edge. In contrast, distinctzero-bias peaks appeared at the zigzag edge of Terminal 2 asindicated by the red dot and curve in Figures 3c and d,respectively. Furthermore, similar peaks were also measurednear the zigzag edge as indicated by green, blue, and pinkcurves. We assigned the zero-bias state to Kondo resonance,arising from the magnetic moment screened by conductionelectrons of the Au(111) surface, since such a peak was absentat the site distant from the zigzag edge (light blue and yellow).We found that Terminal 3 also shows a zero-bias peak in thedI/dV curves (Figures 3e and f). In contrast, the dI/dV curveabove the pentagonal ring fused at armchair edge far awayfrom the zigzag terminus has no zero-bias peak (Figure S16),indicating pentagonal ring itself could not hold net spin. Giventhat the zigzag terminus of a GNR can increase its interactionwith the underlying substrate due to formation of a singleC� Au bond after removing a hydrogen (H) atom,[21] thisbehavior would quench the spin polarization. Thus, we utilizedthe tip manipulation to remove one H atom at the zigzagtermini of Terminal 2 and 3 and found the disappearance ofthe zero-bias peak (Figures S17 and S18). These results furthersuggest that the GNR spin density arises from the zigzag edges.Half width at half maximum (HWHM) of the curve at thezigzag edge of Terminal 3 obtained by Frota fitting (indicatedby dashed lines in Figure 3d and f) is 12.7�0.8 mV, while thatof Terminal 2 is 2.7�0.3 mV. The larger linewidth of Termi-nus 3 might be because of its relatively strong interaction withthe underlying substrate. Nevertheless, the net spin is nolonger suppressed by the Au(111) substrate regardless ofwhether the end of GNR is Terminal 2 or Terminal 3. Thesetypes of termini were observed also in some other GNRs, andthe corresponding STS measurements displayed the consistentresults that Terminal 2 and 3 have zero-bias peak (Figur-es S19–S21). Our study demonstrates the importance of edgeFigure 2. Three types of termini structures of BI-7-AGNR on Au(111).a) STM topography of Terminal 1 with a small node (indicated by anarrow) at the top edge of BI-7-AGNR. b) Constant height dI/dV map ofone Terminal 1. c) Corresponding Laplace filtered image. d) Corre-esponding chemical structure. e) STM topography Terminal 2 with alarge node (indicated by an arrow) at the top edge of BI-7-AGNR.f) Constant height dI/dV map of Terminal 2. g) Corresponding Laplacefiltered image. h) Corresponding chemical structure. i) STM topogra-phy of Terminal 3 with two small nodes at the top edge of BI-7-AGNR.j) Constant height dI/dV map of Terminal 3. k) Corresponding chemicalstructure. l) Corresponding chemical structure. STM measurementparameters: V=200 mV and I=5 pA in (a, i). V=200 mV and I=3 pAin (e). For constant height dI/dV maps: V=1 mV, Vac=10 mV in (b, f).V=0, Vac=1 mV in (j).Figure 3. STS characterization of three types of termini structures.a) Bond-resolved image of Terminal 1 with the chemical structuresuperimposed. b) dI/dV curves recorded at different sites as indicatedby the colored dots in (a). c) Bond-resolved image of Terminal 2 withthe chemical structure superimposed. d) dI/dV curves measured atdifferent sites as indicated by the colored dots in (c). e) Bond-resolvedimage of Terminal 3 with the chemical structure superimposed. f) dI/dV curves measured at different sites as indicated by the colored dotsin (f). The dashed lines overlapping the red lines in (d) and (f) wereobtained from Frota fitting. The bond-resolved images in (a, c, e) areidentical with Figure 2b, f and j, respectively. Measurement parameters:For constant height dI/dV maps: V=1 mV, Vac=10 mV in (a, c). V=0,Vac=1 mV in (e). V=100 mV, I=100 pA, Vac=1 mV for STS in (b).V=30 mV, I=200 pA, Vac=0.3 mV for STS in (d). V=30 mV,I=100 pA, Vac=0.3 mV for STS in (f).AngewandteChemieResearch ArticlesAngew. Chem. Int. Ed. 2023, 62, e202302534 (4 of 6) © 2023 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH 15213773, 2023, 24, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.202302534 by Cochrane Japan, Wiley Online Library on [24/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensemodification to GNR, which is expected to effectivelydecouple GNR from metal substrate, e.g. indeno-fused zigzag-GNR on Au(111).[2c]To study the structure and the electronic state of theGNR adsorbed on Au(111) in depth, DFT calculations werecarried out as they provide key factors for understanding theexperimental results, especially the zero-bias peak in Fig-ure 3. Figures 4a and b show the optimized geometry of theBI-7-AGNR with 4 units containing Terminal 2 (see alsoFigure S22) and the corresponding spin density, respectively.We found the spin density localized around the left terminalof the GNR, which is in agreement with the experimentalresult in Figure 3c. The side view of the DFT optimizedstructure (Figure 4a) shows that the adsorption height of thezigzag edge is slightly larger than that of the rest; the unitwith the edge has a mean adsorption height of 2.65 Å whilethat of the whole GNR is 2.35 Å. Such corrugated structurereduces the interaction between the zigzag edge and theunderlying gold substrate. The more specific details aboutgaps between GNR and substrate can be seen in Figure S23,in which the z-position (perpendicular to the Au(111)surface) of the C atoms is plotted in the bar chart. Toinvestigate the origin of spin density, we performed DFTcalculations on periodic GNR with a pentagon ring fused atthe bulk edge (Figure S24), showing almost no spin density.The spin densities of Terminals 1 and 3 were also calculatedand the results are displayed in Figure S25. DFT calculationsindicated that Terminal 1 exhibits a negligible spin density,while the relatively strong spin density for Terminal 3 wereobserved. Terminal 3 displays more delocalized spin densitythan that of Terminal 2. Thus, Terminal 2 exhibits thehighest zero-bias peak, Terminal 3 lower zero-bias peak, andTerminal 1 no peak, which is in good agreement with theexperimental results (Figure 3). To further analyze therelationship between spin density and the GNR-Au(111)interaction, the GNR terminus was artificially lifted by 1.78–3.28 Å (Figures 4c). The distribution plot clearly shows thatspin density around the GNR terminus significantly enhan-ces with increasing the adsorption height. Thus, DFTcalculations revealed that Terminal 2 and 3 are significantlynonplanar, increasing their adsorption height on Au(111),which decreases the interaction between GNR termini andAu(111) surface, and recovers the spin polarization.ConclusionIn summary, we demonstrate a synthesis of π-extended 7-AGNR with unique edge and termini structures using 7-bromo-12-(10-bromoanthracen-9-yl)tetraphene as the pre-cursor on Au(111). During the process of annealing, weobserved the appearance of 3D-GNR with [6]helicenesubstructures. Further annealing resulted in the formation ofplanarized BI-7-AGNR with three type of termini struc-tures, involving strain-induced skeletal rearrangement at the[6]helicene moieties. A Kondo resonance appeared even ona bare Au(111) surface for nonplanar zigzag termini, whichmay allow further investigations of the magnetic propertiesof the GNRs. DFT calculations revealed that terminalstructural deformation induced a decoupling from thesubstrate, restoring the spin polarization. Although furtheroptimization of the precursor structure is necessary forobtaining GNRs with uniformly fused helicenes, we believethat these results pave the way toward the synthesis of other3D-GNRs with helicene-incorporated edge structures.Moreover, the introduction of the local structural deforma-tion may be the key to realizing magnetic properties of othercarbon-based materials on metal substrates.AcknowledgementsThis work was financially supported in part by Japan Societyfor the Promotion of Science (JSPS) KAKENHI GrantNumbers 19K24686, 21K18885, 21F21058, and 22H00285,and the Okinawa Institute of Science and TechnologyGraduate University. We appreciate the help and supportprovided by the Instrumental Analysis Section of ResearchSupport Division at OIST.Conflict of InterestThe authors declare no conflict of interest.Data Availability StatementThe data that support the findings of this study are availablefrom the corresponding author upon reasonable request.Figure 4. DFT calculations of the GNR segment on Au(111). a) Opti-mized geometry. b) Spin density at the optimized geometry. The upperand lower Figures show the top and side views, respectively. c) Spindensity at the optimized GNR segments at different distances from theAu(111) surface of 1.78–3.28 Å. The spin density isosurface of3×10� 3 e is used for all plots.AngewandteChemieResearch ArticlesAngew. Chem. Int. Ed. 2023, 62, e202302534 (5 of 6) © 2023 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH 15213773, 2023, 24, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.202302534 by Cochrane Japan, Wiley Online Library on [24/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons LicenseKeywords: Graphene Nanoribbons · On-Surface Synthesis ·Scanning Tunneling Microscopy · Spin Polarization · ZigzagTermini[1] a) Y. Yano, N. Mitoma, H. Ito, K. Itami, J. Org. Chem. 2020,85, 4–33; b) A. Borissov, Y. K. Maurya, L. Moshniaha, W.-S.Wong, M. Żyła-Karwowska, M. Stępień, Chem. Rev. 2022, 122,565–788; c) Z. Liu, S. Fu, X. Liu, A. Narita, P. Samorì, M.Bonn, H. I. Wang, Adv. Sci. 2022, 9, 2106055; d) D. Jacob, R.Ortiz, J. Fernández-Rossier, Phys. Rev. B 2021, 104, 075404;e) S. Song, J. Su, M. Telychko, J. Li, G. Li, Y. Li, C. Su, J. Wu,J. Lu, Chem. Soc. Rev. 2021, 50, 3238–3262; f) J. Li, Y. Hu, L.Yu, L. Li, D. Ji, L. Li, W. Hu, H. Fuchs, Small 2021, 17,2100724; g) R. Kawaguchi, K. Hashimoto, T. Kakudate, K.Katoh, M. Yamashita, T. Komeda, Nano Lett. 2023, 23, 213–219.[2] a) L. Talirz, P. Ruffieux, R. Fasel, Adv. Mater. 2016, 28, 6222–6231; b) J. Cai, P. Ruffieux, R. Jaafar, M. Bieri, T. Braun, S.Blankenburg, M. Muoth, A. P. Seitsonen, M. Saleh, X. Feng,K. Müllen, R. Fasel, Nature 2010, 466, 470–473; c) P. Ruffieux,S. Wang, B. Yang, C. Sánchez-Sánchez, J. Liu, T. Dienel, L.Talirz, P. Shinde, C. A. Pignedoli, D. Passerone, T. Dumslaff,X. Feng, K. Müllen, R. Fasel, Nature 2016, 531, 489–492; d) Y.-C. Chen, D. G. de Oteyza, Z. Pedramrazi, C. Chen, F. R.Fischer, M. F. Crommie, ACS Nano 2013, 7, 6123–6128; e) H.Zhang, H. Lin, K. Sun, L. Chen, Y. Zagranyarski, N. Aghdassi,S. Duhm, Q. Li, D. Zhong, Y. Li, K. Müllen, H. Fuchs, L. Chi,J. Am. Chem. Soc. 2015, 137, 4022–4025; f) L. Talirz, H. Söde,T. Dumslaff, S. Wang, J. R. Sanchez-Valencia, J. Liu, P.Shinde, C. A. Pignedoli, L. Liang, V. Meunier, N. C. Plumb, M.Shi, X. Feng, A. Narita, K. Müllen, R. Fasel, P. Ruffieux, ACSNano 2017, 11, 1380–1388; g) J. Liu, B.-W. Li, Y.-Z. Tan, A.Giannakopoulos, C. Sanchez-Sanchez, D. Beljonne, P. Ruffi-eux, R. Fasel, X. Feng, K. Müllen, J. Am. Chem. Soc. 2015,137, 6097–6103; h) P. Han, K. Akagi, F. Federici Canova, H.Mutoh, S. Shiraki, K. Iwaya, P. S. Weiss, N. Asao, T. Hitosugi,ACS Nano 2014, 8, 9181–9187; i) S. Kawai, S. Saito, S. Osumi,S. Yamaguchi, A. S. Foster, P. Spijker, E. Meyer, Nat.Commun. 2015, 6, 8098; j) S. Clair, D. G. de Oteyza, Chem.Rev. 2019, 119, 4717–4776; k) X. Liu, G. Li, A. Lipatov, T. Sun,M. Mehdi Pour, N. R. Aluru, J. W. Lyding, A. Sinitskii, NanoRes. 2020, 13, 1713–1722; l) Q. Fan, L. Yan, M. W. Tripp, O.Krejčí, S. Dimosthenous, S. R. Kachel, M. Chen, A. S. Foster,U. Koert, P. Liljeroth, J. M. Gottfried, Science 2021, 372, 852–856; m) Q. Fan, D. Martin-Jimenez, D. Ebeling, C. K. Krug, L.Brechmann, C. Kohlmeyer, G. Hilt, W. Hieringer, A. Schir-meisen, J. M. Gottfried, J. Am. Chem. Soc. 2019, 141, 17713–17720.[3] a) M. Treier, C. A. Pignedoli, T. Laino, R. Rieger, K. Müllen,D. Passerone, R. Fasel, Nat. Chem. 2011, 3, 61–67; b) C.Rogers, C. Chen, Z. Pedramrazi, A. A. Omrani, H.-Z. Tsai,H. S. Jung, S. Lin, M. F. Crommie, F. R. Fischer, Angew.Chem. Int. Ed. 2015, 54, 15143–15146; c) Q. Fan, D. Martin-Jimenez, S. Werner, D. Ebeling, T. Koehler, T. Vollgraff, J.Sundermeyer, W. Hieringer, A. Schirmeisen, J. M. Gottfried, J.Am. Chem. Soc. 2020, 142, 894–899.[4] a) L. Grill, S. Hecht, Nat. Chem. 2020, 12, 115–130; b) L. Grill,M. Dyer, L. Lafferentz, M. Persson, M. V. Peters, S. Hecht,Nat. Nanotechnol. 2007, 2, 687–691.[5] J. Fernández-Rossier, J. J. Palacios, Phys. Rev. Lett. 2007, 99,177204.[6] a) S. Mishra, D. Beyer, K. Eimre, J. Liu, R. Berger, O.Gröning, C. A. Pignedoli, K. Müllen, R. Fasel, X. Feng, P.Ruffieux, J. Am. Chem. Soc. 2019, 141, 10621–10625; b) J. Su,M. Telychko, P. Hu, G. Macam, P. Mutombo, H. Zhang, Y.Bao, F. Cheng, Z.-Q. Huang, Z. Qiu, S. J. R. Tan, H. Lin, P.Jelínek, F.-C. Chuang, J. Wu, J. Lu, Sci. Adv. 2019, 5,eaav7717; c) J. Li, S. Sanz, J. Castro-Esteban, M. Vilas-Varela,N. Friedrich, T. Frederiksen, D. Peña, J. I. Pascual, Phys. Rev.Lett. 2020, 124, 177201; d) Q. Sun, X. Yao, O. Gröning, K.Eimre, C. A. Pignedoli, K. Müllen, A. Narita, R. Fasel, P.Ruffieux, Nano Lett. 2020, 20, 6429–6436; e) Y. Zheng, C. Li,C. Xu, D. Beyer, X. Yue, Y. Zhao, G. Wang, D. Guan, Y. Li,H. Zheng, C. Liu, J. Liu, X. Wang, W. Luo, X. Feng, S. Wang,J. Jia, Nat. Commun. 2020, 11, 6076.[7] a) S. Mishra, D. Beyer, K. Eimre, S. Kezilebieke, R. Berger, O.Gröning, C. A. Pignedoli, K. Müllen, P. Liljeroth, P. Ruffieux,X. Feng, R. Fasel, Nat. Nanotechnol. 2020, 15, 22–28; b) E.Turco, S. Mishra, J. Melidonie, K. Eimre, S. Obermann, C. A.Pignedoli, R. Fasel, X. Feng, P. Ruffieux, J. Phys. Chem. Lett.2021, 12, 8314–8319; c) S. Mishra, X. Yao, Q. Chen, K. Eimre,O. Gröning, R. Ortiz, M. Di Giovannantonio, J. C. Sancho-García, J. Fernández-Rossier, C. A. Pignedoli, K. Müllen, P.Ruffieux, A. Narita, R. Fasel, Nat. Chem. 2021, 13, 581–586;d) J. Li, S. Sanz, M. Corso, D. J. Choi, D. Peña, T. Frederiksen,J. I. Pascual, Nat. Commun. 2019, 10, 200.[8] M. Ijäs, M. Ervasti, A. Uppstu, P. Liljeroth, J. van der Lit, I.Swart, A. Harju, Phys. Rev. B 2013, 88, 075429.[9] P. Ruffieux, J. Cai, N. C. Plumb, L. Patthey, D. Prezzi, A.Ferretti, E. Molinari, X. Feng, K. Müllen, C. A. Pignedoli, R.Fasel, ACS Nano 2012, 6, 6930–6935.[10] S. Wang, L. Talirz, C. A. Pignedoli, X. Feng, K. Müllen, R.Fasel, P. Ruffieux, Nat. Commun. 2016, 7, 11507.[11] a) S. Mishra, D. Beyer, R. Berger, J. Liu, O. Gröning, J. I.Urgel, K. Müllen, P. Ruffieux, X. Feng, R. Fasel, J. Am. Chem.Soc. 2020, 142, 1147–1152; b) J. Lawrence, P. Brandimarte, A.Berdonces-Layunta, M. S. G. Mohammed, A. Grewal, C. C.Leon, D. Sánchez-Portal, D. G. de Oteyza, ACS Nano 2020,14, 4499–4508.[12] K. Sun, O. J. Silveira, S. Saito, K. Sagisaka, S. Yamaguchi,A. S. Foster, S. Kawai, ACS Nano 2022, 16, 11244–11250.[13] T. Wang, S. Sanz, J. Castro-Esteban, J. Lawrence, A.Berdonces-Layunta, M. S. G. Mohammed, M. Vilas-Varela, M.Corso, D. Peña, T. Frederiksen, D. G. de Oteyza, Nano Lett.2022, 22, 164–171.[14] S. Kawai, O. Krejčí, T. Nishiuchi, K. Sahara, T. Kodama, R.Pawlak, E. Meyer, T. Kubo, A. S. Foster, Sci. Adv. 2020, 6,eaay8913.[15] F.-L. Zhang, K. Hong, T.-J. Li, H. Park, J.-Q. Yu, Science 2016,351, 252–256.[16] S. Blankenburg, J. Cai, P. Ruffieux, R. Jaafar, D. Passerone, X.Feng, K. Müllen, R. Fasel, C. A. Pignedoli, ACS Nano 2012, 6,2020–2025.[17] O. Stetsovych, M. Švec, J. Vacek, J. V. Chocholoušová, A.Jančařík, J. Rybáček, K. Kosmider, I. G. Stará, P. Jelínek, I.Starý, Nat. Chem. 2017, 9, 213–218.[18] a) R. Temirov, S. Soubatch, O. Neucheva, A. C. Lassise, F. S.Tautz, New J. Phys. 2008, 10, 053012; b) L. Gross, F. Mohn, N.Moll, P. Liljeroth, G. Meyer, Science 2009, 325, 1110–1114.[19] J. Björk, C. Sánchez-Sánchez, Q. Chen, C. A. Pignedoli, J.Rosen, P. Ruffieux, X. Feng, A. Narita, K. Müllen, R. Fasel,Angew. Chem. Int. Ed. 2022, 61, e202212354.[20] A. Kinikar, M. Di Giovannantonio, J. I. Urgel, K. Eimre, Z.Qiu, Y. Gu, E. Jin, A. Narita, X.-Y. Wang, K. Müllen, P.Ruffieux, C. A. Pignedoli, R. Fasel, Nat. Synth. 2022, 1, 289–296.[21] J. van der Lit, M. P. Boneschanscher, D. Vanmaekelbergh, M.Ijäs, A. Uppstu, M. Ervasti, A. Harju, P. Liljeroth, I. Swart,Nat. Commun. 2013, 4, 2023.Manuscript received: February 19, 2023Accepted manuscript online: March 16, 2023Version of record online: April 13, 2023AngewandteChemieResearch ArticlesAngew. Chem. Int. Ed. 2023, 62, e202302534 (6 of 6) © 2023 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH 15213773, 2023, 24, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.202302534 by Cochrane Japan, Wiley Online Library on [24/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttps://doi.org/10.1021/acs.joc.9b02814https://doi.org/10.1021/acs.joc.9b02814https://doi.org/10.1021/acs.chemrev.1c00449https://doi.org/10.1021/acs.chemrev.1c00449https://doi.org/10.1002/advs.202106055https://doi.org/10.1039/D0CS01060Jhttps://doi.org/10.1002/smll.202100724https://doi.org/10.1002/smll.202100724https://doi.org/10.1021/acs.nanolett.2c04049https://doi.org/10.1021/acs.nanolett.2c04049https://doi.org/10.1002/adma.201505738https://doi.org/10.1002/adma.201505738https://doi.org/10.1038/nature09211https://doi.org/10.1038/nature17151https://doi.org/10.1021/nn401948ehttps://doi.org/10.1021/ja511995rhttps://doi.org/10.1021/acsnano.6b06405https://doi.org/10.1021/acsnano.6b06405https://doi.org/10.1021/jacs.5b03017https://doi.org/10.1021/jacs.5b03017https://doi.org/10.1021/nn5028642https://doi.org/10.1021/acs.chemrev.8b00601https://doi.org/10.1021/acs.chemrev.8b00601https://doi.org/10.1007/s12274-020-2797-6https://doi.org/10.1007/s12274-020-2797-6https://doi.org/10.1126/science.abg4509https://doi.org/10.1126/science.abg4509https://doi.org/10.1021/jacs.9b08060https://doi.org/10.1021/jacs.9b08060https://doi.org/10.1038/nchem.891https://doi.org/10.1002/anie.201507104https://doi.org/10.1002/anie.201507104https://doi.org/10.1021/jacs.9b10151https://doi.org/10.1021/jacs.9b10151https://doi.org/10.1038/s41557-019-0392-9https://doi.org/10.1038/nnano.2007.346https://doi.org/10.1021/jacs.9b05319https://doi.org/10.1021/acs.nanolett.0c02077https://doi.org/10.1038/s41565-019-0577-9https://doi.org/10.1021/acs.jpclett.1c02381https://doi.org/10.1021/acs.jpclett.1c02381https://doi.org/10.1038/s41557-021-00678-2https://doi.org/10.1021/nn3021376https://doi.org/10.1021/jacs.9b09212https://doi.org/10.1021/jacs.9b09212https://doi.org/10.1021/acsnano.9b10191https://doi.org/10.1021/acsnano.9b10191https://doi.org/10.1021/acsnano.2c04563https://doi.org/10.1021/acs.nanolett.1c03578https://doi.org/10.1021/acs.nanolett.1c03578https://doi.org/10.1126/science.aad7893https://doi.org/10.1126/science.aad7893https://doi.org/10.1021/nn203129ahttps://doi.org/10.1021/nn203129ahttps://doi.org/10.1038/nchem.2662https://doi.org/10.1088/1367-2630/10/5/053012https://doi.org/10.1126/science.1176210https://doi.org/10.1038/s44160-022-00032-5https://doi.org/10.1038/s44160-022-00032-5