Strongly Hole-Doped and Highly Decoupled Graphene on Platinum by

Letter Cite This: J. Phys. Chem. Lett. 2019, 10, 3998−4002

pubs.acs.org/JPCL

Strongly Hole-Doped and Highly Decoupled Graphene on Platinum by Water Intercalation Zhe Li,*,† Shiqi Li,‡ Hsin-Yi Tiffany Chen,§ Nan Gao,‡ Koen Schouteden,∥,⊗ Xiaoming Qiang,‡ Jijun Zhao,*,‡ Steven Brems,⊥ Cedric Huyghebaert,⊥ and Chris Van Haesendonck∥

Downloaded via BUFFALO STATE on July 24, 2019 at 14:22:04 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

†

State Key Laboratory on Tunable Laser Technology, Ministry of Industry and Information Technology Key Lab of Micro-Nano Optoelectronic Information System, School of Science, Harbin Institute of Technology (Shenzhen), Shenzhen 518055, China ‡ Key Laboratory of Materials Modification by Laser, Ion and Electron Beams(Dalian University of Technology), Ministry of Education, Dalian 116024, China § Department of Engineering and System Science, National Tsing Hua University, Hsinchu 30010, Taiwan ∥ Laboratory of Solid-State Physics and Magnetism, KU Leuven, BE-3001 Leuven, Belgium ⊥ Interuniversitair Micro-Electronica Centrum (imec) vzw, Kapeldreef 75, BE-3001 Leuven, Belgium S Supporting Information *

ABSTRACT: Scanning tunneling microscopy and spectroscopy experiments under ultrahigh vacuum and low-temperature conditions have been performed on waterintercalated graphene on Pt(111). We find that the confined water layer, with a thickness around 0.35 nm, induces a strong hole doping in graphene, i.e., the Dirac point locates at round 0.64 eV above the Fermi level. This can be explained by the presence of a single “puckered bilayer” of ice-Ih, which has not been experimentally found on bare Pt(111), being confined in between graphene and Pt(111) surface. Moreover, the water intercalation makes graphene highly decoupled from the substrate, allowing us to reveal the intrinsic graphene phonons and double Rydberg series of even and odd symmetry image-potential states. Our work not only demonstrates that the electronic properties of graphene can be tuned by the confined water layer between graphene and the substrate, but also provides a generally applicable method to study the intrinsic properties of graphene as well as of other supported two-dimensional materials.

W

graphene on bare Pt(111). This indicates that the water layer confined in between graphene and the Pt(111) surface induces a strong p-type doping in graphene. Combining the topographic and electronic properties of graphene, as well as density functional theory (DFT) calculations, we propose that the intercalated water layer is a single “puckered bilayer” of iceIh.10 We also demonstrate that intrinsic properties, i.e., phonon modes and image-potential states (IPSs), of graphene can be revealed due to the pronounced decoupling of the graphene by water intercalation. The growth of graphene (G) on Pt(111) films on sapphire is carried out in a cold-wall chemical vapor deposition reactor (6” Aixtron BM, see details in ref 7). The freshly grown G/Pt is then kept in ultrapure water for about 12 h, resulting in interfacial water layers between graphene and Pt(111).7 The sample is annealed at 500 °C during 12 h in ultrahigh vacuum prior to scanning tunneling microscopy (STM) investigation in order to desorb surface contamination resulting from the exposure to ambient conditions. All STM measurements are performed at 4.5 K using a tungsten tip in an ultrahigh vacuum

ater at surfaces and interfaces is of fundamental interest in multidiscipline research. Two-dimensional materials, especially graphene, have been used as coating layers to study the properties of the water adlayers on various insulating and conducting surfaces.1−6 With respect to possible applications, we recently showed that interfacial water between graphene and its growth substrate Pt(111) is essential for a direct transfer of graphene from a Pt(111) surface to a target wafer.7 Due to the permanent dipole moment of a water molecule, the interfacially confined water layer in turn may influence the electronic properties such as charge doping of the coating graphene layer.6,8,9 For example, it has been reported that, while mica introduces strong p-type doping in graphene, the intercalated water layers result in a significantly reduced hole doping.9 For graphene on metallic surfaces, while ex situ transferred graphene on bare Au(111) surfaces is only slightly electron-doped, the intercalation with water molecules leads to a stronger n-doping in graphene.8 Therefore, the doping effects induced by the intercalated water depend on the type of substrate on which graphene is grown, where the confined interfacial water layer exhibits different structures. Here, we demonstrate that the Dirac point of the waterintercalated graphene shifts toward higher energy and its work function becomes higher compared to the work function of © 2019 American Chemical Society

Received: May 23, 2019 Accepted: July 1, 2019 Published: July 1, 2019 3998

DOI: 10.1021/acs.jpclett.9b01488 J. Phys. Chem. Lett. 2019, 10, 3998−4002

Letter

The Journal of Physical Chemistry Letters system that includes a low-temperature STM setup (Scienta Omicron). (dI/dV)(V) curves are acquired with open and closed feedback loop by lock-in detection using amplitudes of 50 mV and frequencies around 880 Hz. (d2I/dV2)(V) spectra are acquired with open feedback loop by lock-in detection using amplitudes of 5 mV and frequencies around 300 Hz. Prior to performing scanning tunneling spectroscopy (STS) measurements on the graphene sample, the STM tips are first trained and characterized on Au(111) surfaces until a clear onset of the Au(111) surface states is present in the dI/dV curves. Image processing is performed by Nanotec WSxM.11 Figure 1(a) presents the STM topography image of the partially water-intercalated graphene on Pt(111). The brighter

Figure 1. (a) 100 × 100 nm2 STM topography image (V = 0.2 V, I = 0.1 nA) of partially water-intercalated graphene on a Pt(111) substrate. (b) A close-up (13.3 × 6.6 nm2) of the dashed black square in (a), recorded at V = 0.2 V and I = 0.2 nA. The inset image shows the atomic resolution on G/Pt, where both the atomic lattice and a moiré pattern can be resolved. (c) Height profile of G/H2O/Pt along the green dashed line in (b).

Figure 2. (a) dI/dV spectra recorded on G/Pt (dotted black curve) and G/H2O/Pt (solid red curve) regions. The dashed black and the solid red arrows indicate the position of the Dirac point for G/Pt and G/H2O/Pt, respectively. (b) d2I/dV2 spectra recorded on G/Pt (dotted black curve) and G/H2O/Pt (solid red curve) regions. The arrows indicate the inelastic transitions (P1 to P4), which can be attributed to phonons in graphene. (c) 6.5 V × 5.8 nm color visualization of the (dZ/dV)(V) spectra taken along the green arrow in the inset of (d). The horizontal axis shares the same scale with that of (d). The dotted white line indicates the boundary between the G/ Pt and G/H2O/Pt regions along the green arrow. (d) High-voltage dZ/dV spectra recorded with closed feedback loop on G/Pt (dotted black curve) and G/H2O/Pt (solid red curve) regions. The peaks 1 (1′), 2 (2′), and 3 indicate the image-potential states for G/Pt (G/ H2O/Pt). Inset: STM topography, size 8 nm × 8 nm.

meander-type patterns correspond to the areas where there is a water layer intercalated between graphene and Pt(111), which is hereafter referred to as G/H2O/Pt. The darker flat regions are graphene on Pt(111) without water intercalation, which is hereafter referred to as G/Pt. Note that the coverage of the intercalated water in Figure 1(a) is too low for graphene transfer,7 while it is convenient to compare the electronic properties between the G/Pt and G/H2O/Pt regions. The graphene atomic lattice is resolved on both the G/H2O/Pt and the G/Pt regions, while an additional moiré pattern appears on G/Pt, as illustrated in Figure 1(b). The height difference between the G/H2O/Pt and G/Pt regions is measured to be 0.35 ± 0.06 nm [Figure 1(c)]. Considering that the height of a monolayer of a “puckered bilayer” of ordinary ice (ice-Ih) is about 0.37 nm,10 our water layer can be assigned as a monolayer of ice-Ih. The electronic properties of the G/H2O/Pt and G/Pt regions are significantly different, as revealed by a series of STS measurements. Figure 2(a) presents the dI/dV curves recorded on G/Pt and G/H2O/Pt. There occurs a minimum at 0.29 eV for the dI/dV curve taken on a G/Pt region (indicated by the black dashed arrow), which can be associated with the position of the graphene’s Dirac point.12 For G/H2O/Pt, besides the gap-like feature around the Fermi level, an additional minimum appears at 0.64 eV (indicated by the red solid arrow), indicating that the Dirac point shifts toward higher energy when compared to that of G/Pt. The assignment of the Dirac point at around 0.64 eV is further supported by our DFT calculations (see below). The Dirac point shift implies that the

intercalated water layer induces additional hole doping to the graphene. In the dI/dV spectrum of the G/H2O/Pt [red spectrum in Figure 2(a)] there appears a gap-like feature around the Fermi level. It has been demonstrated that such a gap-like feature is present whenever the graphene is very weakly coupled to the substrate and is related to a phonon-assisted inelastic tunneling process.13,14 A further detailed analysis around the gap region allows us to reveal the intrinsic phonon modes of graphene. Figure 2(b) shows the inelastic tunneling spectroscopy (IETS, i.e., d2I/dV2 spectrum) taken on G/H2O/Pt (red solid curve), which shows inversion-symmetry maxima (P1 to P4 indicated by the black arrows). The peaks appearing for positive energies as well as the inverse peaks appearing for negative energies reflect the phonon threshold energies.15 The energies for the four peaks are P1 = 56.4 ± 4.0 meV, P2 = 76.3 ± 5.0 meV, P3 = 148.0 ± 8.0 meV, and P4 = 173.4 ± 10.0 meV, respectively, which is consistent with the theoretically predicted graphene 3999

DOI: 10.1021/acs.jpclett.9b01488 J. Phys. Chem. Lett. 2019, 10, 3998−4002

Letter

The Journal of Physical Chemistry Letters

modeled by a three layer slab and 1.5 nm of empty space is included to avoid spurious interactions between the replicas of the slab model. A 4 × 4 × 1 Monkhorst−Pack grid is used for reciprocal space sampling. DFT-D3 dispersion correction is applied to describe the van der Waals interactions.22 Figures 3(a) and (b) illustrate the optimized structure for a single “puckered bilayer” of ice-Ih between graphene and

phonons (first four phonon energies are 58, 78, 149, and 173 meV).15 However, for G/Pt due to the finite coupling between graphene and Pt(111), the graphene phonons are not revealed since there is neither gap-like feature around the Fermi level in the dI/dV curve [dotted black curve in Figure 2(a)] nor inversion-symmetry feature in the d2I/dV2 curve [dotted black curve in Figure 2(b)]. The presence of a water layer between graphene and Pt(111) makes graphene almost decoupled from the substrate and hence allows access to the graphene phonons. The dZ/dV spectra in Figure 2(d) reveal the presence of several unoccupied states on both the G/Pt and G/H2O/Pt regions, which originate from the IPSs. Figure 2(c) is the spatially resolved dZ/dV spectra taken along a line [the green arrow in the inset of Figure 2(d)] that crosses the G/Pt and G/H2O/Pt regions, which reveals an abrupt change of the IPSs at the border. According to theory, freestanding graphene with two exposed surfaces has a double Rydberg-like series of IPSs of even (n+) and odd (n−) symmetry for each state n.16 When graphene is attached to a support, the even and odd symmetry states cannot be observed due to the interaction with the support,17 as is the case for G/Pt (black dotted curve in Figure 2(d)). As discussed above, the graphene is nearly freestanding for G/H2O/Pt, and it is then expected that the even and odd symmetry states can be observed. Within the measured energy window, four resonances are present at 2.75, 4.18, 5.03, and 6.88 eV, which can be assigned as IPSs with n = 1+, 1−, 2+, 2−, respectively. The first two states agree with the theoretically predicted values for 1+ (between 2.94 and 3.24 eV) and 1− (between 3.69 and 4.16 eV).16 However, our measured values for 2+ and 2− are significantly larger than the theoretical values,16 which can be accounted for by a Stark shift and the stronger coupling of the higher-lying IPSs with the metal substrate.17 Considering high order (n ≥ 2) IPSs, the IPSs of G/H2O/Pt (red solid curve in Figure 2(d)) overall shifts by an amount Δ = 0.32 ± 0.02 eV toward higher energies compared to the IPSs of G/Pt (black dotted curve). The energy shift between the two series of IPSs should be similar for each of the IPS peaks, while the energy difference between the corresponding IPSs of the two supports reflects their different work functions.18 Therefore, we can conclude that a G/H2O/ Pt region has a higher work function than a G/Pt region. We infer a work function difference ΔΦ = ΦG/H2O/Pt − ΦG/Pt = 0.32 ± 0.02 eV between the two regions, which is consistent with the Dirac point shift (0.35 ± 0.02 eV) inferred from Figure 2(a). The work function enhancement on the G/H2O/ Pt regions is consistent with the water-induced additional hole doping of the graphene. We note that the analysis of the IPSs for G/H2O/Pt focuses on the broad regions of G/H2O/Pt islands (typically dimensions larger than a 7 nm × 7 nm area), while for regions with smaller dimensions confinement effects become significant (see Figure S1 in the Supporting Information). In order to further clarify the hole-doping effect of the water layer, we performed spin-polarized DFT calculations, using the generalized gradient approximation (PBE functional19) and the plane wave code VASP.20 The interaction between the ions and the valence electrons is described by the projector augmented wave (PAW) method.21 The graphene (on water layer) on the Pt(111) substrate is modeled by a coincidence structure, obtained by superposing a (4 × 4) graphene unit cell (on a 2 × 2 unit cell of single “puckered bilayer” of ice-Ih) on a (4 × 4) unit cell of the Pt(111) surface. The Pt(111) is

Figure 3. (a) Side view of the proposed structure for a single “puckered bilayer” of ice-Ih intercalated between graphene and Pt(111). The gray, red, white, and blue balls represent C, O, H, and Pt atoms, respectively. (b) Top view of the confined “puckered bilayer” of ice-Ih. The atomic lattices for graphene and Pt are not shown. The dashed lines indicate the hydrogen bonds. (c,d) Band structure plots for G/Pt and G/H2O/Pt, respectively. The Fermi level is at zero energy. The contribution of the graphene to the bands is indicated by the red solid dots. The blue open circles in (d) present the contribution of the H2O molecules.

Pt(111). Given that water molecules induce a hole doping of the graphene, the dipole orientation of the water is expected to be upward, and hence, the hydrogen atoms should face downward to the Pt(111). The calculated band structure in Figure 3(d) indicates that for G/H2O/Pt the Dirac point of the graphene is located at 0.72 eV above the Fermi level. Using the same size of the supercell [Figure 3(a)] without the water layer, the optimized structure of G/Pt results in a position of the Dirac point of graphene at 0.35 eV above the Fermi level. The relative shift of the Dirac point position between G/H2O/ Pt and G/Pt is (0.72 eV − 0.35 eV = ) 0.37 eV, which agrees very well with our experimental observation of 0.35 ± 0.02 eV. We can also calculate the energy difference of the work functions between G/H2O/Pt and G/Pt, resulting in ΔΦ = ΦG/H2O/Pt − ΦG/Pt = 0.34 eV, again consistent with our experimental result of 0.32 ± 0.02 eV. Our DFT calculations consistently confirm that a single “puckered bilayer” of ice-Ih between graphene and Pt(111) indeed results in a strong hole doping of the graphene, and even quantitatively accounts for the experimentally obtained values. In fact, the single “puckered bilayer” of ice-Ih is a ferroelectric structure, where the dipole moments of the water molecules induce electrostatic fields and result in strong hole doping in the graphene and enhance the work function. We also consider the case when hydrogen atoms of the ice layer face upward to graphene. This results in ΔΦ = ΦG/H2O/Pt − ΦG/Pt = −0.53 eV, which fails to reproduce our experimental observation. We note that in our experiment there is no observable preferred rotation angles between the ice lattice and the Pt(111) lattice. This indicates that the rotation angles for different ice layers with respect to the Pt or graphene lattice are randomly distributed. Since the experimentally derived ΔΦ between G/Pt and G/H2O/Pt regions are 0.32 ± 0.02 eV for all measured G/H2O/Pt islands, the effect of the rotation angle on the work function is expected to be limited. 4000

DOI: 10.1021/acs.jpclett.9b01488 J. Phys. Chem. Lett. 2019, 10, 3998−4002

The Journal of Physical Chemistry Letters

■

In conclusion, we demonstrated that graphene on Pt(111) becomes strongly decoupled from the metallic Pt(111) substrate due to the presence of an intercalated water layer between graphene and Pt(111), allowing us to obtain access to the intrinsic properties of the graphene, including the phonons and the double Rydberg series of even and odd symmetry image potential states. In contrast to a water layer between graphene and Au(111), which induces electron doping of the graphene,8 a water layer between graphene and Pt(111) results in hole doping. This indicates that the substrate plays an important role in determining the orientation of the water molecular dipoles and hence also the growth of the confined water layers. Relying on DFT calculations, we show that the intercalated water layer is a single “puckered bilayer” of ice-Ih with the H atoms pointing downward to the Pt(111), which is different from the structure of bare water monolayers on Pt(111).23−25 Since it was recently reported that the singlephoton and multiphoton resonant transitions can be tuned using an electric gate voltage to adjust the chemical potential (Fermi level), i.e., the position of the Dirac point, of graphene, intercalated water induced charge doping in graphene may be used as effective gating for tailoring the third-order optical nonlinearities of graphene.26

■

REFERENCES

(1) Li, Q.; Song, J.; Besenbacher, F.; Dong, M. Two-Dimensional Material Confined Water. Acc. Chem. Res. 2015, 48, 119−127. (2) Bampoulis, P.; Sotthewes, K.; Dollekamp, E.; Poelsema, B. Water Confined in Two-Dimensions: Fundamentals and Applications. Surf. Sci. Rep. 2018, 73, 233−264. (3) Song, J.; Li, Q.; Wang, X.; Li, J.; Zhang, S.; Kjems, J.; Besenbacher, F.; Dong, M. Evidence of Stranski−Krastanov Growth at the Initial Stage of Atmospheric Water Condensation. Nat. Commun. 2014, 5, 4837. (4) Temmen, M.; Ochedowski, O.; Schleberger, M.; Reichling, M.; Bollmann, T. R. J. Hydration Layers Trapped between Graphene and a Hydrophilic Substrate. New J. Phys. 2014, 16, 053039. (5) Bampoulis, P.; Siekman, M. H.; Kooij, E. S.; Lohse, D.; Zandvliet, H. J. W.; Poelsema, B. Latent Heat Induced Rotation Limited Aggregation in 2D Ice Nanocrystals. J. Chem. Phys. 2015, 143, 034702. (6) Bollmann, T. R. J.; Antipina, L. Y.; Temmen, M.; Reichling, M.; Sorokin, P. B. Hole-Doping of Mechanically Exfoliated Graphene by Confined Hydration Layers. Nano Res. 2015, 8, 3020−3026. (7) Verguts, K.; Schouteden, K.; Wu, C.-H.; Peters, L.; Vrancken, N.; Wu, X.; Li, Z.; Erkens, M.; Porret, C.; Huyghebaert, C.; Van Haesendonck, C.; De Gendt, S.; Brems, S. Controlling Water Intercalation Is Key to a Direct Graphene Transfer. ACS Appl. Mater. Interfaces 2017, 9, 37484−37492. (8) Cao, P.; Varghese, J. O.; Xu, K.; Heath, J. R. Visualizing Local Doping Effects of Individual Water Clusters on Gold(111)-Supported Graphene. Nano Lett. 2012, 12, 1459−1463. (9) Shim, J.; Lui, C. H.; Ko, T. Y.; Yu, Y.-J.; Kim, P.; Heinz, T. F.; Ryu, S. Water-Gated Charge Doping of Graphene Induced by Mica Substrates. Nano Lett. 2012, 12, 648−654. (10) Xu, K.; Cao, P.; Heath, J. R. Graphene Visualizes the First Water Adlayers on Mica at Ambient Conditions. Science 2010, 329, 1188. (11) Horcas, I.; Fernandez, R.; Gomez-Rodriguez, J. M.; Colchero, J.; Gomez-Herrero, J.; Baro, A. M. WSxM: A Software for Scanning Probe Microscopy and a Tool for Nanotechnology. Rev. Sci. Instrum. 2007, 78, 013705−8. (12) Ugeda, M. M.; Fernandez-Torre, D.; Brihuega, I.; Pou, P.; Martinez-Galera, A. J.; Perez, R.; Gomez-Rodriguez, J. M. Point Defects on Graphene on Metals. Phys. Rev. Lett. 2011, 107, 116803. (13) Zhang, Y.; Brar, V. W.; Wang, F.; Girit, C.; Yayon, Y.; Panlasigui, M.; Zettl, A.; Crommie, M. F. Giant Phonon-Induced Conductance in Scanning Tunnelling Spectroscopy of Gate-Tunable Graphene. Nat. Phys. 2008, 4, 627−630. (14) Kim, H. W.; Ko, W.; Ku, J.; Jeon, I.; Kim, D.; Kwon, H.; Oh, Y.; Ryu, S.; Kuk, Y.; Hwang, S. W.; Suh, H. Nanoscale Control of Phonon Excitations in Graphene. Nat. Commun. 2015, 6, 7528. (15) Natterer, F. D.; Zhao, Y.; Wyrick, J.; Chan, Y.-H.; Ruan, W.-Y.; Chou, M.-Y.; Watanabe, K.; Taniguchi, T.; Zhitenev, N. B.; Stroscio, J. A. Strong Asymmetric Charge Carrier Dependence in Inelastic Electron Tunneling Spectroscopy of Graphene Phonons. Phys. Rev. Lett. 2015, 114, 245502. (16) Silkin, V. M.; Zhao, J.; Guinea, F.; Chulkov, E. V.; Echenique, P. M.; Petek, H. Image Potential States in Graphene. Phys. Rev. B: Condens. Matter Mater. Phys. 2009, 80, 121408. (17) Armbrust, N.; Güdde, J.; Höfer, U. Formation of ImagePotential States at the Graphene/Metal Interface. New J. Phys. 2015, 17, 103043. (18) Li, Z.; Chen, H.-Y. T.; Schouteden, K.; Janssens, E.; Van Haesendonck, C.; Lievens, P.; Pacchioni, G. Spontaneous Doping of Two-Dimensional NaCl Films with Cr Atoms: Aggregation and Electronic Structure. Nanoscale 2015, 7, 2366−2373. (19) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple. Phys. Rev. Lett. 1996, 77, 3865−3868. (20) Kresse, G.; Hafner, J. Ab Initio Molecular Dynamics for Liquid Metals. Phys. Rev. B: Condens. Matter Mater. Phys. 1993, 47, 558−561. (21) Blöchl, P. E. Projector Augmented-Wwave Method. Phys. Rev. B: Condens. Matter Mater. Phys. 1994, 50, 17953−17979.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpclett.9b01488.

■

Letter

Image-potential states on difference sizes of G/H2O/Pt islands (PDF)

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. ORCID

Zhe Li: 0000-0002-6428-1222 Hsin-Yi Tiffany Chen: 0000-0002-9651-3200 Jijun Zhao: 0000-0002-3263-7159 Present Address ⊗

Semiconductor Physics Section, Department of Physics and Astronomy, KU Leuven, BE-3001 Leuven, Belgium Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research was supported by the National Natural Science Foundation of China (11704057, 11574040), the starting grants from HIT-Shenzhen (HA45001082), the Research FoundationFlanders (FWO), and the Flemish Concerted Research Action program (BOF KU Leuven, Project No. GOA/14/007). Z.L. and K.S. acknowledge additional support from the FWO through a postdoctoral grant. H.-Y.T.C. acknowledges financial support provided by the Ministry of Science and Technology, Taiwan (MOST 106-2112-M-007001-MY3) and the computing resource of TAIWANIA in the National Center for High-Performance Computing (NCHC), Taiwan. J.Z. acknowledges the Supercomputing Center of Dalian University of Technology. 4001

DOI: 10.1021/acs.jpclett.9b01488 J. Phys. Chem. Lett. 2019, 10, 3998−4002

Letter

The Journal of Physical Chemistry Letters (22) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A Consistent and Accurate Ab Initio Parametrization of Density Functional Dispersion Correction (DFT-D) for the 94 Elements H-Pu. J. Chem. Phys. 2010, 132, 154104. (23) Nie, S.; Feibelman, P. J.; Bartelt, N. C.; Thürmer, K. Pentagons and Heptagons in the First Water Layer on Pt(111). Phys. Rev. Lett. 2010, 105, 026102. (24) Standop, S.; Redinger, A.; Morgenstern, M.; Michely, T.; Busse, C. Molecular Structure of the H2O Wetting Layer on Pt(111). Phys. Rev. B: Condens. Matter Mater. Phys. 2010, 82, 161412. (25) Maier, S.; Lechner, B. A. J.; Somorjai, G. A.; Salmeron, M. Growth and Structure of the First Layers of Ice on Ru(0001) and Pt(111). J. Am. Chem. Soc. 2016, 138, 3145−3151. (26) Jiang, T.; Huang, D.; Cheng, J.; Fan, X.; Zhang, Z.; Shan, Y.; Yi, Y.; Dai, Y.; Shi, L.; Liu, K.; Zeng, C.; Zi, J.; Sipe, J. E.; Shen, Y.-R.; Liu, W.-T.; Wu, S. Gate-Tunable Third-Order Nonlinear Optical Response of Massless Dirac Fermions in Graphene. Nat. Photonics 2018, 12, 430−436.

4002

DOI: 10.1021/acs.jpclett.9b01488 J. Phys. Chem. Lett. 2019, 10, 3998−4002