Control of Carbon Nanotube Electronic Properties by Lithium CationIntercalationOleksandr M. Korsun,† Oleg N. Kalugin,*,† and Oleg V. Prezhdo*,‡

†Department of Inorganic Chemistry, V. N. Karazin Kharkiv National University, Kharkiv 61022, Ukraine‡Department of Chemistry, University of Southern California, Los Angeles, California 90089, United States

*S Supporting Information

ABSTRACT: We show that the electronic properties of single walled carbon nanotubes(SWCNTs) can be tuned continuously from semiconducting to metallic by varying thelocation of ions inside the tubes. Focusing on the Li+ cation inside the (26,0) zigzagsemiconducting and (15,15) armchair metallic SWCNTs, we found that the Li+-SWCNTinteraction is attractive. The interaction is stronger for the metallic SWCNT, indicating inparticular that metallic tubes can enhance performance of lithium-ion batteries. Theelectronic properties of the metallic SWCNT are virtually independent of the presence ofions: Li+ creates an energy level in the valence band slightly below the Fermi energy. Onthe contrary, the semiconducting SWCNT can be made metallic by placing ions close tothe tube axis: Li+ generates a new bottom of the conduction band. Letting the ionsapproach SWCNT walls recovers the semiconducting behavior.

SECTION: Energy Conversion and Storage; Energy and Charge Transport

Carbon nanotubes (CNTs) constitute an important novelclass of materials that exhibit unique physical and

chemical properties, leading to a variety of promisingtechnologies including optoelectronic devices, lithium-ionbatteries (LIBs), catalytic and sensing devices, mechanical andbiomedical tools, and so on.1−21 CNTs are a likely candidatefor use in LIBs due to their unique set of electrochemical andmechanical properties.22 Although lithium intercalation intocarbon nanomaterial, including CNTs, has been an interestingtopic of electrochemistry over the years, the importance of suchmaterials gained renewed attention, requiring fundamentalunderstanding of interaction of lithium atoms and cationswithin the carbon matrix at the atomic level.It was shown in a series of theoretical calculations on Li and

Li+ adsorbed on CNTs23−26 that (i) lithium prefers to localizenear CNT sidewalls, (ii) small diameter zigzag tubes could beplausible candidates for Li-ion battery applications, (iii) Liinsertion modified the electronic band structure of CNTs byshifting the Fermi level to a region with a higher density ofstates. Experimental investigations have proved the influence oftopology of single-walled CNTs (SWCNTs) on reversible Li+

ion storage capacity,27 as well as the ability of inserted lithiumto modify the CNT band structure.28,29 It should be noted thatmajority of the ab initio calculations were done on finite-lengthSWCNTs terminated by hydrogen atoms. One expects thatperiodic calculations provide a better representation of therelatively weak interaction between the intercalated lithium andquasi-infinite SWCNTs.The literature survey shows that the properties of Li+ inside

CNTs of different topologies require further quantitative

investigations, especially for electrochemical applications. Oneexpects that Li+ inside a CNT can strongly influence CNTelectronic properties, creating novel possibilities for develop-ment of electronic devices. The present investigation elucidatesthe energetic and electronic properties of Li+ ions insideSWCNTs, and compares the effects of Li+ on the electronicstructure of metallic and semiconducting SWCNTs. The studyshows that the electronic and thermodynamic properties ofLi+@SWCNT complexes depend strongly on the SWCNTtopology and location of the cation inside the CNT. Theinteraction between Li+ and SWCNTs is attractive and isstronger for metallic CNTs, suggesting in particular thatmetallic CNTs can provide enhanced Li+ ion storage capacity inLIBs. The electronic properties of semiconducting SWCNTsvary dramatically depending on the cation location, changingfrom semiconducting in pristine CNTs and CNTs with ionslocated near the wall to metallic in CNTs with ions located nearthe central axis. One can control the position of Li+ withrespect to CNTs using an insulating AFM tip30 functionalizedwith a charged cluster of Li atoms or a metallic Li STM tip withvarying applied voltage.31 The geometric and electronicstructure of the closely related fullerene systems with imbeddedLi has been mapped using STM previously.32

The polarizing action of the Li+ cation on the CNTelectronic properties was investigated by employing the (15,15)conducting armchair and (26,0) semiconducting zigzag

Received: October 13, 2014Accepted: November 14, 2014Published: November 14, 2014

Letter

pubs.acs.org/JPCL

© 2014 American Chemical Society 4129 dx.doi.org/10.1021/jz502175e | J. Phys. Chem. Lett. 2014, 5, 4129−4133

Dow

nloa

ded

via

UN

IV O

F SO

UT

HE

RN

CA

LIF

OR

NIA

on

Nov

embe

r 21

, 201

9 at

02:

17:5

7 (U

TC

).Se

e ht

tps:

//pub

s.ac

s.or

g/sh

arin

ggui

delin

es f

or o

ptio

ns o

n ho

w to

legi

timat

ely

shar

e pu

blis

hed

artic

les.

SWCNTs. These achiral CNTs have very similar internal radii,which are formally equal to 1.0171 and 1.0185 nm for the(15,15) armchair and (26,0) zigzag CNTs, respectively.33 TheCNTs have relatively large radii, compared to the tubesinvestigated in the earlier ab initio studies. The radii wereselected in order to properly represent systems employed in Li-ion batteries and to allow incorporation of solvent molecules infuture studies. To achieve the main goal of the study, extensivequantum chemical calculations of pristine SWCNTs andLi+@SWCNT complexes were carried out. The four types ofinvestigated objects are shown in Figure 1. Calculations on theLi+@SWCNT complexes were performed for varying radialshifts of Li+ with respect to CNT main symmetry axis, r(Li+).

The calculations were performed using periodic densityfunctional theory, as described in the Supporting Information.To validate the quantum-chemical level of theory used in the

present study, the electronic gaps and entropies wereinvestigated for the isolated SWCNTs. The electronic energygaps at the center of the Brillouin zone (Γ-point) are 0.06 and0.35 eV for the (15,15) metallic armchair and (26,0)semiconducting zigzag SWCNTs, respectively, in agreementwith the SWCNT types.The changes in the specific free energy functional (FEF),

ΔFFEF, and electronic entropy, ΔSel, are shown in Figure 2 asfunctions of the Li+ radial shift from the CNT axis. As followsfrom the negative ΔFFEF values, formation of the Li+@SWCNTcomplexes from the isolated Li+ cation and SWCNT isenergetically favorable for both nanotubes topologies. Thebiggest gain in the formation energy corresponds to the Li+ ionposition shifted by 0.8 nm from the SWCNT main axis. That is,the most energetically stable Li+@(15,15) and Li+@(26,0)complexes are those with the cation situated near the internalwall of the SWCNTs. The specific interaction energy risessharply starting from the 0.8 nm because Li+ approachesSWCNT walls, and contribution of the local electronicrepulsion dominates. Notably, the metallic (15,15) CNTgives a significantly larger energy gain than the semiconducting(26,0) CNT. This important result indicates that metallicCNTs are more favorable for construction of LIBs.The two SWCNTs have almost equal diameters, and the

Li+/C atom ratio is similar in both cases, ∼1:200. Therefore,one can expect that the Li+ ion affinity and, consequently,capacity should be fundamentally greater for metallic SWCNTsthan semiconducting SWCNTs. It should be noted thatcapacity is a collective property of a macroscopic sample athigh Li+ concentration. The binding energy of a single Li+ ionmay differ from the corresponding value at high Li+

concentration. Other factors, such as volumetric and weightcapacity, should be considered as well. Still, the effect willremain in realistic systems because the stronger binding of Li+

to the metallic tube compared to its semiconducting counter-part arises due to higher polarizability of the former.The right panel of Figure 2 shows that insertion of Li+ into

both types of SWCNTs increases the electronic entropy. Thisfact highlights the difference in the electronic structure of thepristine SWCNTs and the Li+@SWCNTs complexes and

Figure 1. Cross sections of the (15,15) armchair and (26,0) zigzagSWCNTs (upper panel) and corresponding SWCNTs with Li+ inside(bottom panel).

Figure 2. Changes in the specific free energy functional (ΔFFEF) and specific electronic entropy (ΔSel) due to insertion of the Li+ ion as functions ofthe Li+ radial shift from the SWCNT main axis, r(Li+).

The Journal of Physical Chemistry Letters Letter

dx.doi.org/10.1021/jz502175e | J. Phys. Chem. Lett. 2014, 5, 4129−41334130

indicates that the latter possess higher conductivity. However,the entropy gain becomes smaller as Li+ approaches thenanotube wall. When Li+ is located at the most energeticallypreferable position inside the SWCNTs, the gain in theelectronic entropy essentially vanishes.The Fermi−Dirac distributions shown in Figure 3 provide

additional details regarding the polarizing action of the Li+

cation on the SWCNT electronic structure. These data arepresented for both isolated SWCNTs and Li+@SWCNTscomplexes with different locations of the Li+ cation. The Fermilevels in Figure 3 correspond to the inflection points of theFermi−Dirac distribution functions ( f = 1.0). The isolated

(15,15) armchair and (26,0) zigzag SWCNTs exhibit sym-metrical Fermi−Dirac distributions. The electronic gaps at theΓ-point are 0.06 eV for the conducting (15,15) CNT and 0.35eV for the semiconducting (26,0) CNT.Cation insertion introduces a new electronic level, labeled by

Li+ in Figures 3 and 4. The total number of valence electrons inthe systems does not change because isolated Li+ has novalence electrons. The new electronic level can be interpretedas the partly occupied 2s sublevel of lithium. After Li+ insertion,the electronic density of SWCNTs is redistributed between theLi+ and SWCNT subsystems. The influence of the Li+ cation onthe nanotube electronic structure is distinctly different for the

Figure 3. Fermi−Dirac distributions, showing electronic level occupation, f, vs energy, ε, for the (15,15) armchair and (26,0) zigzag SWCNTs withand without the Li+ ion, color and black circles respectively, for different radial shifts of Li+ from the SWCNTs main axis, r(Li+). The circlescorrespond to the energy levels at the Γ-point. The boxes emphasize the changes in the HOMO and LUMO occupations for SWCNT due to the Li+

insertion and displacement. The “Li+” label indicates the level arising due to the ion. The Fermi energy is located at f = 1.0.

Figure 4. Electronic energy levels at the Γ-point of the (15,15) armchair and (26,0) zigzag SWCNTs with and without the Li+ ion, color and blacklines respectively, for different radial shifts of Li+ from the SWCNT main axis, r(Li+). “2” indicates 2-fold degeneracy of a given level, and “Li+” showsthat the level arises due to the ion. Gray lines connected by dashes show the Fermi energy. It depends strongly and nonuniformly on the location ofthe Li+ ion. The effect of Li+ is qualitatively different for the metallic and semiconducting tubes. In the metallic (15,15) SWCNT, Li+ creates a level inthe valence band slightly below the Fermi energy. In contrast, in the semiconducting (26,0) SWCNT, Li+ is responsible for a new bottom of theconduction band. Regardless of the SWCNT topology, the energy of the Li+ level grows rapidly and enters the SWCNT conduction band. Note thatthe energy scales are shifted in different parts of the figure.

The Journal of Physical Chemistry Letters Letter

dx.doi.org/10.1021/jz502175e | J. Phys. Chem. Lett. 2014, 5, 4129−41334131

(15,15) armchair and (26,0) zigzag SWCNTs. Insertion of Li+

inside the armchair nanotube increases the density of states atthe Fermi level. The increase is independent of the Li+ radialshift, indicating that Li+ insertion into the (15,15) armchairSWCNT generates minor perturbations in the CNT electronicstructure. Rather than perturbing the CNT, Li+ creates a newstate near the Fermi energy. The electrons located in this stateincrease the overall electronic entropy and the electricalconductivity in comparison with the pristine (15,15) armchairSWCNT.On the contrary, localization of the Li+ cation in the middle

of the (26,0) zigzag SWCNT dramatically modifies its densityof states near the Fermi level. The density becomes metal-like,with a very low electronic gap value around 0.05 eV. ShiftingLi+ away from the CNT axis restores the electronic gap. As theLi+ ion approaches the CNT wall, the value of the electronicgap for the pristine CNT is recovered. The calculations showthat the electronic gap and, therefore, the conductive orsemiconductive nature of the Li+@(26,0) complex dependsdirectly and strongly on the radial position of Li+ inside thetube. Because the CNT/Li+ interaction is long-ranged(Coulomb) and involves electron density transfer, the observedeffects should persist in CNTs with larger diameters, up to 2−3nm, and different Li+ ion concentrations.Orbital-energy diagrams clearly characterize the perturbation

induced by the Li+ cation in the SWCNT electronic structurenear the Fermi level. Figure 4 presents such diagrams for theisolated SWCNTs and the Li+@SWCNTs complexes withvarying shifts of Li+ from the nanotube axis. The energydiagrams of the isolated SWCNTs are symmetric with respectto the Fermi level. The Fermi levels are located between doublydegenerate electronic states at the Γ-points. Insertion of Li+

into the SWCNTs lowers the absolute values of the electroniclevel energies by about 0.15 and 0.13 eV for the (15,15)armchair and (26,0) zigzag SWCNTs, correspondingly.The energy levels arising directly from the SWCNTs

significantly change due to strong interaction with the cation.The CNT/Li+ interaction causes electron density transfer fromthe SWCNTs to Li+. The transfer has two effects on theSWCNTs levels, as illustrated with Figures 3 and 4. The Fermienergies shift down significantly, and the states at the edge ofthe valence bands of both SWCNTs become partially vacant.Insertion of the Li+ cation into the investigated SWCNTsmodifies the density of states near the Fermi level; however, itdoes not disrupt the electronic structure of pristine CNTsentirely. The 2-fold degeneracy of some electronic levels islifted, whereas other levels remain doubly degenerate. Thedegeneracies are lifted most strongly when the Li+ ionapproaches the energetically favorable location near theSWCNT internal wall, that is, for r(Li+) = 0.7 and 0.8 nm.At the same time, the electronic gaps of the Li+@SWCNTcomplexes at these locations of the Li+ ion are closest to thoseof isolated SWCNTs.For Li+ shifted by 0 to 0.6 nm with respect to the SWCNT

symmetry axis, one observes qualitative differences in theposition of the nondegenerate states of Li+ relative thenanotube valence and conduction bands. In the case of theLi+@(15,15) complex, the nondegenerate state is locatedslightly below the Fermi energy and near the CNT valenceband edge. In contrast, the nondegenerate state in theLi+@(26,0) complex forms the new bottom of the conductionband. As a result, the formerly semiconducting zigzag (26,0)tube becomes conducting, when the Li+ ion is placed near the

SWCNT axis. The energetic diagrams in the vicinity of theFermi level of the Li+@SWCNTs systems become very similarto the corresponding diagrams of the isolated SWCNTs, Figure4, when the Li+ ion is located in the FEF minimum, left panel ofFigure 2. This result is in full agreement with the vanishingchange in the specific electronic entropy for large shifts of theLi+ cation away from the SWCNT symmetry axis, right panel ofFigure 2.The placement of Li+ ions can be used to make

semiconducting SWCNTs conducting, to a varying extent,depending on the Li+ location inside the tube. Focusing on theright panel of Figure 4, corresponding to the (26,0)semiconducting SWCNT, one observes in the pristine (26,0)tube that the Fermi energy is located in the middle between theedges of the valence and conduction bands. When Li+ is placedon the SWCNT axis, the Fermi energy drops and approachesclosely the edge of the (26,0) valence band. This happensbecause Li+ introduced a new energy level close to the valenceband edge. At a finite temperature, a fraction of the electronicdensity transfers from the (26,0) valence band edge to thevacant Li+ level, and SWCNT acquires hole conductivity. Acation placed on the SWCNT axis affects the properties of thewhole tube. As Li+ approaches the SWCNT wall, the energy ofthe Li+ level rises, and the Fermi energy moves up. When Li+ isat the SWCNT wall, its level is above the (26,0) conductionband edge, and the electronic gap becomes the same as in thepristine (26,0) SWCNT. Close to the tube wall, Li+ only affectsSWCNT locally, and the global properties of the SWCNT, suchas the band gap, remain unchanged.The reported quantum-chemical study of the inner-tube

complexes of the Li+ cation with the (15,15) armchair metallicand (26,0) zigzag semiconducting SWCNTs showed that theelectronic properties and thermodynamics of the systemsdepend crucially on the two factors: (i) SWCNT topology and(ii) radial position of the Li+ cation. Both Li+@(15,15) andLi+@(26,0) complexes are more stable than their isolatedcomponents. In either case, the Li+ cation prefers to be near theSWCNT inner wall. At the same time, the Li+@(15,15)complex involving the metallic CNT is more energeticallystable than the Li+@(26,0) complex involving the semi-conducting CNT. This result is particularly important fordesign of high-capacity lithium-ion batteries, since it shows thathigher lithium affinity can be obtained with metallic CNTs. Theelectronic properties of the (26,0) zigzag semiconductingSWCNT are strongly dependent on the cation location. TheCNT electronic gap varies from 0.35 eV, for the isolatedSWCNT and the Li+@SWCNT complex with Li+ near theCNT wall, to 0.05 eV, for the Li+@SWCNT complex with Li+

in the center of the nanotube. The former value corresponds toa semiconductor, whereas the latter value is characteristic ofconducting materials. A possibility of creating a new class ofelectronic devices based on SWCNTs with controlledconductivity has been demonstrated. The conductivity can becontrolled by varying the position of a cation with respect tothe SWCNT main axis. In order to achieve this goal, one canuse AFM and STM techniques.30−32

■ ASSOCIATED CONTENT

*S Supporting InformationDetails of the quantum chemical calculations. This material isavailable free of charge via the Internet at http://pubs.acs.org.

The Journal of Physical Chemistry Letters Letter

dx.doi.org/10.1021/jz502175e | J. Phys. Chem. Lett. 2014, 5, 4129−41334132

■ AUTHOR INFORMATIONCorresponding Authors*E-mail: onkalugin@gmail.com.*E-mail: prezhdo@usc.edu.

NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThe simulations were performed using the joint computationalcluster of the Institute for Single Crystals and Institute forScintillation Materials of the National Academy of Science ofthe Ukraine incorporated into the Ukrainian National Grid.O.M.K. acknowledges the Fund of the fundamental, applied,and searching research works of the V. N. Karazin KharkivNational University for the financial support, grant0111U006845. O.V.P. acknowledges funding from the Officeof Basic Energy Sciences of the U.S. Department of Energy,grant DE-SC0006527.

■ REFERENCES(1) Kamat, P. V. Graphene-Based Nanoassemblies for EnergyConversion. J. Phys. Chem. Lett. 2011, 2, 242−251.(2) Dillon, A. C. Carbon Nanotubes for Photoconversion andElectrical Energy Storage. Chem. Rev. 2010, 110, 6856−6872.(3) Cui, K.; Chiba, T.; Omiya, S.; Thurakitseree, T.; Zhao, P.; Fujii,S.; Kataura, H.; Einarsson, E.; Chiashi, S.; Maruyama, S. Self-Assembled Microhoneycomb Network of Single-Walled CarbonNanotubes for Solar Cells. J. Phys. Chem. Lett. 2013, 4, 2571−2576.(4) Ulissi, Z. W.; Shimizu, S.; Lee, C. Y.; Strano, M. S. CarbonNanotubes as Molecular Conduits: Advances and Challenges forTransport through Isolated Sub-2 Nm Pores. J. Phys. Chem. Lett. 2011,2, 2892−2896.(5) Schnorr, J. M.; Swager, T. M. Emerging Applications of CarbonNanotubes†. Chem. Mater. 2010, 23, 646−657.(6) De Volder, M. F. L.; Tawfick, S. H.; Baughman, R. H.; Hart, A. J.Carbon Nanotubes: Present and Future Commercial Applications.Science 2013, 339, 535−539.(7) Mohanraj, J.; Armaroli, N. Luminophores and Carbon Nano-tubes: An Odd Combination? J. Phys. Chem. Lett. 2013, 4, 767−778.(8) Liu, H.; Cooper, V. R.; Dai, S.; Jiang, D.-e. Windowed CarbonNanotubes for Efficient CO2 Removal from Natural Gas. J. Phys. Chem.Lett. 2012, 3, 3343−3347.(9) Mishra, S.; Song, K.; Koza, J. A.; Nath, M. Synthesis ofSuperconducting Nanocables of Fese Encapsulated in CarbonaceousShell. ACS Nano 2013, 7, 1145−1154.(10) Chaban, V. V.; Prezhdo, O. V. Water Boiling inside CarbonNanotubes: Toward Efficient Drug Release. ACS Nano 2011, 5, 5647−5655.(11) Sun, Y. K.; Chen, Z. H.; Noh, H. J.; Lee, D. J.; Jung, H. G.; Ren,Y.; Wang, S.; Yoon, C. S.; Myung, S. T.; Amine, K. NanostructuredHigh-Energy Cathode Materials for Advanced Lithium Batteries. Nat.Mater. 2012, 11, 942−947.(12) Dunn, B.; Kamath, H.; Tarascon, J. M. Electrical Energy Storagefor the Grid: A Battery of Choices. Science 2011, 334, 928−935.(13) Li, C. Z.; Chueh, C. C.; Yip, H. L.; Ding, F. Z.; Li, X. S.; Jen, A.K. Y. Solution-Processible Highly Conducting Fullerenes. Adv. Mater.2013, 25, 2457−2461.(14) Hartschuh, A.; Pedrosa, H. N.; Peterson, J.; Huang, L.; Anger,P.; Qian, H.; Meixner, A. J.; Steiner, M.; Novotny, L.; Krauss, T. D.Single Carbon Nanotube Optical Spectroscopy. ChemPhysChem 2005,6, 577−582.(15) Kim, J.; Lee, S. W.; Carlton, C.; Shao-Horn, Y. Pt-CoveredMultiwall Carbon Nanotubes for Oxygen Reduction in Fuel CellApplications. J. Phys. Chem. Lett. 2011, 2, 1332−1336.(16) Chen, X. Y.; Zhu, H. L.; Chen, Y. C.; Shang, Y. Y.; Cao, A. Y.;Hu, L. B.; Rubloff, G. W. Mwcnt/V2O5 Core/Shell Sponge for High

Areal Capacity and Power Density Li-Ion Cathodes. ACS Nano 2012,6, 7948−7955.(17) Ueta, A.; Tanimura, Y.; Prezhdo, O. V. Distinct Infrared SpectralSignatures of the 1,2-and 1,4-Fluorinated Single-Walled CarbonNanotubes: A Molecular Dynamics Study. J. Phys. Chem. Lett. 2010,1, 1307−1311.(18) Chaban, V. V.; Prezhdo, V. V.; Prezhdo, O. V. Confinement byCarbon Nanotubes Drastically Alters the Boiling and Critical Behaviorof Water Droplets. ACS Nano 2012, 6, 2766−2773.(19) Evanoff, K.; Benson, J.; Schauer, M.; Kovalenko, I.; Lashmore,D.; Ready, W. J.; Yushin, G. Ultra Strong Silicon-Coated CarbonNanotube Nonwoven Fabric as a Multifunctional Lithium-Ion BatteryAnode. ACS Nano 2012, 6, 9837−9845.(20) de Juan, A.; Pouillon, Y.; Ruiz-Gonzalez, L.; Torres-Pardo, A.;Casado, S.; Martin, N.; Rubio, A.; Perez, E. M. MechanicallyInterlocked Single-Wall Carbon Nanotubes. Angew. Chem., Int. Ed.2014, 53, 5394−5400.(21) Kalugin, O. N.; Chaban, V. V.; Loskutov, V. V.; Prezhdo, O. V.Uniform Diffusion of Acetonitrile Inside Carbon Nanotubes FavorsSupercapacitor Performance. Nano Lett. 2008, 8, 2126−2130.(22) Landi, B. J.; Ganter, M. J.; Cress, C. D.; DiLeo, R. A.; Raffaelle,R. P. Carbon Nanotubes for Lithium Ion Batteries. Energy Environ. Sci.2009, 2, 638−654.(23) Kar, T.; Pattanayak, J.; Scheiner, S. Insertion of Lithium Ionsinto Carbon Nanotubes: An Ab Initio Study. J. Phys. Chem. A 2001,105, 10397−10403.(24) Li, J.; Li, H. M.; Liang, X. Q.; Zhang, S. O.; Zhao, T.; Xia, D. G.;Wu, Z. Y. First Principles Study on the Diffusion of Alkali-Metal Ionson the Armchair Single-Wall Nanotubes. J. Phys. Chem. A 2009, 113,791−796.(25) Song, B.; Yang, J. W.; Zhao, J. J.; Fang, H. P. Intercalation andDiffusion of Lithium Ions in a Carbon Nanotube Bundle by Ab InitioMolecular Dynamics Simulations. Energy Environ. Sci. 2011, 4, 1379−1384.(26) Quinonero, D.; Garau, C.; Frontera, A.; Ballester, P.; Costa, A.;Deya, P. M. Structure and Binding Energy of Anion-Pi and Cation-PiComplexes: A Comparison of MP2, RI-MP2, DFT, and DF-DFTMethods. J. Phys. Chem. A 2005, 109, 4632−4637.(27) Kawasaki, S.; Hara, T.; Iwai, Y.; Suzuki, Y. Metallic andSemiconducting Single-Walled Carbon Nanotubes as the AnodeMaterial of Li Ion Secondary Battery. Mater. Lett. 2008, 62, 2917−2920.(28) Petaccia, L.; Goldoni, A.; Lizzit, S.; Larciprete, R. ElectronicProperties of Clean and Li-Doped Single-Walled Carbon Nanotubes. J.Electron Spectrosc. Relat. Phenom. 2005, 144, 793−797.(29) Zhong, J.; Song, L.; Chiou, J. W.; Dong, C. L.; Liang, X. Q.;Chen, D. L.; Xie, S. S.; Pong, W. F.; Chang, C. L.; Guo, J. H.; et al.Electronic Structure Study of Li+/OH− Modified Single-WalledCarbon Nanotubes by Soft-X-ray Absorption and Resonant EmissionSpectroscopy. Appl. Phys. Lett. 2010, 96, 213112.(30) Nirmalraj, P. N.; Boland, J. J. Selective Tuning and Optimizationof the Contacts to Metallic and Semiconducting Single-Walled CarbonNanotubes. ACS Nano 2010, 4, 3801−3806.(31) Yarotski, D. A.; Kilina, S. V.; Talin, A. A.; Tretiak, S.; Prezhdo,O. V.; Balatsky, A. V.; Taylor, A. J. Scanning Tunneling Microscopy ofDNA-Wrapped Carbon Nanotubes. Nano Lett. 2009, 9, 12−17.(32) Feng, M.; Zhao, J.; Petek, H. Atomlike, Hollow-Core-BoundMolecular Orbitals of C-60. Science 2008, 320, 359−362.(33) TubeASP (http://k.1asphost.com/tubeasp/tubeasp.asp), Veiga,R. G. A., Universidade Federal de Uberlandia, Brazil, and Tomanek,D., MSU, 2007.

The Journal of Physical Chemistry Letters Letter

dx.doi.org/10.1021/jz502175e | J. Phys. Chem. Lett. 2014, 5, 4129−41334133