Nanoscale

PAPER

Cite this: Nanoscale, 2016, 8, 6014

Received 7th January 2016,Accepted 17th February 2016

DOI: 10.1039/c6nr00138f

www.rsc.org/nanoscale

Pressure-driven opening of carbon nanotubes

Vitaly V. Chaban*a and Oleg V. Prezhdob

The closing and opening of carbon nanotubes (CNTs) is essential for their applications in nanoscale

chemistry and biology. We report reactive molecular dynamics simulations of CNT opening triggered by

internal pressure of encapsulated gas molecules. Confined argon generates 4000 bars of pressure inside

capped CNT and lowers the opening temperature by 200 K. Chemical interactions greatly enhance the

efficiency of CNT opening: fluorine-filled CNTs open by fluorination of carbon bonds at temperature and

pressure that are 700 K and 1000 bar lower than for argon-filled CNTs. Moreover, pressure induced CNT

opening by confined gases leaves the CNT cylinders intact and removes only the fullerene caps, while the

empty CNT decomposes completely. In practice, the increase in pressure can be achieved by near-infrared

light, which penetrates through water and biological tissues and is absorbed by CNTs, resulting in rapid local

heating. Spanning over a thousand of bars and Kelvin, the reactive and non-reactive scenarios of CNT

opening represent extreme cases and allow for a broad experimental control over properties of the CNT

interior and release conditions of the confined species. The detailed insights into the thermodynamic con-

ditions and chemical mechanisms of the pressure-induced CNT opening provide practical guidelines for the

development of novel nanoreactors, catalysts, photo-catalysts, imaging labels and drug delivery vehicles.

Introduction

Carbon nanotubes (CNTs) are tubular cylinders of carbonatoms exhibiting unusual strength, elasticity, electrical con-ductivity, thermal conductivity and other unique properties.1–5

CNTs find applications in electronics, optics, catalysis, energystorage, imaging, sensing, medicine, etc.6–16 The number ofresearch papers dedicated to CNTs, their derivatives, and CNTcontaining composite materials continues to grow exponen-tially, despite 24 years since the first publication.

CNTs are employed in medical treatments and drug deliv-ery, owing to their ability to penetrate lipid bilayers, reachcellular nucleus, and encapsulate drug molecules, DNA frag-ments and other biologically relevant guest species.14,17–26 Inaddition, CNTs absorb near-infrared light that penetratesthrough water and biological tissues, allowing one to activatetissue imbedded CNTs remotely and locally by a laser, formingthe basis for cancer treatments. Chemical reactions performedinside CNT nanoreactors exhibit different thermodynamics,kinetics and product yields, compared to unconfinedsystems.27–29 The CNT inner space creates a possibility forasymmetric catalysis and production of different isomers start-ing from the same reactants and depending on the CNT dia-

meter. This feature opens a wide avenue to tune selectivity andadjust catalytic conditions.30,31 Similarly, photo-catalystsexhibit different and often enhanced performance insideCNTs.32 In the course of the above processes, a cargo or reac-tant molecule has to be sealed inside a CNT and then releasedfrom the CNT, when required. It is particularly important inthis context to establish and understand the features andtrends associated with the sealing and opening of carbonnanoscale pores.

Recent chemical literature contains multiple success storiesof delivering biologically relevant molecules to target cells.Even though the toxicity issues of CNTs are not fully resolvedyet, rapid development of different CNT containing drug deliv-ery systems is continuously reported. Rusling and coworkers17

developed the first targeted killing of cancer cells in vivo usinga drug–CNT bioconjugate. The first line anticancer agent cispla-tin and an epidermal growth factor were attached to single-walled CNTs to deliberately target squamous cancer cells.According to the authors, the efficacy of their drug deliverysystem is better than that of the non-targeted bioconjugates. Fanand coworkers18 prepared a CNT–graphene nanosheet–ironoxide nanoparticle hybrid to perform efficient loading and deliv-ery of anticancer drugs. This hybrid exhibits superparamagneticproperties and strongly binds 5-fluorouracil with an unexpect-edly high load capacity. The authors conclude that the preparedsystem will find interesting applications in drug delivery.

Recently, Giri and coworkers19 contributed a system for asustained release of hydrophobic drugs, such as diclofenac

aInstituto de Ciência e Tecnologia, Universidade Federal de São Paulo, 12231-280,

São José dos Campos, SP, Brazil. E-mail: vvchaban@gmail.combDepartment of Chemistry, University of Southern California, Los Angeles, CA 90089,

USA

6014 | Nanoscale, 2016, 8, 6014–6020 This journal is © The Royal Society of Chemistry 2016

Publ

ishe

d on

19

Febr

uary

201

6. D

ownl

oade

d by

Uni

vers

ity o

f So

uthe

rn C

alif

orni

a on

11/

8/20

19 1

2:13

:24

AM

.

View Article OnlineView Journal | View Issue

sodium. The new system uses acrylic acid grafted guar gum-carboxy functionalized multi-walled CNTs in in situ compositemembranes. The CNTs increased hydrophobicity of all acrylicacid grafted membranes through an interaction with thematrix. Transdermal drug delivery was achieved by Gharib andcoworkers21 using a CNT–polyimide composite material. Thismaterial was used to fabricate hollow microneedles. Patternedbundles of CNTs were used as a porous scaffold to define thegeometry of the microneedle. CNT–dendrimer hybrid nano-materials were utilized by Rashidian and coworkers.33 Theresulting drug delivery systems readily penetrate cell mem-branes, allow high drug loading and target tumor cells bymeans of a magnetic field assigned to γ-F2O3 nanoparticles.Chaban and Prezhdo34 combined CNT efficacy in photo-induced cancer treatment and drug delivery, proposing aunified protocol for focused delivery and release of drugmolecules.

Liu and coworkers35 contributed a detailed and insightfulstudy of the thermal decomposition of nitromethane on func-tionalized graphene sheets, which were obtained by introdu-cing oxygen containing groups. The research wasaccomplished by ab initio molecular dynamics. It was shownthat carbon vacancy defects within the plane of the functiona-lized graphene greatly accelerate thermal decomposition ofnitromethane and its derivatives. The graphene sheetsremained intact during the simulated reactions. Zhu and co-workers36 explored the microscopic dynamic response anddecomposition mechanism of β-octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocin (an important explosive) under shock con-ditions using tight-binding density functional theory. The reac-tion of the explosive was initiated by fission of the molecularring at the carbon–nitrogen bond located in the vicinity of themajor axis. Subsequently, the carbon–hydrogen bonds andnitrogen–nitrogen bonds were destroyed. The authors under-lined that the simulated dynamic response behavior dependedon the angle between the chemical bond and the shockdirection.

Chemical reactions inside CNTs were exemplified by Eliseevand coworkers,28 who proposed a simple two-step approachfor an assured filling of CNT channels with AIIBVI and AIVBVI

one-dimensional chalcogenide nanocrystals. The proposedmethod is based on an intermediate metal iodide and amolten medium technique, with a subsequent chalcogenationinside CNT. Khlobystov and coworkers27 studied reactions ofthe inner surface of CNTs and nanoprotrusion processes.Importantly, in contrast to the original belief, they proved thatthe inner surface of CNT is reactive in the presence of catalyti-cally active rhenium atoms. According to Smith and co-workers,29 CNTs can dictate stereo- and regio-selectivity ofDiels–Alder reactions. Molecular dynamics simulationsdemonstrated that the 1,4-exo adduct is produced instead ofthe expected 9,10-adduct. It was noted that the reactant sizeand the CNT radius play major roles in the determination ofthe final product structure. A comprehensive review of theeffects of confinement inside CNTs on chemical catalysis canbe found in ref. 37.

Motivated by the broad interest to obtain efficient androbust CNT opening on demand, this work reports a theore-tical investigation of pressure- and chemistry-driven CNTdecomposition as initiated, for instance, by laser-inducedheating. We simulate CNT opening in real-time using a reac-tive atomistically-precise interaction model and varying thenature of the filler. The considered scenarios cover a broadrange of local temperature and pressure, demonstrating thepossibility of extensive experimental control over CNT interiorproperties and conditions for the release of confined species.

Methodology

The simulations were performed using reactive moleculardynamics (RMD) with the force field obtained from van Duinand coworkers.38,39 RMD differs from classical MD, since itemploys much more sophisticated intra- and inter-molecularpotentials, which aim to reproduce a realistic potential energysurface for a given system. In particular, RMD allows chemicalbonds to be broken and formed, while classical MD typicallydoes not. RMD is different from quantum chemical methods,since it does not employ time-consuming iterative proceduresand is generally much less computationally demanding. Alinear scaling of performance can be theoretically achieved inRMD much easier than with any quantum chemical method.The RMD simulations were carried out in ADF2013.

CNT (10,10) with a length of ca. 13.5 Å was considered as amodel of a single-walled nanotube with an ideal geometry.Three primary RMD systems involving CNT (10,10) werestudied, including an empty CNT, and argon(Ar)-filled CNT,and a fluorine(F2)-filled CNT. The RMD simulations wererepeated at a number of temperatures ranging from 2000 to4800 K, for 2000 ps each. An integration time-step of 0.1 fs wasselected to suitably reproduce high-frequency bond vibrationsupon decomposition at high temperatures. The Andersenthermostat was employed to control constant temperature witha relaxation time constant of 0.1 ps.

Results and discussion

Fig. 1 depicts empty and gas-filled CNT (10,10). The CNT struc-ture in all RMD systems is stable, although it is somewhatdeformed due to thermal motion under room conditions. Notethat argon and molecular fluorine are gases at room tempera-ture. Visual examination of the CNTs indicates that deviationsfrom the ideal CNT geometry are more significant in the filledCNTs. This observation originates from local pressures gene-rated by the gases inside the filled CNTs, whereas the internaland external pressures are equal in the case of the empty CNT.In the following, we will demonstrate that the internal pressureundermines CNT stability and that this effect can be quanti-fied from the reactive simulations.

The constructed systems (Table 1) were equilibrated at298 K using 50 ps long MD runs. The used equilibration times

Nanoscale Paper

This journal is © The Royal Society of Chemistry 2016 Nanoscale, 2016, 8, 6014–6020 | 6015

Publ

ishe

d on

19

Febr

uary

201

6. D

ownl

oade

d by

Uni

vers

ity o

f So

uthe

rn C

alif

orni

a on

11/

8/20

19 1

2:13

:24

AM

. View Article Online

are sufficient, since the fillers retain gas-like structures underconfinement, whereas the CNT structure is relaxed based onthe intra-molecular potentials, whose corresponding frequen-cies are below 1.0 fs. Thus, the RMD systems obtain theirglobal free energy configurations quickly. Afterwards, the MDsystems were heated up to 2000–4800 K and their stabilitieswere investigated at a number of temperatures.

It should be noted that the decomposition temperatures ofideal CNTs exceed the experimentally observed values, ca.1000 K,40–42 significantly. Ideal CNTs are extremely strong andelastic, as is known from the respective mechanical properties.However, structural defects, which are omnipresent in com-mercial CNTs and are a direct consequence of the chosen pro-duction method, decrease their thermal stability. Furthermore,the armchair CNT (10,10) is narrow with a diameter of ca.1 nm, while experimentally identified CNTs reach up to 50 nmin diameter. Wide single-walled CNTs are typically less stablethan narrow CNTs. Oxidants in the atmosphere favor low-temperature oxidation reactions, impacting stability of theentire nanostructure.

Fig. 2 depicts the evolution of the selected energy terms inthe case of the argon-filled CNT (10,10) at 4200 K. Other RMDsystems exhibit qualitatively similar trends. All intra-molecularenergy components increase in the course of the RMD simu-lation, since the systems lose their structures (increaseentropy) at high temperatures. Decrease of the van der Waalsenergy is due to the formation of multiple separate particles,

Fig. 1 Atomic configurations of the simulated systems at room temp-erature: (a) empty CNT (10,10); (b) CNT filled by argon atoms; (c) CNTfilled by molecular fluorine, F2.

Table 1 The times of CNT decomposition start and CNT decompo-sition end at different temperatures and with different fillers. CNTdecomposition is considered to begin when the number of molecules inthe system changes for the first time. CNT decomposition is consideredto finish when the number of molecules in the RMD system stabilizesunder given conditions

System

Start ofdecomposition,#iterations per 103

End ofdecomposition,#iterations per 103

Simulatedtemperature, K

Empty CNT(10,10)

— — 298— — 4000— — 4200600 1800 4400220 1600 4600100 370 4800

CNT (10,10)+ 50 Ar

— — 298— — 4000400 2200 4200100 1100 4400180 730 4600120 350 4800

CNT (10,10)+ 50 F2

— — 2000— — 2500— — 3000315 1750 3500200 750 3800

Fig. 2 Evolution of energy terms upon explosion of CNT (10,10) filledwith 50 argon atoms: (a) conjugation energy between carbon atoms; (b)carbon–carbon covalent energy; (c) van der Waals energy for bondedand non-bonded C–C, Ar–Ar, and C–Ar interactions; (d) carbon–carbon–carbon valence angle energy.

Paper Nanoscale

6016 | Nanoscale, 2016, 8, 6014–6020 This journal is © The Royal Society of Chemistry 2016

Publ

ishe

d on

19

Febr

uary

201

6. D

ownl

oade

d by

Uni

vers

ity o

f So

uthe

rn C

alif

orni

a on

11/

8/20

19 1

2:13

:24

AM

. View Article Online

which can approach one another more efficiently, as comparedto the static CNT structure. The decomposition of CNT, as wellas any other thermal decomposition, is entropically favorable.

Destruction of the argon-filled CNT occurs at 200 K lowerthan the destruction of the empty CNT, which occurs at4400 K. This is an effect of the local internal pressure. Argon isan example of a gas that does not interact chemically withCNTs. Thus, the pressure effect can be separated. We believethat the observed 200 K decrease in the decomposition temp-erature of the ideal CNT is a practically interesting pheno-

menon. Being corrected with respect to CNT natural defects, itoffers a mechanism to release encapsulated agents from CNTs.

Chemical interactions greatly enhance the efficiency of theCNT opening; see the fluorine-filled CNT case (Table 1). Here,the temperature is decreased by 900 K compared to the emptyCNT. Molecular fluorine is a strong oxidant, especially at hightemperatures. Thus, in addition to the internal pressure, thedecomposition is fostered by fluorination and, consequently,partial destruction of aromaticity. This CNT opening mechan-ism is another option to remove CNT caps. Opening frominside is different from the conventional opening solutions,which involve various oxidation agents. Our results are inqualitative agreement with those of Khlobystov and co-workers.27 For certain applications, e.g. in vivo, aggressive oxi-dants cannot be safely supplied. In those cases,decomposition from inside is particularly promising. The oxi-dants can be encapsulated inside the CNT. When needed, thereaction is initiated and the tube loses its integrity. The pro-ducts of this reaction are carbon fragments functionalizedwith the reactants, e.g. fluorinated carbon. These productsexhibit lower toxicity than the original reactants. The simplestinitiation pathway, as exemplified in the present work, is atemperature increase, which can be achieved by laserheating.43,44

The remainder of the work characterizes decompositionpathways and provides certain quantitative descriptors. Fig. 3shows the local internal pressures inside CNT (10,10). Thepressure vs. temperature is systematically higher in the case of

Fig. 3 Pressure generated inside CNT (10,10) by confined fluorinemolecules (red solid line) and by confined argon atoms (green dashedline) at elevated temperatures.

Fig. 4 Mechanisms of CNT (10,10) opening: (a) empty CNT at T = 4400 K; (b) argon-filled CNT at T = 4200 K; (c) fluorine-filled CNT at T = 3500 K.Opening of CNT (10,10) system starts from the region of fullerene cap in all systems, although the mechanism depends on the system. Empty CNTdecomposes when the energy of thermal vibrations, kT, exceeds the potential energy of carbon–carbon covalent bonds. When bonds break, carbonatoms form long unsaturated chains. The argon-filled CNT opens at 200 K lower than the empty CNT due to additional internal pressure of the gas(Fig. 3). The fullerene cap is removed and argon atoms leave the tube. After argon atoms escape and the pressure difference vanishes, the CNTdecomposition stops. Opening of fluorine-filled CNT (10,10) takes place at a significantly lower temperature (3500 K). Fluorine atoms attach to theCNT sidewalls and cap, and promote chemical decomposition.

Nanoscale Paper

This journal is © The Royal Society of Chemistry 2016 Nanoscale, 2016, 8, 6014–6020 | 6017

Publ

ishe

d on

19

Febr

uary

201

6. D

ownl

oade

d by

Uni

vers

ity o

f So

uthe

rn C

alif

orni

a on

11/

8/20

19 1

2:13

:24

AM

. View Article Online

argon. This is in accordance with our expectations, since argonatoms interact weakly with one another, deviate less from theideal gas behavior, and therefore, are more mobile. Addition-ally, a certain number of fluorine molecules are involved in thereaction with CNT, which decreases pressure further. Theobserved pressures range from 3400 to 4600 bar. The pressurevs. temperature dependence is linear.

The CNT opening reactions exhibit some similarities andsome differences (Fig. 4). All CNTs start decomposition fromthe CNT cap. This is in line with general knowledge suggestingthat the cap is less stable than the CNT body. A high tempera-ture, 4400 K, is required to initiate breakage of the firstcarbon–carbon bonds. Afterwards, the nanotube unknits rela-tively quickly.

Selected geometrical features of the CNTs near thedecomposition temperature are given in Fig. 5 and 6. At 298 K,the C–C bond length in the cap amounts to 1.44 Å with thestandard deviation of 0.02 Å. These values correspond to agood thermal stability of the capped CNT (10,10). Thermalexpansion occurs at 4000 K. The C–C lengths in the sidewallbecome 1.46 Å, whereas the C–C lengths in the cap become1.45 Å. In turn, the C–C bonds in the five-membered rings ofthe cap are 1.48 Å. These distances constitute a good illus-

tration that pressure-driven CNT decomposition initially startsfrom the five-membered rings of the CNT cap, whose bondlengths fluctuate by ±0.08 Å. The C–C–C angle in the pentagonis 107° before breakage.

Somewhat unexpectedly, the diameter of the empty CNTdecreases upon heating, compare 13.70 Å at 298 K to 13.58 Åat 4200 K (Fig. 6). These diameters were defined as distancesbetween the opposite carbon atoms, averaged over time. Inturn, the diameter of the argon-filled CNT at 4200 K is 13.78 Å.That is, the CNT expands only when it contains an encapsu-lated gas. Furthermore, the gaseous filler quenches the magni-tude of radial fluctuations by a factor of two: 0.5 Å vs. 0.9 Å(T = 4200 K). Nevertheless, the reduced fluctuation does notmake the argon-filled CNTs more stable than the empty CNT(Table 1).

Conclusions

To recapitulate, we demonstrated an extensive control over themechanism, temperature, pressure and products of CNTopening for applications as nanoreactors, catalysts, imagingand drug delivery vehicles. Reactive simulations of empty andfilled CNTs showed that pressure of a confined gas can be uti-lized to remove CNT caps in a predictable manner. The

Fig. 5 Representative (a) covalent C–C bond lengths and (b) valenceC–C–C angles in the argon-filled CNT (10,10) vs. time near thedecomposition temperature, Table 1. The bonds and angles from theCNT sidewall are red solid lines; the bonds and angles in the six-mem-bered rings of the CNT cap are green dashed lines; the bonds andangles in the five-membered ring of the fullerene cap are blue dash-dotted lines. The inset in (a) depicts evolution of a carbon–carbon bondlength at 298 K for comparison.

Fig. 6 Evolution of the (a) length and (b) diameter of CNT (10,10):empty CNT at 298 K (red solid line), empty CNT at 4000 K (dashed greenline), argon-filled CNT at 4000 K (blue dash-dotted line).

Paper Nanoscale

6018 | Nanoscale, 2016, 8, 6014–6020 This journal is © The Royal Society of Chemistry 2016

Publ

ishe

d on

19

Febr

uary

201

6. D

ownl

oade

d by

Uni

vers

ity o

f So

uthe

rn C

alif

orni

a on

11/

8/20

19 1

2:13

:24

AM

. View Article Online

decomposition temperature of the argon-filled CNT (10,10) isca. 200 K smaller than that of the empty CNT (10,10). Chemi-cal interactions greatly facilitate CNT opening. When CNT(10,10) was filled with an oxidant gas, i.e., molecular fluorine,the CNT decomposition temperature decreased by ca. 900 Kdue to the fluorination of the CNT caps, ultimately leading tofluorinated hydrocarbon products. In certain applications, e.g.in vivo, aggressive oxidants cannot be supplied from theoutside, and decomposition from inside becomes particularlypromising. Importantly, pressure induced CNT openingremoved only CNT caps, leaving the CNT cylinders intact. Incontrast, thermal decomposition of the empty CNT destroyedthe entire tube. The uncovered details of the CNT decompo-sition mechanisms create a broad range of possibilities fortuning of CNT thermal stability, internal pressure, temperatureand opening protocol.

We employed perfect CNT models. Commercially availableCNTs are known to be less stable due to defects, dopants, sur-factants and other species. In such cases, decomposition startsfrom the defects, and the reported decomposition tempera-tures will have to be corrected for each practical application inmind. The pressure and temperature control over the CNTopening should be possible in realistic systems as well.

The reported results are important for a broad range ofapplications, in which CNT opening on demand is necessary.Such applications include reactions within a nanoscale con-finement and targeted drug delivery. Local heating by a lasercan be employed to generate the correct temperature. Laserheating is particularly facile with CNTs, since they absorbnear-infrared light that penetrates through water and biologi-cal tissues.

Acknowledgements

V. V. C. is funded through CAPES. O. V. P. acknowledgessupport from the US Department of Energy (Grant No.DE-SC0014429).

References

1 J. T. Hu, T. W. Odom and C. M. Lieber, Chemistry andphysics in one dimension: Synthesis and properties of nano-wires and nanotubes, Acc. Chem. Res., 1999, 32, 435–445.

2 D. Tasis, N. Tagmatarchis, A. Bianco and M. Prato, Chem-istry of carbon nanotubes, Chem. Rev., 2006, 106, 1105–1136.

3 Q. Tang, Z. Zhou and Z. F. Chen, Graphene-related nano-materials: tuning properties by functionalization, Nano-scale, 2013, 5, 4541–4583.

4 A. Nitzan and M. A. Ratner, Electron transport in molecularwire junctions, Science, 2003, 300, 1384–1389.

5 M. Pumera and Y. Miyahara, What amount of metallicimpurities in carbon nanotubes is small enough notto dominate their redox properties?, Nanoscale, 2009, 1,260–265.

6 R. H. Baughman, A. A. Zakhidov and W. A. de Heer,Carbon nanotubes - the route toward applications, Science,2002, 297, 787–792.

7 M. J. Zhi, C. C. Xiang, J. T. Li, M. Li and N. Q. Wu, Nano-structured carbon-metal oxide composite electrodes forsupercapacitors: a review, Nanoscale, 2013, 5, 72–88.

8 A. S. Arico, P. Bruce, B. Scrosati, J. M. Tarascon and W. Van-Schalkwijk, Nanostructured materials for advanced energyconversion and storage devices, Nat. Mater., 2005, 4, 366–377.

9 P. Avouris, Z. H. Chen and V. Perebeinos, Carbon-basedelectronics, Nat. Nanotechnol., 2007, 2, 605–615.

10 M. A. Shannon, P. W. Bohn, M. Elimelech, J. G. Georgiadis,B. J. Marinas and A. M. Mayes, Science and technology forwater purification in the coming decades, Nature, 2008,452, 301–310.

11 S. Park, M. Vosguerichian and Z. A. Bao, A review of fabri-cation and applications of carbon nanotube film-basedflexible electronics, Nanoscale, 2013, 5, 1727–1752.

12 P. Jena, Materials for Hydrogen Storage: Past, Present, andFuture, J. Phys. Chem. Lett., 2011, 2, 206–211.

13 L. L. Zhang and X. S. Zhao, Carbon-based materials assupercapacitor electrodes, Chem. Soc. Rev., 2009, 38, 2520–2531.

14 K. Y. Choi, G. Liu, S. Lee and X. Y. Chen, Theranostic nano-platforms for simultaneous cancer imaging and therapy:current approaches and future perspectives, Nanoscale,2012, 4, 330–342.

15 D. S. Yu and L. M. Dai, Self-Assembled Graphene/CarbonNanotube Hybrid Films for Supercapacitors, J. Phys. Chem.Lett., 2010, 1, 467–470.

16 A. Kaniyoor, R. I. Jafri, T. Arockiadoss and S. Ramaprabhu,Nanostructured Pt decorated graphene and multi walledcarbon nanotube based room temperature hydrogen gassensor, Nanoscale, 2009, 1, 382–386.

17 A. A. Bhirde, V. Patel, J. Gavard, G. F. Zhang, A. A. Sousa,A. Masedunskas, R. D. Leapman, R. Weigert, J. S. Gutkindand J. F. Rusling, Targeted Killing of Cancer Cells in Vivoand in Vitro with EGF-Directed Carbon Nanotube-BasedDrug Delivery, ACS Nano, 2009, 3, 307–316.

18 X. J. Fan, G. Z. Jiao, L. Gao, P. F. Jin and X. Li, The prepa-ration and drug delivery of a graphene-carbon nanotube-Fe3O4 nanoparticle hybrid, J. Mater. Chem. B, 2013, 1,2658–2664.

19 A. Giri, T. Bhunia, L. Goswami, A. B. Panda andA. Bandyopadhyay, Fabrication of acrylic acid grafted guargum-multiwalled carbon nanotube hydrophobic mem-branes for transdermal drug delivery, RSC Adv., 2015, 5,41736–41744.

20 L. W. Zhang, P. F. Rong, M. L. Chen, S. Gao and L. Zhu,A novel single walled carbon nanotube (SWCNT)functionalization agent facilitating in vivo combinedchemo/thermo therapy, Nanoscale, 2015, 7, 16204–16213.

21 B. J. Lyon, A. I. Aria and M. Gharib, Fabrication of carbonnanotube-polyimide composite hollow microneedles fortransdermal drug delivery, Biomed. Microdevices, 2014, 16,879–886.

Nanoscale Paper

This journal is © The Royal Society of Chemistry 2016 Nanoscale, 2016, 8, 6014–6020 | 6019

Publ

ishe

d on

19

Febr

uary

201

6. D

ownl

oade

d by

Uni

vers

ity o

f So

uthe

rn C

alif

orni

a on

11/

8/20

19 1

2:13

:24

AM

. View Article Online

22 L. Minati, V. Antonini, M. Dalla Serra and G. Speranza,Multifunctional Branched Gold-Carbon Nanotube Hybridfor Cell Imaging and Drug Delivery, Langmuir, 2012, 28,15900–15906.

23 J. M. Tan, G. Karthivashan, P. Arulselvan, S. Fakurazi andM. Z. Hussein, Sustained Release and Cytotoxicity Evalu-ation of Carbon Nanotube-Mediated Drug Delivery Systemfor Betulinic Acid, J. Nanomater., 2014, 862148.

24 L. Muzi, C. Menard-Moyon, J. Russier, J. Li, C. F. Chin,W. H. Ang, G. Pastorin, G. Risuleo and A. Bianco, Dia-meter-dependent release of a cisplatin pro-drug from smalland large functionalized carbon nanotubes, Nanoscale,2015, 7, 5383–5394.

25 Y. H. Xiao, X. X. Ye, L. He and J. F. Che, New carbonnanotube-conducting polymer composite electrodes fordrug delivery applications, Polym. Int., 2012, 61, 190–196.

26 S. L. Yoong, W. L. Lau, A. Y. Liu, D. Prendergast, H. K. Ho,V. C. K. Yu, C. Lee, W. H. Ang and G. Pastorin, Mitochon-dria-acting hexokinase II peptides carried by short-lengthcarbon nanotubes with increased cellular uptake, endo-somal evasion, and enhanced bioactivity against cancercells, Nanoscale, 2015, 7, 13907–13917.

27 T. W. Chamberlain, J. C. Meyer, J. Biskupek, J. Leschner,A. Santana, N. A. Besley, E. Bichoutskaia, U. Kaiser andA. N. Khlobystov, Reactions of the inner surface of carbonnanotubes and nanoprotrusion processes imaged at theatomic scale, Nat. Chem., 2011, 3, 732–737.

28 A. A. Eliseev, M. V. Chernysheva, N. I. Verbitskii,E. A. Kiseleva, A. V. Lukashin, Y. D. Tretyakov, N. A. Kiselev,O. M. Zhigalina, R. M. Zakalyukin, A. L. Vasiliev, et al.,Chemical Reactions within Single-Walled Carbon Nano-tube Channels, Chem. Mater., 2009, 21, 5001–5003.

29 N. M. Smith, K. S. Iyer and B. Corry, The confined spaceinside carbon nanotubes can dictate the stereo- and regio-selectivity of Diels–Alder reactions, Phys. Chem. Chem.Phys., 2014, 16, 6986–6989.

30 S. A. Miners, G. A. Rance and A. N. Khlobystov, Regio-selective control of aromatic halogenation reactions incarbon nanotube nanoreactors, Chem. Commun., 2013, 49,5586–5588.

31 K. Hashimoto, N. Kumagai and M. Shibasaki, A CarbonNanotube Confinement Strategy to Implement Homo-geneous Asymmetric Catalysis in the Solid Phase, Chem. –Eur. J., 2015, 21, 4262–4266.

32 X. F. Cui, Y. J. Wang, G. Y. Jiang, Z. Zhao, C. M. Xu,A. J. Duan, J. Liu, Y. C. Wei and W. K. Bai, The encapsu-lation of CdS in carbon nanotubes for stable andefficient photocatalysis, J. Mater. Chem. A, 2014, 2,20939–20946.

33 E. Mehdipoor, M. Adeli, M. Bavadi, P. Sasanpour andB. Rashidian, A possible anticancer drug delivery systembased on carbon nanotube-dendrimer hybrid nano-materials, J. Mater. Chem., 2011, 21, 15456–15463.

34 V. V. Chaban and O. V. Prezhdo, Water Boiling InsideCarbon Nanotubes: Toward Efficient Drug Release, ACSNano, 2011, 5, 5647–5655.

35 L. M. Liu, R. Car, A. Selloni, D. M. Dabbs, I. A. Aksay andR. A. Yetter, Enhanced Thermal Decomposition of Nitro-methane on Functionalized Graphene Sheets: Ab InitioMolecular Dynamics Simulations, J. Am. Chem. Soc., 2012,134, 19011–19016.

36 Z. H. He, J. Chen, G. F. Ji, L. M. Liu, W. J. Zhu and Q. Wu,Dynamic Responses and Initial Decomposition underShock Loading: A DFTB Calculation Combined with MSSTMethod for beta-HMX with Molecular Vacancy, J. Phys.Chem. B, 2015, 119, 10673–10681.

37 X. L. Pan and X. H. Bao, The Effects of Confinement insideCarbon Nanotubes on Catalysis, Acc. Chem. Res., 2011, 44,553–562.

38 E. C. Neyts, K. Ostrikov, Z. J. Han, S. Kumar, A. C. T. vanDuin and A. Bogaerts, Defect Healing and EnhancedNucleation of Carbon Nanotubes by Low-Energy Ion Bom-bardment, Phys. Rev. Lett., 2013, 110, 065501.

39 A. M. Kamat, A. C. T. van Duin and A. Yakovlev, MolecularDynamics Simulations of Laser-Induced Incandescence ofSoot Using an Extended ReaxFF Reactive Force Field,J. Phys. Chem. A, 2010, 114, 12561–12572.

40 M. C. L. de Oliveira, G. Ett and R. A. Antunes, Corrosionand thermal stability of multi-walled carbon nanotube-graphite-acrylonitrile-butadiene-styrene composite bipolarplates for polymer electrolyte membrane fuel cells, J. PowerSources, 2013, 221, 345–355.

41 A. M. F. Lima, A. W. Musumeci, H. W. Liu, E. R. Waclawikand G. G. Silva, Purity evaluation and influence of carbonnanotube on carbon nanotube/graphite thermal stability,J. Therm. Anal. Calorim., 2009, 97, 257–263.

42 X. Z. Xu, J. Zhou, L. Jiang, G. Lubineau, S. A. Payne andD. Gutschmidt, Lignin-based carbon fibers: Carbon nano-tube decoration and superior thermal stability, Carbon,2014, 80, 91–102.

43 T. R. Nelson, V. V. Chaban, O. N. Kalugin andO. V. Prezhdo, Vibrational Energy Transfer between CarbonNanotubes and Liquid Water: A Molecular Dynamics Study,J. Phys. Chem. B, 2010, 114, 4609–4614.

44 L. M. Maestro, P. Haro-Gonzalez, B. del Rosal, J. Ramiro,A. J. Caamano, E. Carrasco, A. Juarranz, F. Sanz-Rodriguez,J. G. Sole and D. Jaque, Heating efficiency of multi-walledcarbon nanotubes in the first and second biologicalwindows, Nanoscale, 2013, 5, 7882–7889.

Paper Nanoscale

6020 | Nanoscale, 2016, 8, 6014–6020 This journal is © The Royal Society of Chemistry 2016

Publ

ishe

d on

19

Febr

uary

201

6. D

ownl

oade

d by

Uni

vers

ity o

f So

uthe

rn C

alif

orni

a on

11/

8/20

19 1

2:13

:24

AM

. View Article Online