ARTICLE

Engineering Radial Deformations in Single-Walled Carbon and Boron Nitride Nanotubes Using Ultrathin NanomembranesMeng Zheng, Lian-feng Zou, Howard Wang, Cheol Park, ,§ and Changhong Ke ,* Department of Mechanical Engineering, State University of New York at Binghamton, Binghamton, New York 13902, United States, National Institute of Aerospace, 100 Exploration Way, Hampton, Virginia 23666, United States, and§Department of Mechanical and Aerospace Engineering, University of Virginia, Charlottesville, Virginia 22904, United States adial deformations of one-dimensional tubular nanostructures, such as carbon nanotubes (CNTs) and boronnitride nanotubes (BNNTs), are of great importance to their respective electrical properties and electronics applications (e.g., eld e ect transistors (FETs),1 singleelectron transistors (SETs),2 and optoelectronic devices3). From the structural point of view, CNTs and BNNTs are made of alike hexagonal sp2 covalent CÀC and partially ionic BÀN bonding networks, respectively. Both tubes have extraordinary mechanical and thermal properties. For instance, they have comparable Young's moduli, with reported values of up to 1.2À1.3 TPa.4À10 Their thermal conductivities are reportedly to be above 3000 W 3 mÀ1 3 KÀ1.11,12 However, the electrical properties of these two types of tubes are of distinct di erence. Single-walled CNTs (SWCNTs) are either metallic or semiconductive depending on the tube chirality, while BNNTs are excellent insulators with band-gaps of about 5À6 eV and being largely independent of the tube chirality.13À15 Both experiments and theoretical modeling demonstrate that the radial deformation of SWCNTs can greatly in uence their electronic structures (e.g., band-gaps) and result in semiconductorto-metal or metal-to-semiconductor transitions in reportedly reversible fashions.16À21 In addition, studies show that radial deformation can reduce the band gap of BNNTs from insulator to semiconductor or even conductor. 18,22 Tuning the band-gaps of these tubular nanostructures through engineering their transverse deformations will greatly impact many of their novel electronics applications.ZHENG ET AL .

R

ABSTRACT Radial deformations of car-

bon and boron-nitride nanotubes are of great importance to their respective electronic properties and applications. In this paper, we present a simple and practical approach of engineering radial deformations in singlewalled carbon and boron-nitride nanotubes (SWCNTs and SW-BNNTs) through covering individual nanotubes lying on at substrates with subnanometer-thick monolayer graphene oxide (GO) nanomembranes. The GO membrane conforms to and transversely compresses the underlying nanotube as a result of its adhesion binding interaction with the substrate. Our atomic force microscopy (AFM) imaging measurements reveal that the engineered net radial deformations of both types of tubes increase with the tube diameter and are more f

or SWBNNTs compared with SWCNTs of the same tube diameter. Our results capture the net crosssection height reductions of up to 44.1% for SW-BNNTs and up to 29.7% for SWCNTs. Our work clearly demonstrates the e ectiveness of our proposed approach for engineering and controlling the radial deformation in one-dimensional tubular nanostructures and opens a promising route for mechanical tuning of their electronic properties for novel nanoelectronics applications.KEYWORDS: boron nitride nanotubes . carbon nanotubes . graphene-oxide nanosheets . radial deformation . atomic force microscopy

Prior studies show that noticeable tube deformation occurs along its transverse direction when a carbon nanotube is deposited on a at substrate, which is due to the van der Waal (vdw) interaction between the nanotube and the substrate.23À25 However, such radial deformation is only substantial for nanotubes of very large diameters (e.g., 4À5 nm). For SWCNTs of relatively small diameters, none-to-little radial deformation occurs. For instance, MD simulations show that vdw adhesion-induced radial deformation for a (20, 20) SWCNT (2.70 nm in diameter) is about 13% of its original tubeVOL. 6’

* Address correspondence to cke@binghamton.edu. Received for review December 13, 2011 and accepted January 20, 2012. Published online January 26, 2012 10.1021/nn2048813C 2012 American Chemical Society

NO. 2

’

1814– 1822

’

2012

1814

ARTICLE

diameter, while merely 2% for a (10,10) tube of 1.34 nm in diameter.23 It is noted that substantial radial deformations, as higher as 50% of the original tube diameter,18,19 are desired for inducing signi cant band gap and electrical conductance changes in nanotubes. Signi cant radial deformations on individual tubes can be engineered using nanomanipulation with sharp scanning probe microscopy (SPM) probes.2,21,26 However, this method is a slow and sequential approach, which is only useful for a very small quantity of nanotubes and is not feasible for large-scale processing. Introducing signi cant radial deformations in nanotubes with scalability for large-scale integration is still not available. In this paper, we propose a simple and practical approach for engineering radial deformations in individual nanotubes, and demonstrate its e ectiveness through engineering radial deformations in SWCNTs and single-walled BNNT (SW-BNNTs). Our proposed approach is to cover or partially cover individual nanotubes lying on at substrates with subnanometer-thick nanomembranes, as illustrated in Figure 1a. Figure 1b shows schematically the cross sections of a nanotube with a height of h0 (left) and the membrane-covered nanotube (right) staying on a at substrate, respectively. The cross-section height of the membranecovered nanotube, h, is measured as the height di erence of the membrane from the position right on top of the underlying nanotube to the position of the at covering of the substrate. Th

e nanomembrane conforms to the underlying nanotube as a result of a balanced competition between the in-plane membrane rigidity and the adhesion interaction between the nanomembrane and the substrate/nanotube. Consequently, the nanomembrane imposes a compressive load to mechanically deform the underlying nanotube in its transverse direction. The radial deformation of the nanotube covered by the nanomembrane can be quanti ed by measuring parameters h and h0. We choose monolayer graphene oxide (GO) sheet as the nanomembrane in our study mainly for the following factors: (1) the monolayer GO sheet is～0.7 nm in thickness and readily conforms to the underlying nanotube in a relatively uniform fashion; (2) the GO sheet has a relatively high Young's modulus (EGO= 207 GPa27), which has the advantage of inducing large compressive loads on the underlying nanotube; (3) the GO sheet is water-soluble and can be processed and patterned by existing micro/nanofabrication techniques;28,29 (4) the GO sheet is electrically insulating30 and does not directly interfere with the electron transport in the underlying nanotube. In addition, if desired, the GO sheet can also be readily reduced to graphene.28,30,31 In this study, we investigate both the cross-section deformations of the GO sheet-covered individual SWCNTs and SW-BNNTs and the structural morphology of the covering monolayer GO sheetsZHENG ET AL .

Figure 1. (a) Schematic of a nanotube on a at substrate partially covered by a thin membrane: a is the length of the nanotube segment covered by the membrane. (b) Schematics of the cross sections of a nanotube (left) and a membrane-covered nanotube (right) on a at substrate. (c) Schematic drawings and representative HRTEM images of SW-BNNTs and SWCNTs. Both tubes in the drawings are of zigzag (14, 0) chirality, in which the red, blue, and black dots represent nitrogen, boron, and carbon atoms, respectively. The scale bars represent 2 nm.

using atomic force microscopy (AFM) and nonlinear continuum mechanics theories. The schematics and representative high resolution transmission electron microscopy (HRTEM) images of both types of tubes are presented in Figure 1c. Our AFM measurements reveal that the engineered radial deformations for both types of tubes increase with the tube diameter, and are more for SW-BNNTs compared with SWCNTs of the same tube diameter. Our results capture net radial height reductions of up to 44.1% for SW-BNNTs and up to 29.7% for SWCNTs, which are substantially higher compared with the radial height reduction ascribed to the vdw interaction between the nanotube and the substrate. The observed di erence in radial height reduction between SW-BNNTs and SWCNTs is ascribed to their di erence in e ective radial modulus.32 Our results also reveal that the structural morphology of the covering monolayer GO sheet is governed by both membrane bending and stretching e ects. The adhesion binding energy between monolayer GO sheets an

d at SiO2/Si substrates is found to be about 0.038 J/ m2. Our proposed approach of engineering radial deformations in nanotubes is compatible with existing micro/nano-fabrication techniques and thus suitableVOL. 6’

NO. 2

’

1814– 1822

’

2012

1815

ARTICLE

for large-scale processing and integration. Our work contributes directly to the study of engineering and controlling the radial deformation of one-dimensional tubular nanostructures, and opens a promising route for mechanical tuning of their electronic properties for novel nanoelectronic applications. RESULTS AND DISCUSSION AFM Measurements of GO Sheet-Covered Nanotubes. Figure 2a shows a representative tapping mode AFM image of one GO sheet deposited on a silicon substrate. The measured height of the GO sheet is 1.5 ( 0.15 nm, which is consistent with the data reported in the literature about the monolayer GO sheet.29,33,34 All the results and analysis presented in this paper are based on the monolayer GO sheets similar to the one shown in Figure 2a. Figure 2b shows one representative AFM image exhibiting three scenarios of GO sheet covered-CNTs, which are also observed for GO sheet-covered BNNTs. In this image, CNT#1 is partially covered by the GO sheet with only one tube end sticking out, CNT#2 is partially covered by the GO sheet in the central potion with both tube ends exposed, while CNT#3 is fully covered by the GO sheet. Both CNTs#1 and#2 are considered appropriate samples for our study because both h0 and h can be directly measured, while only h can be measured for CNT#3. The uncovered tube cross-section heights (h0) of CNTs#1 and#2 are measured to be 2.36 and 4.61 nm, respectively. Therefore, only CNT#1 may be a SWCNT (which is confirmed by our later analysis), while CNT#2 is most likely a multiwalled tube. The measured cross-section heights of CNT#1 in Figure 1b along its longitude direction are presented in the inset plot in Figure 2c, which displays a prominent transition around the edge of the GO sheet. The nanotube cross-section height varies on a very tight range (<0.5Å) for both the uncovered and the GO sheet-covered portions away from the transition position, which can be ascribed to the variation of the tube diameter, the roughness of the substrate, and the noise in the AFM measurements. The red and blue curves in Figure 2c show the respective cross-section profiles of the uncovered and the GO sheetcovered portions of CNT#1 shown in Figure 1b based on the AFM-measured height data marked in the red and blue boxes. The tube height for the covered portion of this nanotube is measured to be h= 1.72 nm. Therefore, the tube height reduction as a result of the monolayer GO sheet covering isΔh= 0.64 nm. Similar tube cross-section height reduction because of the GO sheet covering is also observed on GO sheet-covered BNNTs. The CNT height profiles presented in Figure 2c are also contrasted with the respective cross-section profiles

for one selected SW-BNNT of a similar original tube height. The measured cross-section height reduction for this BNNT isΔh= 0.74 nm, which is noticeably higher than that of the SWCNT of the same tube diameter. Such a differenceZHENG ET AL .

Figure 2. (a) AFM image of a monolayer graphene oxide (GO) sheet on a fresh silicon substrate. (b) AFM image of dispersed CNTs on a fresh silicon substrate covered or partially covered by a monolayer GO nanosheet. (c) Comparison of the respective cross-section pro les of the uncovered portion (red curve) and the GO-sheet-covered portion (blue curve) of CNT#1 shown in panel b. The pink and green curves are the respective cross-section pro les for one SW-BNNT of similar diameter to CNT#1 in panel b. The inset curve shows the variation of the measured tube height along the tube longitude axis for CNT#1 in panel b around the edge of the covering GO sheet.

in cross-section height reduction between GO-covered SW-BNNTs and SWCNTs are consistently observed on other measured samples. Figure 3a shows the dependences of the nanotube cross-section height reduction as a result of the monolayer GO sheet covering on the nanotube diameter based on measurements of 27 different SW-BNNTs and 15 different SWCNTs, whose original tube diameters are measured to be within 0.5 to 2.8 nm. The nanotube diameter here is defined as Dnt= h0Àtnt, in which tnt= 0.34 nm is the interlayer distance of both the graphene and BÀN sheets.35,36 The key experimentally measured and theoretically predicted parameters on six selected GO-covered SWCNTs and SW-BNNTs are listed in Table 1. Our results show that the measured radial height reduction for both types of tubes increases with the nanotube diameter. SW-BNNTs are found to consistently deform more in their transverse direction due to the GO sheet covering compared with SWCNTs of the same tube diameter. This result suggests that the radial rigidity of SW-BNNTs is lower compared with SWCNTs of the same diameter, which is consistent with our recently reported results on the e ective radial moduli of SW-BNNTs and SWCNTs.32 It is noted thatVOL. 6’

NO. 2

’

1814– 1822

’

2012

1816

ARTICLE

Figure 3. (a) The dependencies of the measured crosssection height reduction for SWCNTs and SW-BNNTs due to the covering of monolayer GO sheets on the nanotube diameter. (b) The respective per-unit-length compressive loads applied on SWCNTs and SW-BNNTs versus their tube diameters. (c) Comparison of the net radial deformation strain for SWCNTs and SW-BNNTs as a result of the covering of monolayer GO sheets. The solid lines are the linear- tting curves of the respective experimental data sets.

double-walled (or even triple-walled) CNTs and BNNTs may also have their diameters within the diameter range shown in Figure 3a. However, such tubes have much higher radial rigidity compared with singlewalled counterparts of the same outer diameter. Therefore, their radial he

ight reduction due to the GO sheetZHENG ET AL .

covering is expected to be substantially lower compared with single-walled tubes, which is con rmed by our measurements (see Figure S1 in the Supporting Materials). In addition, we performed minimal crosssection height measurements on selected nanotubes to determine their tube-layer numbers based on the fact that the cross-section heights of highly compressed single-walled CNTs or BNNTs will be close to 0.68 nm (i.e., twice of their interlayer distance values of 0.34 nm), while 1.36 nm for double-walled tubes and 2.04 nm for triple-walled tubes. The number of tube walls determined by our minimal cross-section height measurements is consistent with our analysis based on the data shown in Supporting Information, Figure S1. The detailed measurements of determining the number of tube layers in CNTs and BNNTs using this method will be reported in a separate publication.37 Therefore, all the data shown in Figure 3a represent only SWCNTs and SW-BNNTs. Radial Deformation of GO Sheet-Covered Nanotubes. Because the elastic modulus of GO sheets is much higher than the effective radial moduli of both SWCNTs and SW-BNNTs as well as the elastic modulus of the substrate (native oxide, 74 GPa), the deformation of the GO sheet along its thickness direction can be reasonably ignored. Therefore, the height reduction of the nanotube,Δh= h0À h, can be ascribed to the cross-section deformation of the nanotube and the deformation of the substrate. Since the AFM tipÀsample interaction was controlled in the attractive force regime in the AFM tapping mode scanning process, the gentle contact of the AFM tip with the nanotube is not expected to result in any height reduction of the nanotube. Here, we reasonably assume that the mechanical deformations of the nanotube and the substrate are within the elastic regime, and therefore can be modeled using wellestablished contact mechanics theories. In our contact mechanics model, the GO sheet-covered nanotube is simplified as an elastic cylindrical body in contact with a flat substrate under a uniformly distributed compressive line load P. In our data analysis, the nanotube is assumed to undertake a circular cross-section on the substrate before being covered by the GO sheet, thus neglecting the effect of vdw interactions between the tube and the substrate on the flattening of the tube cross-section.23 The combined nanotubeÀsubstrate deformation equals to the cross-section height reduction of the nanotubeΔh and is given by38 ! 2 2 1À vsub 1À vntΔh¼ PþπEsubπEnt 0 13 2 B C 6 B C7 6 7 2a2 B C7 6 ! C7 61þ logB B 1À v2 C7 6 2 1À vsub@ A5 4 subþ P 3 h0 3πEsubπEsubVOL. 6’

(1)

NO. 2

’

1814– 1822

’

2012

1817

ARTICLE

TABLE 1. List of the Key Experimentally Measured and Theoretically Predicted Parameters for Selected Monolayer GO Sheet-Covered SWCNT and SW-BNNT Samples L (nm) Dnt L (nm) L (nm)

2 (nm) a (nm)Δh (nm) P

(N/m)εrad nt (%) (theoretical-bending only) (theoretical-bendingþstretching) (experimental) M0 (nN) W at (J/m )

SWCNT#1 SWCNT#2 SWCNT#3 SW-BNNT#1 SW-BNNT#2 SW-BNNT#3

1.09 1.46 1.96 0.83 1.42 1.97

185 165 245 225 250 170

0.21 0.33 0.57 0.26 0.43 0.70

0.72 0.77 0.80 0.82 0.51 0.47

14.15 18.88 26.00 13.10 21.14 29.31

12.60 13.07 13.69 10.91 14.52 15.84

18.31 20.64 23.50 14.11 21.91 26.10

17.54 23.70 24.89 16.09 21.93 23.76

0.699 0.717 0.718 0.787 0.551 0.512

0.0400 0.0423 0.0423 0.0509 0.0249 0.0215

where Ent and Esub are the effective radial modulus of the nanotube and the Young's modulus of the substrate, respectively.νnt and vsub are the Poisson's ratios of the nanotube and the substrate, respectively. A native oxide layer of～2 nm in thickness was measured on the surface of the Si substrates using ellipsometry (Model Uvisel, Horiba, Ltd.). The materials for the substrate are considered to be native silicon oxide with an elastic modulus of 74 GPa and a Poisson's ratio of 0.16. Poisson's ratios for SWCNTs and SW-BNNTs are assumed to be 0.17 and 0.2,32 respectively. a is the length of the nanotube segment covered by the GO sheet, as illustrated in Figure 1a. The recently reported effective radial elastic moduli of both SW-BNNTs and SWCNTs32 can be reasonably approximated as simple power functions of the tube diameter (Dnt), given by 2.203 E nt= 13.428DÀ (for SW-BNNTs) and Ent= ntÀ1.931 26.602Dnt (for SWCNTs), in which Ent and Dnt are in units of GPa and nm, respectively. The compressive load P is calculated using eq 1 for all the measured nanotubes shown in Figure 3a, and the results are presented in Figure 3b. For SW-BNNTs, P is found to be within the range of 0.4À1.0 N/m, displaying a decreasing trend with the increase in the tube diameter, while for SWCNTs, the calculated P is about 0.7 N/m and exhibits no appreciable dependence on the tube diameter. The net cross-section deformation of the nanotubeδd can be estimated using eq 1 by considering the substrate as a rigid body. Figure 3c shows the dependence of the net nanotube rad radial deformation strainεrad nt, de ned asεnt=δd/dnt, on the tube diameter for both SW-BNNTs and SWCNTs. Our results reveal thatεrad nt increases, in nearly linear fashions, with the tube diameter for both types of tubes. SW-BNNTs are found to consistently deform more along their transverse directions than comparable SWCNTs, with the di erence of about 10À15% in radial deformation strain within the measured tube diameter range. For SWCNTs, the net radial deformation strain is found to be about 18.4% for a tube of 1.4 nm in diameter and 29.7% for a tube of 2.7 nm in diameter, both of which are signi cantly higher than the previously reported radial strain values ascribed to the tube-substrate vdw interaction.23 The net radial deformation strain reaches a remarkable value ofZHENG ET AL .

44.1% for a SW-BNNT of 2.6 nm in diameter. Our results clearly demonstrate that su

bstantial nanotube crosssection attening is produced as a result of covering individual nanotubes with monolayer GO sheets. We want to point out that the above-calculated net nanotube cross-section height reduction does not take into account the tube radial deformation induced by the vdw interaction between the tube and the substrate, which e ectively results in an underestimation of the net nanotube radial height reduction and the corresponding strain rate. It is noted that the consideration of the vdw interactions among all the elastic bodies in our model requires advanced continuum and/or atomistic level modeling techniques (e.g., nite element methods and molecule dynamics simulations), which is beyond the scope of this study. Structural Morphology of the Covering GO Sheet. The structural morphology of the GO sheet covered on top of the nanotube is of importance for understanding the compressive load applied on the nanotube and the adhesion interaction between the GO sheet and the substrate. It is noted that the substrate-regulated nanomembrane morphological deformations have recently been investigated based on monolayer graphene models using both continuum mechanics and molecular mechanics simulations.39À43 In this study, the GO sheet is considered as a membrane with uniform thickness, and its cross-section deformation on top of the nanotube is theoretically modeled using continuum mechanics theory. With the consideration of both membrane bending and stretching effects, the in-plane deformation of the membrane, w(x), is governed by d2 w P L DGO 2¼ N 3 wþ M0ÀÀx (2) dx 2 2 where DGO= (EGOt3)/(12(1Àν2 GO)) is the bending stiffness of the GO sheet in which EGO= 207 GPa, t= 0.7 nm and vGO= 0.17 are the Young's modulus, thickness, and Poisson's ratio of the monolayer GO sheet, respectively. N is the stretching force in the membrane. L is the spanning width of the GO sheet (see Figure 1b). M0 is the bending moment at x= (L/2. The boundary conditions for the deflected membrane include w= dw/dx= 0 at x= (L/2. It is noted that theVOL. 6’

NO. 2

’

1814– 1822

’

2012

1818

ARTICLE

Figure 4. (a) The comparison of the experimentally measured and theoretically predicted cross-section morphological pro les of the GO-sheet on top of CNT#1 as shown in Figure 1b. The red curve represents the theoretical prediction based on the model considering both membrane bending and stretching e ects, while the blue curve is based on the model considering only the membrane bending e ect. (b) The dependencies of the experimentally measured and theoretically predicted GO sheet covering length L on the nanotube diameter for both SWCNTs and SW-BNNTs. (c) The dependencies of the ratio L/Dnt on the nanotube diameter Dnt based on the measured values L for both the tested SWCNTs and SW-BNNTs.

maximum deformation of the membrane w(x= 0)= h. Both the displacement of the GO sheet w(x) and its spanning width L

can be obtained using eq 2 based on the calculated line load P shown in Figure 3b. When theZHENG ET AL .

stretching force is neglected (i.e., N= 0 in eq 2), the maximum deformation of the membrane due to pure membrane bending is given by h= (PL3)/(192DGO), and the deformation profile of the membrane is given by w(x)= 4h[3(1/2À x/L)2À 4(1/2À x/L)3] for x∈[0, L/2]. The deformation profile of the GO sheet covered on top of the nanotube is theoretically predicted for both scenarios (i.e., membrane bending only versus both membrane bending and stretching). Figure 4a shows the comparison of the theoretically predicted deformation profiles for CNT#1 shown in Figure 2b and the experimentally measured profile by AFM. The radius of curvature (Rtip) of the employed AFM tips is estimated to be within 15À25 nm based on the geometrical deconvolution relationship Rtip=ω2/8Δ, in whichΔ andω are the measured height and apparent width of a nanotube in the captured AFM images.44 Because the radius of curvature of the employed AFM tip is much smaller than the radius of curvature of the deformed GO sheet, the AFM-measured GO sheet deformation profiles are free of the well-known tipÀsample convolution artifacts. Our results shown in Figure 4a clearly reveal that the AFM-measured deformation profile of the GO sheet is in reasonably good agreement with the predicted deformation profile using eq 2 by considering both membrane bending and stretching effects. The measured GO sheet spanning width (L) using AFM is 24 nm, which is very close to the theoretically predicted value of 23 nm. It is noted that the theoretically predicted value of L using the model considering only the membrane bending effect is only 14 nm, indicating that the negligence of the stretching force in the membrane will substantially underestimate this quantity. Figure 4b shows the comparison of the theoretically predicted and experimentally measured spanning width of the GO sheet for all the tested nanotube samples, which consistently shows the good agreement on the GO sheet spanning width between the experimental measurements and the theoretically predicted values considering both the membrane bending and stretching effects. Figure 4c reveals that the ratios between the measured value of L and the nanotube diameter Dnt for all the tested SWCNTs and SW-BNNTs exhibit two decreasing and well-overlapping trends with the nanotube diameter. This observation suggests that the GO sheet wraps around the tube more tightly for tubes of larger diameters, which is consistent with the previously reported results on the wrapping configuration of graphene sheets around silicon nanowires on flat SiO2 substrates obtained using continuum and molecular mechanics simulations.39,43 The Adhesion Energy between the GO Sheet and the Substrate. The adhesion interaction between the GO sheet and the substrate influences both the covering morphology of the GO sheet and the compressive load applied on the

nanotube. The structural morphology of the GO sheet as shown in Figure 4a can be consideredVOL. 6’

NO. 2

’

1814– 1822

’

2012

1819

ARTICLE

Figure 5. (a) The dependencies of the per-unit-length bending moment in the GO sheet at the delamination interface with the substrate on the nanotube diameter for both SWCNTs and SW-BNNTs. (b) The dependencies of the perunit-area binding energy of the GO sheet to the substrate on the nanotube diameter based on the measurements for GO-covered SWCNTs and SW-BNNTs.

as a result of the delamination of the GO sheet from the substrate due to the existence of the underlying nanotube, which is ascribed to a balanced competition between the bending moment and the adhesion between the GO sheet and the substrate at the delamination front (x= (L/2). The reaction moment at the delamination front is equal to the bending moment in the GO sheet in the vicinity of the delamination front, that is, M0. The adhesion energy per unit area or energy release rate at the delamination front is given by45À47 G¼ 1 M0 2 2DGO (3)

sheet-covered SW-BNNTs displays a decreasing trend with the increase in the tube diameter, while this trend is not appreciable for GO sheet-covered SWCNTs. It is important to highlight here that the adhesion energy between the GO sheet and the substrate is not only due to the long-range nonbonded van der Waals interaction between the GO sheet and the substrate, but also due to other factors such as the adhesion between the GO sheet and the nanotube and the residual humidity in the chamber where the experiments were performed. It is noted that our estimated value of the adhesion energy between monolayer GO sheets and silicon substrates with native oxide layers is substantially lower than the recently reported data regarding the adhesion energy between monolayer graphene sheets and SiO2 substrates (0.15À0.45 J/m2).48,49 We argue that the functional groups on the basal graphene plane in the GO sheet increases the distance between the graphene basal plane to the substrate, thus signi cantly reducing the long-range nonbonded van der Waals interaction. It is noted that the elastic modulus of monolayer GO sheets was obtained from stretching experiments based on a thickness value of 0.7 nm that was measured separately by AFM thickness tests.27 Therefore, the calculated value of the bending sti ness (DGO) may be overestimated because the bending sti ness of a thin plate is a cubic function of its thickness. While the inaccuracy in the value of DGO will not have any material e ect on our results regarding the radial deformation of the nanotube, it will a ect the calculated deformational pro le of the GO sheet and its binding energy with the substrate. Our analysis shows that the actual GO sheet spanning with L may be lower, while the binding energy G may be higher, than our reported values as a result of the possible overestimation of DGO. CONCLUSIONS We investigate a ne

w approach of engineering radial deformations in SWCNTs and SW-BNNTs using monolayer GO nanomembranes. By using a combined experimentalÀtheoretical approach, we demonstrate that the GO sheet covering can induce substantial radial deformation in the underlying nanotube. Our results show that the radial deformation of the nanotube due to the GO sheet covering increases with the tube diameter and SW-BNNTs deforms more than SWCNTs of the same diameter. Our results capture net radial height reduction of up to 44.1% for SWBNNTs and up to 29.7% for SWCNTs, which are signi cantly higher than the radial deformation induced purely by the vdw interaction between the nanotube and the substrate. The engineered substantial radial deformation in both carbon and boron nitride nanotubes is expected to have an in uential impact on their electrical properties and thus their electronics applications. Since the GO sheet is water-soluble and can beVOL. 6’

Figure 5a shows the calculated bending moment at the delamination front for both the tested SW-BNNTs and SWCNTs. Figure 5b shows the dependencies of the corresponding binding energy between the GO sheet and the substrate on the tube diameter. For GO sheetcovered SWCNTs, the average binding energy and the corresponding deviation is calculated to be 0.038 ( 0.005 N/m, while it is 0.037 ( 0.013 for GO sheet-covered SW-BNNTs. The binding energy for GOZHENG ET AL .

NO. 2

’

1814– 1822

’

2012

1820

ARTICLE

deposited and patterned on the substrate using well-established micro- and nanofabrication techniques, it is plausible that our proposed approach can be applicable for large-scale manipulation of the radial

deformation and electrical properties of nanotubes, which can be useful for the development of large-scale nanoelectronic devices and nanoelectromechanical systems (NEMS).

EXPERIMENTS AND METHODSThe employed SWCNTs, purchased from Sigma-Aldrich Co. in the form of dried powders, were synthesized using chemical vapor deposition (CVD) methods. The employed SW-BNNTs were synthesized using pressurized vapor/condenser (PVC) methods.35 The GO sheets were prepared following a previously reported method50 and were measured to have a carbonto-oxygen atomic ratio of 2.06 using X-ray photoelectron spectroscopy (XPS). The XPS measurements on the GO sheet samples were performed at room temperature and under ultrahigh vacuum conditions using a PHI 5000 VersaProbe instrument with monochromatic Al KR X-rays at 1486.6 eV, a 200μm diameter spot, and constant pass energy of 23.5 eV at a power of 50 W. Both CNTs and BNNTs were dispersed in aqueous solution using ultrasonication with the aid of ionic surfactant sodium dodecylbenzenesulfonate (NaDDBS). 32,37 The supernatant of the dispersed nanotube solutions was deposited on clean Si wafers using spin-coating, and was subsequently rinsed multiple times with deionized (DI) water to remove residue surfactants. The synthesized GO sheets dissolved

in DI water were deposited on both clean Si wafers and Si wafers with dispersed nanotubes on top through spincoating. All the GO sheet and GO sheet-covered nanotube samples were dried in an oven set at 90°C for 12 h before AFM imaging measurements. All AFM imaging measurements reported in this paper were performed using a Park Systems XE-70 AFM operating at room temperature inside a sealed chamber back lled with dry nitrogen gas. The employed AFM is incorporated with a closed-loop feedback control feature in the XYZ axes. Tapping-mode AFM was performed with the tipÀsample interaction controlled in the attractive force regime using silicon AFM probes (Applied Nanostructures, Inc.) with a nominal spring constant of 48 N/m and resonant frequency of 190 kHz. The fast scanning direction of the AFM tip was controlled to be perpendicular to the nanotube longitude axis, and only straight tubes are selected for the analysis. All AFM images were collected at a scan rate of 1.0 Hz, a scan line of 1024Â 1024 pixels, and scan sizes of 1000À2000 nm. The root-mean-square (rms) noise in the resonating amplitude of the AFM cantilever is measured to be 0.46Å using the laser re ection scheme. The high resolution transmission electron microscopy (HRTEM) characterization of CNTs and BNNTs was performed using a JEM 2100F TEM (JEOL Ltd.) operated at accelerating voltages of 120À200 kV. Con ict of Interest: The authors declare no competing nancial interest. Acknowledgment. This work was funded by US Air Force O ce of Scienti c ResearchÀLow Density Materials Program under Grant Nos. FA9550-11-1-0042 and FA9550-10-1-0451, and was also partially supported by the American Chemistry Society Petroleum Research Fund. We thank Dr. In-Tae Bae for his assistance with the TEM characterization. The XPS, the ellipsometry, and the HRTEM measurements were performed using the facilities in the Analytical and Diagnostics Laboratory at Binghamton University's Small Scale Systems Integration and Packaging Center (S3IP). Supporting Information Available: Additional data about the nanotube radial height reduction due to the covering of monolayer graphene oxide sheets for both BNNTs and CNTs. This material is available free of charge via the Internet at http:// .

REFERENCES AND NOTES1. Tans, S. J.; Verschueren, A. R. M.; Dekker, C. RoomTemperature Transistor Based on a Single Carbon Nanotube. Nature 1998, 393, 49–52. 2. Postma, H. W. C.; Sellmeijer, A.; Dekker, C. Manipulation and Imaging of Individual Single-Walled Carbon Nanotubes with an Atomic Force Microscope. Adv. Mater. 2000, 12, 1299–1302. 3. Chen, Z. G.; Zou, J.; Liu, G.; Li, F.; Cheng, H. M.; Sekiguchi, T.; Gu, M.; Yao, X. D.; Wang, L. Z.; Lu, G. Q. Long Wavelength Emissions of Periodic Yard-Glass Shaped Boron Nitride Nanotubes. Appl. Phys. Lett. 2009, 94, 023105. 4. Wei, X. L.; Wang, M. S.; Bando, Y.; Golberg, D. Tensile Tests on Individual Multi-walled Boron Nitride Nanotubes. Adv. Mater.

2010, 22, 4895–4899. 5. Chopra, N. G.; Zettl, A. Measurement of the Elastic Modulus of a Multi-wall Boron Nitride Nanotube. Solid State Commun. 1998, 105, 297–300. 6. Arenal, R.; Wang, M. S.; Xu, Z.; Loiseau, A.; Golberg, D. Young Modulus, Mechanical and Electrical Properties of Isolated Individual and Bundled Single-Walled Boron Nitride Nanotubes. Nanotechnology 2011, 22, 265704. 7. Hernandez, E.; Goze, C.; Bernier, P.; Rubio, A. Elastic Properties of C and BxCyNz Composite Nanotubes. Phys. Rev. Lett. 1998, 80, 4502–4505. 8. Suryavanshi, A. P.; Yu, M. F.; Wen, J. G.; Tang, C. C.; Bando, Y. Elastic Modulus and Resonance Behavior of Boron Nitride Nanotubes. Appl. Phys. Lett. 2004, 84, 2527–2529. 9. Golberg, D.; Costa, P. M. F. J.; Lourie, O.; Mitome, M.; Bai, X. D.; Kurashima, K.; Zhi, C. Y.; Tang, C. C.; Bando, Y. Direct Force Measurements and Kinking under Elastic Deformation of Individual Multiwalled Boron Nitride Nanotubes. Nano Lett. 2007, 7, 2146–2151. 10. Ghassemi, H. M.; Lee, C. H.; Yap, Y. K.; Yassar, R. S. Real-Time Fracture Detection of Individual Boron Nitride Nanotubes in Severe Cyclic Deformation Processes. J. Appl. Phys. 2010, 108, 024314. 11. Xiao, Y.; Yan, X. H.; Cao, J. X.; Ding, J. W.; Mao, Y. L.; Xiang, J. Speci c Heat and Quantized Thermal Conductance of Single-Walled Boron Nitride Nanotubes. Phys. Rev. B 2004, 69, 205415. 12. Pop, E.; Mann, D.; Wang, Q.; Goodson, K.; Dai, H. J. Thermal Conductance of an Individual Single-Wall Carbon Nanotube above Room Temperature. Nano Lett. 2006, 6, 96–100. 13. Rubio, A.; Corkill, J. L.; Cohen, M. L. Theory of Graphitic BoronÀNitride Nanotubes. Phys. Rev. B 1994, 49, 5081– 5084. 14. Blase, X.; Rubio, A.; Louie, S. G.; Cohen, M. L. Stability and Band-Gap Constancy of BoronÀNitride Nanotubes. Europhys. Lett. 1994, 28, 335–340. 15. Lee, C. H.; Xie, M.; Kayastha, V.; Wang, J. S.; Yap, Y. K. Patterned Growth of Boron Nitride Nanotubes by Catalytic Chemical Vapor Deposition. Chem. Mater. 2010, 22, 1782–1787. 16. Gulseren, O.; Yildirim, T.; Ciraci, S.; Kilic, C. Reversible BandGap Engineering in Carbon Nanotubes by Radial Deformation. Phys. Rev. B 2002, 65, 155410. 17. Kinoshita, Y.; Ohno, N. Electronic Structures of Boron Nitride Nanotubes Subjected to Tension, Torsion, and Flattening: A First-Principles DFT Study. Phys. Rev. B 2010, 82, 085433. 18. Kim, Y. H.; Chang, K. J.; Louie, S. G. Electronic Structure of Radially Deformed BN and BC3 Nanotubes. Phys. Rev. B 2001, 63, 205408.

ZHENG ET AL .

VOL. 6

’

NO. 2

’

1814– 1822

’

2012

1821

ARTICLE

19. Shan, B.; Lakatos, G. W.; Peng, S.; Cho, K. J. First-Principles Study of Band-Gap Change in Deformed Nanotubes. Appl. Phys. Lett. 2005, 87, 2067697. 20. Crespi, V. H.; Cohen, M. L.; Rubio, A. In Situ Band Gap Engineering of Carbon Nanotubes. Phys. Rev. Lett. 1997, 79, 2093–2096. 21. Barboza, A. P. M; Gomes, A. P.; Archanjo, B. S.; Araujo, P. T.; Jorio, A.; Ferlauto, A. S.; Mazzoni, M. S. C.

; Chacham, H.; Neves, B. R. A. Deformation Induced Semiconductor-Metal Transition in Single Wall Carbon Nanotubes Probed by Electric Force Microscopy. Phys. Rev. Lett. 2008, 100, 256804. 22. Bai, X. D.; Golberg, D.; Bando, Y.; Zhi, C. Y.; Tang, C. C.; Mitome, M.; Kurashima, K. Deformation-Driven Electrical Transport of Individual Boron Nitride Nanotubes. Nano Lett. 2007, 7, 632–637. 23. Hertel, T.; Walkup, R. E.; Avouris, P. Deformation of Carbon Nanotubes by Surface Van Der Waals Forces. Phys. Rev. B 1998, 58, 13870–13873. 24. Hertel, T.; Martel, R.; Avouris, P. Manipulation of Individual Carbon Nanotubes and Their Interaction with Surfaces. J. Phys. Chem. B 1998, 102, 910–915. 25. Tang, T.; Jagota, A.; Hui, C. Y. Adhesion between SingleWalled Carbon Nanotubes. J. Appl. Phys. 2005, 97, 074304. 26. Tombler, T. W.; Zhou, C. W.; Alexseyev, L.; Kong, J.; Dai, H. J.; Lei, L.; Jayanthi, C. S.; Tang, M. J.; Wu, S. Y. Reversible Electromechanical Characteristics of Carbon Nanotubes under Local-Probe Manipulation. Nature 2000, 405, 769–772. 27. Suk, J. W.; Piner, R. D.; An, J. H.; Ruo , R. S. Mechanical Properties of Mono-layer Graphene Oxide. ACS Nano 2010, 4, 6557– 6564. 28. Cote, L. J.; Cruz-Silva, R.; Huang, J. X. Flash Reduction and Patterning of Graphite Oxide and Its Polymer Composite. J. Am. Chem. Soc. 2009, 131, 11027–11032. 29. Robinson, J. T.; Zalalutdinov, M.; Baldwin, J. W.; Snow, E. S.; Wei, Z. Q.; Sheehan, P.; Houston, B. H. Wafer-Scale Reduced Graphene Oxide Films for Nanomechanical Devices. Nano Lett. 2008, 8, 3441–3445. 30. Gomez-Navarro, C.; Weitz, R. T.; Bittner, A. M.; Scolari, M.; Mews, A.; Burghard, M.; Kern, K. Electronic Transport Properties of Individual Chemically Reduced Graphene Oxide Sheets. Nano Lett. 2007, 7, 3499–3503. 31. Stankovich, S.; Piner, R. D.; Chen, X. Q.; Wu, N. Q.; Nguyen, S. T.; Ruo , R. S. Stable Aqueous Dispersions of Graphitic Nanoplatelets via the Reduction of Exfoliated Graphite Oxide in the Presence of Poly(sodium 4-styrenesulfonate). J. Mater. Chem. 2006, 16, 155–158. 32. Zheng, M.; Chen, X.; Bae, I.-T.; Ke, C.; Park, C.; Smith, M. W.; Jordan, K. Radial Mechanical Properties of Single-Walled Boron Nitride Nanotubes. Small 2012, 8, 116–122. 33. Chen, C. M.; Yang, Q. H.; Yang, Y. G.; Lv, W.; Wen, Y. F.; Hou, P. X.; Wang, M. Z.; Cheng, H. M. Self-Assembled FreeStanding Graphite Oxide Membrane. Adv. Mater. 2009, 21, 3007–3011. 34. Pandey, D.; Reifenberger, R.; Piner, R. Scanning Probe Microscopy Study of Exfoliated Oxidized Graphene Sheets. Surf. Sci. 2008, 602, 1607–1613. 35. Smith, M. W.; Jordan, K. C.; Park, C.; Kim, J. W.; Lillehei, P. T.; Crooks, R.; Harrison, J. S. Very Long Single- and Few-Walled Boron Nitride Nanotubes via the Pressurized Vapor/Condenser Method. Nanotechnology 2009, 20, 505604. 36. DeBorde, T.; Joiner, J. C.; Leyden, M. R.; Minot, E. D. Identifying Individual Single-Walled and Double-Walled Carbon Nanotubes by Atomic Force Microscopy. Nano Lett. 2008,

8, 3568–3571. 37. Zheng, M.; Ke, C.; Bae, I.-T.; Park, C.; Smith, M. W.; Jordan, K. Radial Elasticity of Multi-walled Boron Nitride Nanotubes. Nanotechnology 2012, in press. 38. Puttock, M. J.; Thwaite, E. G. Elastic Compression of Spheres and Cylinders at Point and Line Contact; Commonwealth Scienti c and Industrial Research Organization: Melbourne, Australia, 1969. 39. Zhang, Z.; Li, T. Graphene Morphology Regulated by Nanowires Patterned in Parallel on a Substrate Surface. J. Appl. Phys. 2010, 107, 103519.

40. Li, T.; Zhang, Z. Substrate-Regulated Morphology of Graphene. J. Phys: D-Appl. Phys. 2010, 43, 075303. 41. Li, T.; Zhang, Z. Snap-through Instability of Graphene on Substrates. Nanoscale Res. Lett. 2010, 5, 169–173. 42. Aitken, Z. H.; Huang, R. E ects of Mismatch Strain and Substrate Surface Corrugation on Morphology of Supported Monolayer Graphene. J. Appl. Phys. 2010, 107, 123531. 43. Zhang, Z.; Li, T. A Molecular Mechanics Study of Morphologic Interaction between Graphene and Si Nanowires on a SiO2 Substrate. J. Nanomater. 2011, 2011, 374018. 44. Gibson, C. T.; Watson, G. S.; Myhra, S. Scanning Force Microscopy;Calibrative Procedures for 'Best Practice'. Scanning 1997, 19, 564–581. 45. Ke, C.; Zheng, M.; Zhou, G.; Cui, W.; Pugno, N.; Miles, R. N. Mechanical Peeling of Free-Standing Single-Walled Carbon Nanotube Bundles. Small 2010, 6, 438–445. 46. Zheng, M.; Ke, C. H. Elastic Deformation of Carbon-Nanotube Nanorings. Small 2010, 6, 1647–1655. 47. Ke, C. H.; Zheng, M.; Bae, I. T.; Zhou, G. W. Adhesion-Driven Buckling of Single-Walled Carbon Nanotube Bundles. J. Appl. Phys. 2010, 107. 48. Koenig, S. P.; Boddeti, N. G.; Dunn, M. L.; Bunch, J. S. Ultrastrong Adhesion of Graphene Membranes. Nat. Nanotechnol. 2011, 6, 543–546. 49. Zong, Z.; Chen, C. L.; Dokmeci, M. R.; Wan, K. T. Direct Measurement of Graphene Adhesion on Silicon Surface by Intercalation of Nanoparticles. J. Appl. Phys. 2010, 107, 026140. 50. Hummers, W. S.; O eman, R. E. Preparation of Graphitic Oxide. J. Am. Chem. Soc. 1958, 80, 1339–1339.

ZHENG ET AL .

VOL. 6

’

NO. 2

’

1814– 1822

’

2012

1822