(PDF) Control of Shape and Material Composition of Solid-State Nanopores - DOKUMEN.TIPS (2024)

(PDF) Control of Shape and Material Composition of Solid-State Nanopores - DOKUMEN.TIPS (1)

Control of Shape and MaterialComposition of Solid-State NanoporesMeng-Yue Wu,† Ralph M. M. Smeets,† Mathijs Zandbergen,‡ Ulrike Ziese,†Diego Krapf,†,§ Philip E. Batson,‡ Nynke H. Dekker,† Cees Dekker,†and Henny W. Zandbergen*,†

KaVli Institute of Nanoscience, Delft UniVersity of Technology, Lorentzweg 1,2628 CJ Delft, The Netherlands, and IBM Thomas J. Watson Research Center,Yorktown Heights, New York 10598

Received November 28, 2008

ABSTRACT

Solid-state nanopores fabricated by a high-intensity electron beam in ceramic membranes can be fine-tuned on three-dimensional geometryand composition by choice of materials and beam sculpting conditions. For similar beam conditions, 8 nm diameter nanopores fabricated inmembranes containing SiO2 show large depletion areas (70 nm in radius) with small sidewall angles (55°), whereas those made in SiN membranesshow small depletion areas (40 nm) with larger sidewall angles (75°). Three-dimensional electron tomograms of nanopores fabricated in aSiO2/SiN/SiO2 membrane show a biconical shape with symmetric top and bottom and indicate a mixing of SiN and SiO2 layers up to 30 nmfrom the edge of nanopore, with Si-rich particles throughout the membrane. Electron-energy-loss spectroscopy (EELS) reveals that the oxygen/nitrogen ratio near the pore depends on the beam sculpting conditions.

Solid-state nanopores are used to detect and characterizeDNA and RNA with single-molecule resolution,1 comparableto the translocation measurements performed on biologicalphospholipid-embedded protein channels.2 They have sig-nificant advantages over protein channels as they arefunctional in a wider range of temperatures, solvents, andvoltages and offer possibilities for device integration andtunability in the pore dimensions. A nanopore can be readilyformed in thin membranes by an electron beam in a TEM.3-5

This TEM-based method provides an advantage of excellentsize control as the nanopore can directly be visualized duringdrilling process. In addition, the shape of the nanopore canbe controlled, provided that the electron beam is very wellaligned and the specimen drift is low. The sidewall abrupt-ness, which is an important factor in the analysis of DNAtranslocation through nanopores, can be tuned with thedrilling conditions (beam size, beam intensity) and on thesample composition. Notably, a very different sidewallabruptness was observed in nanopores fabricated in 40 nmSiO2/SiN4 and 50 nm Si3N4 membranes,5,6 demanding furtherinvestigation. Another very important factor in the analysisof DNA translocation is the surface composition of thenanopore, as Smeets et al. 7 have suggested following

examination of a range of nanopores. Since SiN and SiO2

may yield different surface compositions resulting in differenthydrophilicities, the control of surface charge via TEM-engineered material properties is an interesting perspective.

To better control the shape and material composition ofsolid-state nanopores, we here use electron-energy-lossspectroscopy (EELS), energy-filtered TEM (EFTEM), elec-tron tomography, scanning transmission-electron microscopy(STEM), and high-resolution transmission-electron micros-copy (HRTEM) to measure the shape and composition ofthe nanopore as a function of electron-beam drilling proce-dures and membrane composition. Electron tomographyshows that the shape of the pore is mirror symmetric aboutthe middle plane of the membrane and yields a thicknessvariation in good agreement with EFTEM data. By com-bining EELS and HRTEM, we obtain information on thematerial surrounding the pore; we show that small Si-richparticles are formed by a high-intensity electron beam in themembranes that contain SiO2, but not in those formed inpure SiN. These small Si-rich particles are found throughoutthe membrane in the vicinity of the pore’s rim, and indicatea thorough local mixing of nitride and oxide layers that iscontrolled by the electron beam. Furthermore, by combiningEELS and STEM, we show that the oxygen and nitrogencomposition in the region surrounding the nanopore issensitive to the drilling conditions in the TEM.

Membranes were fabricated using standard semiconductormicrofabrication processes. First, the following three-layer

* To whom correspondence should be addressed. E-mail:[emailprotected]. Fax: +31-15-2786730.

† Delft University of Technology.‡ IBM Thomas J. Watson Research Center.§ Present address: Electrical and Computer Engineering, Colorado State

University, Fort Collins, Colorado 80523-1373.

NANOLETTERS

2009Vol. 9, No. 1

479-484

10.1021/nl803613s CCC: $40.75 2009 American Chemical SocietyPublished on Web 12/09/2008

(PDF) Control of Shape and Material Composition of Solid-State Nanopores - DOKUMEN.TIPS (2)

structure was deposited on both sides of a 20-30 Ωcm,boron-doped, Si <100> wafer by low-pressure chemicalvapor deposition (LPCVD): 20-60 nm of low-stress silicon-rich silicon nitride (SiN), followed by 200 nm of SiO2, anda 500 nm thick capping layer of low stress SiN. Next, squarewindows were patterned in the backside layers using e-beamlithography and CHF3 reactive ion etching. Using thebackside SiN as a mask, the Si substrate was anisotropicallyetched in KOH solution (29 wt %) at 80 °C during 8 h. Free-standing 50 × 50 µm2 membranes consisting of the three-layer stack were realized. In the middle of the membrane, a5 µm wide region was thinned to expose only the lowermost20-60 nm SiN layer. For this, the capping SiN and SiO2

layers were removed by reactive ion etching and wet HFetching, respectively. The final fabrication step for the triplelayer membrane involves the deposition of a 20 nm of siliconoxide film by sputtering in Ar plasma on both sides of theSiN membrane. A thin foil sample of amorphous SiO2 wasprepared using standard plan-view TEM sample preparationmethods.

The nanopores were drilled and monitored using a fieldemission FEI Tecnai (S)TEM operated at an acceleratingvoltage of 200 kV. The (S)TEM system included a mono-chromator and a high-resolution Gatan Imaging Filter(HRGIF).8,9 EELS spectra were obtained with the mono-chromator in off-mode since the intrinsic widths of the EELSpeaks of Si, N, and O do not require the high resolution.Moreover, the higher brightness in off-mode allows a quickerformation of the nanopores. An electron beam with adiameter between 2 to 10 nm (full width at half-maximumheight, fwhm) and a beam current of 2-7 nA was used fordrilling. The N and O compositions were acquired with themicroscope operant in STEM mode with a spot size about0.5 nm, a camera length of 30-50 mm, a HRGIF entranceaperture of 2 mm, and an energy dispersion of either 0.5 or0.2 eV/pixel. The energy-filtered transmission-electron mi-croscopy (EFTEM) images were typically recorded using a5 mm HRGIF entrance aperture and a 4 eV energy window(with the exception of a 10 eV energy window for thethickness map). Note that all experiments were repeatedseveral times but only representative results are shown here.

TEM tomography using a Tecnai 20 was performed on a20 nm nanopore in a SiO2/SiN/SiO2 membrane. Thisnanopore was enlarged by a ∼20 nm electron probe startingfrom an 8 nm nanopore that had previously been fabricatedby a ∼10 nm electron probe with a beam current of 10-15nA in a Philips 300UT microscope operated at 300 kV.Double-axis tilt series of 140 images for each axis wereacquired from -70 to 70° with a tilt increment of 1°. IMODsoftware10 was used to compute 3D reconstructions from thetilt series. A pixel size of 0.72 nm was obtained for the finalreconstruction. The resulting three-dimensional structure ofthe nanopore was visualized using the surface reconstructionfunction included in Amira 4.1 (Mercury). To avoid electron-beam-induced changes during the tilt series, a low currentdensity was used, and the total dose was ∼8 × 104 e/nm2.No changes on the nanopore could be detected during thetilt series recording.

For the experiments involving drilling using a very smallspot size, a STEM located at IBM in Yorktown Heights wasused, which is based on a VG Microscopes HB501 STEMwith the addition of a quadrupole-octupole aberrationcorrector.11 The electron beam employed had a spot size of0.1-0.15 nm and a current of 30-50 pA at 120 kV, andthese experiments were performed on a 40 nm thick SiNmembrane.

Nanopores can be fabricated in numerous types ofmembranes. Here we compare the thickness profiles ofnanopores drilled in a single SiN membrane (20-60 nm inthickness), a triple layer SiO2/SiN/SiO2 membrane (20/20/20 nm) and also a pure SiO2 foil. In Figure 1, we plot thethickness variations surrounding a 8 nm diameter nanoporein a 60 nm thick SiO2/SiN/SiO2 membrane, an ∼10 nmdiameter nanopore in an ∼60 nm SiN membrane, and an∼10 nm diameter nanopore in a pure amorphous SiO2

membrane fabricated under similar drilling conditions (twobeam sizes were used: a 9 nm (fwhm) beam for drilling anda 36-54 nm beam for short inspection of the nanopore (∼3%of drilling time)). A mean free path of 180 nm (calculatedusing the formulas in ref 12) was used to obtain the absolutethickness value of all three membranes. An uncertainty of∼3% in the mean free path is expected as compositionchanges from pure Si to triple layer. The thickness profilein SiN membrane shows a sidewall angle of about 75° anda depletion area (<90% original thickness) of about 40 nmin radius, which is similar to the result reported by Kim etal.5,6 In contrast, the edge of the nanopore in a SiO2/SiN/SiO2 membrane is more wedge-shaped with a sidewall angleabout 55° and a depletion area of about 70 nm in radius,similar to that in a SiN/SiO2 double layer membrane thatwe reported previously.4 Nanopores formed in pure SiO2

membranes were found to have a geometry that is compa-rable to that found in nanopores formed in the triple layer

Figure 1. Measured and calculated thickness variation profiles. (a)Black, an 8 nm nanopore in 60 nm SiO2/SiN/SiO2 membrane; red,a 10 nm nanopore in ∼60 nm SiN membrane; green, a calculated8 nm nanopore assuming a linear mass loss with respect to theelectron beam intensity. Electron probes (9 and 45 nm) were used.The total dose of the 45 nm probe is 1/10 of the total dose of the9 nm probe. (b) A 10 nm nanopore in an amorphous SiO2 foil.

480 Nano Lett., Vol. 9, No. 1, 2009

(PDF) Control of Shape and Material Composition of Solid-State Nanopores - DOKUMEN.TIPS (3)

(Figure 1b). In all three cases, the large extent of the depletionarea compared to the size of the electron beam is remarkable.Assuming a Gaussian beam profile and a material loss thatis linear with beam dose, the material depletion profile forthe experimental conditions we used is calculated (Figure1, green line). Comparing this estimated depletion profilewith the experimental shapes (Figure 1, red and black dottedlines), we conclude that there must be a lateral displacementof material. Similar phenomena were also observed when avery small beam (0.1-0.15 nm) was used to drill nanoporesin 40 nm thick SiN membrane whereby the membrane wasthinned in an area with a radius of 4 nm (data not shown).

In addition to the difference in the shape of the nanoporesfabricated in SiN membranes versus SiO2/SiN/SiO2 or pureSiO2 membranes, we also observed an effect of membranecomposition on the formation of small Si-rich particles inthe membranes. Figure 2a,b shows images of two 8 nmdiameter nanopores fabricated in a 20 nm SiN and in a 60nm SiO2/SiN/SiO2 membranes under similar drilling condi-tions, respectively. Small particles with a size of about 3nm can be clearly seen in all the membranes that contain aSiO2 layer (Figure 2b), but not in the SiN single layermembranes (Figure 2a). To exclude an effect of the filmthickness, we also confirmed that 40 and 60 nm SiNmembranes did not show the formation of particles. Thesesmall particles were identified as Si-rich particles by EFTEMusing a 15-19 eV energy window, which included the bulkplasmon peak of Si at 16.7 eV, while excluding the plasmonpeaks of SiN and SiO2 near 22.8 eV. Figure 1c-e showsEFTEM images of a 14 nm nanopore in SiO2/SiN/SiO2

membrane with different energy windows: an elastic image(-2 to +2 eV; Figure 1c), an EFTEM image around the Sibulk plasmon (Figure 2d), and an EFTEM image around thebulk plasmon for SiN and SiO2 (Figure 2e). The smallparticles are clearly visible in the 15-19 eV EFTEM image(Figure 2d), showing that these particles are Si-rich. No

lattice fringes were found in HREM images of these particles.More detailed information of valence states of Si in theparticles cannot be given due to their intrinsic energy widths.Dori et al.13 have reported that in substoichiometric oxidesexcess silicon is present either as nanometer-sized siliconislands or as submicroscopic silicon oxides of varyingstoichiometry. However, the possibility that the electronbeam might modify the valence state of Si was not discussedby Dori et al. Chen et al.14 found that amorphous Si dots, orwires surrounded by SiO2-x, can be formed in amorphous 15nm thick SiO2 films by a high-intensity 100 keV focusedelectron probe.

To determine the membrane composition at which thesmall Si-rich particles start to form, a line scan wasperformed across the nanopore (see the inset profile in Figure2b). Around the nanopores in SiO2/SiN/SiO2 membranes, theaverage radius of the O depletion area is about 75 nm,whereas that of the N depletion area is about 30 nm.Comparing the locations of the small Si-rich particles in theTEM image with the oxygen composition distribution at therim of the nanopore, it can be seen that the particles arepresent in the area where ∼20% oxygen is lost. Interestingly,the area of nitrogen loss in the case of a sandwich betweentwo SiO2 layers is larger (by about 10 nm) than the lossarea of a single SiN layer, which holds for both 20 and 60nm thick SiN films. This indicates that SiN reacts with theSiO2 layers in that area of the triple layer.

Using our control of the electron beam, we can fine-tunethe size of the nanopore. In a previous paper, we have alreadyreported that the nanopore size can be reduced, but here wedemonstrate that enlargement is also possible via thistechnique.15 Two methods were used for pore enlargement:(1) we defocused the small beam so that the rim of thenanopore is equally irradiated, and (2) we continue thedrilling with the tail of the focused beam. Figure 3 plots thethickness profiles of two nanopores that are enlarged from8 to 18 nm in SiO2/SiN/SiO2 triple layer membranes. Theprofiles shown in Figure 3a are from the same pore beforeand after enlargement with method (1) using a ∼20 nmdefocused electron beam, whereas the profiles in Figure 3care from the initial small pore and its enlargement throughmethod (2) using a focused ∼10 nm electron beam. Bothenlargements result in changes in the nanopore geometry.However, the changes are different, as can be seen from thedifference profiles of the thicknesses before and afternanopore enlargements by these two methods (Figure 3b,d).When method (2) is employed, the maximum loss of materialoccurs at ∼10 nm from the edge whereas material is lost ina region up to ∼ 20 nm from the edge of nanopore. Thisresults in a steeper sidewall, for example, the nanoporebecomes more tunnel-like. In contrast, when method (1) isemployed, the maximum loss of matter occurs at ∼20 nmfrom the edge and the overall material loss range is in a∼45 nm region from the edge of the nanopore. Consequently,with a defocused beam the surrounding of the nanoporebecomes thinner while keeping the sidewall angle nearlyunchanged.

Figure 2. TEM images of two nanopores with sizes of 8 nmfabricated in (a) 20 nm SiN membrane and (b) 60 nm SiO2/SiN/SiO2 membrane under similar drilling conditions. The inset showsHRTEM images of these nanopores. The inset profile in (b) is thedistribution of O and N. (c-e) EFTEM images of a 14 nm nanoporein a SiO2/SiN/SiO2 membrane acquired with different energywindows (c) -2-2 eV, (d) 15-19 eV and (e) 21-25 eV.

Nano Lett., Vol. 9, No. 1, 2009 481

(PDF) Control of Shape and Material Composition of Solid-State Nanopores - DOKUMEN.TIPS (4)

The two different ways to enlarge a small nanopore in a∼60 nm SiO2/SiN/SiO2 membrane also lead to different O/Nratios near the edge of the nanopore, as can be seen in Figure4. In Figure 4a,b, we compare the O and N compositionbefore and after enlargement for the two enlargementtechniques. Using the defocused beam, the O/N ratio issmaller in the depletion area than before enlargement (seeFigure 4a). Through the use of the highly focused beam, theO/N ratio does not change significantly (see Figure 4b). Notethat the O and N spectra were acquired simultaneously underlow beam intensity conditions to avoid additional materialalteration by the electron probe. This results in higher scatterin the N signal since the O/N atomic ratio is about 3.5 inthe initial sample. Consequently the O/N profiles aresomewhat noisy, but the signal-to-noise is sufficient to allowus to reproducibly observe the trends of the changes in O/Nby the two different enlargement methods.

Finally, to determine the 3D geometry of the nanopore,TEM tomography was carried out on a ∼20 nm nanoporefabricated in a ∼60 nm SiO2/SiN/SiO2 membrane. Theresulting three-dimensional shape is shown in Figure 5a. Notethat the shape of the pore is mirror symmetric, with the mirrorplane in the middle of the trilayer. Although this 20 nmnanopore was fabricated with different beam settings in theprimary beam energy and beam current, the thicknessvariation is similar to that of the ∼20 nm nanopore fabricatedin a ∼60 nm SiO2/SiN/SiO2 membrane shown in Figure 3a.A comparison of the two thickness profiles is given in Figure

5b. The sidewall angle indicated in the inset of Figure 5bcan be simply calculated from the sidewall angle R measuredfrom the thickness profile as ) 2 arctan(tan(R)/2).

In addition to the three-dimensional shape of nanopore,we obtained the distribution of Si-rich particles from TEMtomography. Several slices of the tomogram are depicted inFigure 5c-h. In the first 30 nm from the edge of the pore,the Si-rich particles are present throughout the entire thick-ness of the film. In the area between 30 and 70 nm from theedge, the Si rich particles decrease gradually in numberdensity in SiO2 layer, and they are not present in the middleSiN layer. This last observation, in combination with the factthat there is no significant loss in N in this area, indicatesthat in this area there is no change in the SiN layer both isshape and in composition, and thus that no mixing with theSiO2 layers occurred.

The tomography allows us to distinguish two areassurrounding the nanopore (see Figure 5): Area 1 with a radiusof about 30 nm, including the pore and its direct surround-ings, and Area 2 in a region with a radius between 30 and70 nm radii, where there is significant O loss and no N loss.We assume that in Area 1, which is the area irradiated witha high intensity electron beam, the material almost behavesas a liquid during the drilling. This assumption is supported

Figure 3. Thickness variation profiles (acquired by EFTEM) aroundnanopores before and after enlargement and the respective differenceprofiles via the use of (a) an ∼20 nm defocused beam and (c) an∼10 nm focused beam.

Figure 4. The distribution of N and O atoms before and afterenlargement by the electron beam in two different ways. (a)Defocused beam and (b) highly focused beam. An O/N cross sectionratio of 1.87 was used in O/N atomic ratio profile as appropriatefor our experiment condition. The data of all O/N profiles are plottedstarting at the edge of the large nanopore.

482 Nano Lett., Vol. 9, No. 1, 2009

(PDF) Control of Shape and Material Composition of Solid-State Nanopores - DOKUMEN.TIPS (5)

by the constant changes in contrast observed (movie,Supporting Information), which indicates very rapid changesin the atoms’ positions. This is quite plausible as 200 keVelectrons can easily break O and N bonds. It is also supportedby experiments by Kimoto et al.16 on SiO2/Si3N4/SiOxNy/Simultilayers, which showed for instance that O atoms can bedragged by the electron beam from the SiO2 layer into theSi3N4 layer. Such liquidlike behavior would lead to mixingof O and N over the liquidlike volume, and concomitantlySi rich particles all throughout the membrane thickness, aswe observed. If the beam is displaced or enlarged, newvolume is added to the liquidlike part, and if the addedvolume has a different composition, the overall compositionof the liquidlike volume will change. This explains why moreN-rich material is found near the edge of the pore when anenlarged beam is used. This effect allows us to change thecomposition of the rim. Another effect of the liquidlikebehavior is that during the removal of material, the surfaceson the top and bottom of the membrane are governed by aminimization of the surface energy, and thus that top andbottom parts of the pore will have very similar shapes andwill be mirror symmetric in Area 1. This is in contrast withcrystalline materials where the formation of the pore by ahigh electron beam starts with the formation of voids and ahole in the bottom side of the specimen and where the topside at first shows almost no changes.17,18 Our results implythat one could in principle deposit a layer on the membraneafter the pore formation and mix this into the membranematerial with the electron beam, thereby fine-tuning thechemical composition. In Area 2, diffusion along themembrane normal does not occur because the negligiblenitrogen loss in Area 2 indicates that the SiN layer remainsintact in this region. Thus, lateral diffusion should take place.From the mirror symmetry in Area 2, we conclude thatsputtering is not a dominant mechanism, which is consistentwith the low exposure to the electron beam.

Our results suggest that the shape of the edge of thenanopore can be optimized by the choice of the compositionof the membrane in particular using multiphase membranes,like the SiO2/SiN/SiO2 membranes used here. Since the local

surface roughness and hydrophobicity are speculated to bethe reason for nanobubble formation, a prominent source ofnoise in solid-state nanopores,16 it is advantageous to be ableto engineer the composition near the nanopore. Control overthe three-dimensional structure and local properties ofnanopores is of importance to improve the reliability of thesesensors,19,20 and to interpret results of translocation measure-ments.7

In summary, solid-state nanopores can be fabricated inmembranes with a range of SiN/SiO2 compositions by a high-intensity electron beam. The final geometry of the nanoporeis dependent on material composition and drilling conditions.In addition, once a nanopore is created, the local compositionat the edge can be changed by mixing in material of themembrane part that has a composition different from theedge, which is possible due to the fluidlike behavior of thearea that is irradiated with an intense electron beam. Theresults given in this paper enable the fabrication of improvedand well-characterized nanopores.

Acknowledgment. This work was financially supportedby the Dutch foundation for Fundamental Research on Matter(FOM) and The Netherlands Organization for ScientificResearch (NWO). N.H.D. acknowledges funding from theEuropean Young Investigators program (EURYI) of theEuropean Science Foundation (ESF). The authors would liketo thank Anna Carlsson of the FEI Company for imageacquisition for 3D tomography and Gregory Pandraud of theDelft Institute of Microelectronics and Submicron Technol-ogy, Delft University of Technology for a supply of SiNsamples.

Supporting Information Available: High resolutionelectron microscopy movie of the formation of a nanoporein a 60 nm SiO2/SiN/SiO2 membrane. An ∼10 nm electronprobe with primary energy of 300 keV and beam current of7 nA is used for drilling. Because of the dynamical range ofthe CCD, there is no contrast in the area with very highintensity. An enlarged beam has to be used to view thechanges induced by the focused beam in the material. The

Figure 5. Three-dimensional structure of a 20 nm nanopore fabricated in SiO2/SiN/SiO2 membrane. (a) Surface reconstruction image. (b)Comparison of the thickness profiles deduced from the three-dimensional reconstruction and the one of the 18 nm nanopore shown inFigure 3a. (c-h) Five slices through the membrane with cutting positions indicated in panel a. The blue circles in panels c-h have 30 and70 nm radii delineating the two different areas discussed in the text.

Nano Lett., Vol. 9, No. 1, 2009 483

(PDF) Control of Shape and Material Composition of Solid-State Nanopores - DOKUMEN.TIPS (6)

material under the focused beam becomes thinner andthinner. Once we see clearly a 5 nm nanopore is created,we defocus the beam. The nanopore increases in size untilthe electron beam is indeed defocused sufficiently. Thenanopore shrinks continually to 5 nm as we refocus the beam(∼100 nm). This material is available free of charge via theInternet at http://pubs.acs.org.

References(1) Dekker, C. Nat. Nanotech. 2007, 2, 209.(2) Kasianowicz, J. J.; Brandin, E.; Branton, D.; Demer, D. W. Proc. Nat.

Acad. Sci. U.S.A. 1996, 93, 13770.(3) Storm, A. J.; Chen, J. H.; Ling, X. S.; Zandbergen, H. W.; Dekker,

C. Nat. Mater. 2003, 2, 537.(4) Wu, M.-Y.; Krapf, D.; Zandbergen, M.; Zandbergen, H. W.; Batson,

P. E. Applied Phys. Lett. 2005, 87, 113106.(5) Kim, M. J.; Wanunu, M.; Bell, D. C.; Meller, A. AdV. Mater. 2006,

18, 3149.(6) Kim, M. J.; McNally, B.; Murata, K.; Meller, A. Nanotechnology 2007,

18, 205302.(7) Smeets, R. M. M.; Keyser, U. F.; Krapf, D.; Wu, M.-Y.; Dekker, N. H.;

Dekker, C. Nano Lett. 2006, 6, 89.

(8) Tiemeijer, P. C.; van Lin, J; H, A.; de Jong, A. F. Microsc. Microanal.2001, 7 (suppl. 2), 1130–1131.

(9) Lazar, S.; Botton, G. A.; Wu, M.-Y.; Tichelaar, F. D.; Zandbergen,H. W. Ultramicroscopy 2003, 96, 535.

(10) Kremer, J. R.; Mastronarde, D. N.; McIntosh, J. R. J. Struct. Biol.1996, 116, 71.

(11) Batson, P. E.; Dellby, N.; Krivanek, O. L. Nature 2002, 418, 8.(12) Egerton, R. F. Plenum Press: New York, 1996; Vol. 305.(13) Dori, L.; Bruley, J.; DiMaria, D. J.; Batson, P. E.; Tornello, J.; Arienzo,

M. J. Appl. Phys. 1991, 69 (4), 2317–2323.(14) Chen, G. S.; Boothroyd, C. B.; Humphreys, C. J. Appl. Phys. Lett.

1993, 62, 1949.(15) Note that it also can be shrunk by a less focused beam about 100

times lower in current density than the focused one. See also ref 4.(16) Kimoto, K; Isakozawa, S.; Aoyama, T.; Mataui, Y. J. Electron Microsc.

2001, 50, 523.(17) Bullough, T. J.; Devenish, R. W.; Humphreys, C. J. Inst. Phys. Conf.

Ser. 1990, 98, 267.(18) Tang, B. B; Jones, I. P.; Lai, W. S.; Bacon, D. J. Philos. Mag. 2005,

17, 1805.(19) Smeets, R. M. M.; Keyser, U. F.; Wu, M. Y.; Dekker, N. H.; Dekker,

C. Phys. ReV. Lett. 2006, 97, 088101.(20) Chen, P.; Mitsui, T.; Farmer, D. B.; Golovchenko, J.; Gordon, R. G.;

Branton, D. Nano Lett. 2004, 4, 1333.

NL803613S

484 Nano Lett., Vol. 9, No. 1, 2009

(PDF) Control of Shape and Material Composition of Solid-State Nanopores - DOKUMEN.TIPS (2024)
Top Articles
Latest Posts
Article information

Author: Catherine Tremblay

Last Updated:

Views: 6439

Rating: 4.7 / 5 (47 voted)

Reviews: 94% of readers found this page helpful

Author information

Name: Catherine Tremblay

Birthday: 1999-09-23

Address: Suite 461 73643 Sherril Loaf, Dickinsonland, AZ 47941-2379

Phone: +2678139151039

Job: International Administration Supervisor

Hobby: Dowsing, Snowboarding, Rowing, Beekeeping, Calligraphy, Shooting, Air sports

Introduction: My name is Catherine Tremblay, I am a precious, perfect, tasty, enthusiastic, inexpensive, vast, kind person who loves writing and wants to share my knowledge and understanding with you.