鈴木 哲

Optical absorption efficiency of graphene integrated onto a silicon photonic platform was characterized at around 1.55-mu m optical telecommunications wavelength. Micro-Raman spectroscopy performed after the completion of all fabrication processes confirmed that transferred chemical-vapor-deposited graphene is a single layer (>90% coverage) without any significant damage. Absorption efficiencies of the single-layer graphene (SLG) on a silicon wire waveguide, obtained by measuring different lengths (cutback method) of the SLG from 2.5 to 200 mu m, were 0.09 and 0.05 dB/mu m for TE- and TM-polarized light. The unusual relationship in the polarization dependency can be explained by strong surface-plasmon-porlariton support in the TM mode. (c) The Japan Society of Applied Physics

Large-area, atomically thin hexagonal boron nitride (h-BN) thin films were grown simply by annealing in a vacuum from solid sources deposited on Ni or Co foils. Either a sputter-deposited amorphous boron nitride film or a spin-coated borane ammonia film can be used as the B and N source. The h-BN films were formed on the back surface of the metal foil following diffusion of B and N atoms through the foil of similar to 20 mu m-thick, although N is almost completely insoluble in these metals. The atomically thin h-BN film largely prevented the foil from oxidizing. The h-BN film formation was found to be restricted by the provision of N atoms. The authors propose that the mass transport of N atoms in the foil and on the back surface of the foil is dominated by grain boundary diffusion and surface migration. (C) 2013 American Vacuum Society.

Nanometer-thick hexagonal boron nitride thin films were grown by thermal chemical vapor deposition on polycrystalline Ni, Co, and Cu substrates. A thicker and more regularly stacked film was grown on a Ni substrate, whereas a smaller film thickness and more turbostratic stacking or smaller domain size were observed on Cu. Intermediate situations were observed on Co. The substrate material dependence strongly suggests that the substrate plays an important role in h-BN growth, like it does in graphene growth, although nitrogen is almost insoluble even in Ni. Grain boundaries may accommodate boron and nitrogen atoms at a growth temperature. © 2012 The Surface Science Society of Japan.

By analytically constructing the matrix elements of an electron-phonon interaction for the D band in the Raman spectra of armchair graphene nanoribbons, we show that pseudospin and momentum conservation result in (i) a D band consisting of two components, (ii) a D band Raman intensity that is enhanced only when the polarizations of the incident and scattered light are parallel to the armchair edge, and (iii) the D band softening/hardening behavior caused by the Kohn anomaly effect is correlated with that of the G band. Several experiments are mentioned that are relevant to these results. It is also suggested that pseudospin is independent of the boundary condition for the phonon mode, while momentum conservation depends on it.

Atomically thin hexagonal boron nitride films were grown on both the top and bottom surfaces of a polycrystalline Co or Ni film by annealing a Co (Ni)/amorphous boron nitride/SiO2 structure in vacuum. This method of growing hexagonal boron nitride is much simpler than other methods, such as thermal chemical vapour deposition. B and N atoms diffuse through the metal film, although N is almost completely insoluble in both Co and Ni, and precipitation occurs at the topmost surface. The mass transport is considered to be caused by grain boundary diffusion.

We grew graphene films on polycrystalline metal substrates using spin-coated polystyrene or polyaniline films as a carbon source by heating the films in an Ar atmosphere or in a vacuum. In an Ar atmosphere of 6.7 x 10(4) Pa, precise control of the initial polymer film thickness is crucial for few-layer graphene growth. In a vacuum, few-layer graphene films were obtained regardless of the initial polymer film thickness, because excess carbon atoms are removed from the surface when the polymer is thermally decomposed. This latter method does not require any gas and is thus a very simple and easy way to grow few-layer graphene. (C) 2012 The Japan Society of Applied Physics

By considering analytical expressions for the self-energies of intervalley and intravalley phonons in graphene, we describe the behavior of D, 2D, and D' Raman bands with changes in doping (mu) and light-excitation energy (E-L). Comparing the self-energy with the observed mu dependence of the 2D bandwidth, we estimate the wave vector q of the constituent intervalley phonon at (h) over bar vq similar or equal to E-L/1.6 (v is the electron's Fermi velocity) and conclude that the self-energy makes a major contribution (60%) to the dispersive behavior of the D and 2D bands. The estimate of q is based on a concept of shifted Dirac cones in which the resonance decay of a phonon satisfying q >omega/v (omega is the phonon frequency) into an electron-hole pair is suppressed when mu < (<(h)over bar>vq - h omega)/2. We highlight the fact that the decay of an intervalley (and intravalley longitudinal optical) phonon with q = omega/v is strongly suppressed by electron-phonon coupling at an arbitrary mu. This feature is in contrast with the divergent behavior of an intravalley transverse optical phonon, which bears a close similarity to the polarization function relevant to plasmons.

Boron and nitrogen-incorporated graphene thin films were grown on polycrystalline Ni substrates by thermal chemical vapor deposition using separate boron- and nitrogen-containing feedstocks. Boron and nitrogen atoms were incorporated in the film in almost equal amounts and the total content reached similar to 28%. The film predominantly consisted of separate graphene and boron nitride domains. Carrier concentration in the graphene domains was estimated to be about 1 x 10(-3) e/atom (3.8 x 10(12) cm(-2)) from G band shift in Raman spectra. (C) 2011 Elsevier B.V. All rights reserved.

Low-energy irradiation-induced conductivity decrease of a single-wall carbon nanotube has been well established experimentally. However, its origin is still controversial. Irradiation effects on suspended single-wall carbon nanotubes, which are much less affected by substrate charging effects, were studied to distinguish possible origins. The results indicate that the conductivity decrease is not caused by substrate charging, but by irradiation-induced defect formation. © 2011 The Surface Science Society of Japan.

Short-wavelength electroluminescence (EL) emission is observed from unipolar and ambipolar carbon nanotube field-effect transistors (CNFETs) under high bias voltage. EL measurements were carried out with an unsuspended single-walled carbon nanotube (SWNT) in high vacuum to prevent the oxidation damage Induced by current heating. Short-wavelength emission under high bias voltage is obtained because of the Schottky barrier reduction and the electric field Increase in a SWNT. The simultaneous measurements of transport and EL spectra revealed the excitation mechanism of impact excitation or electron and hole injection dependent on the conduction type of unipolar or ambipolar characteristics. In addition to the EL emission, blackbody radiation was also observed in a p-type CNFET. Taking into account the device temperature estimated from blackbody radiation, the contribution of impact excitation and thermal effect to the exciton production rate was evaluated.

InAs semiconductor nanowires are expected to have applications in high-mobility nanoelectronics. Understanding the dopant distribution will be critical for the fabrication of high-performance devices based on nanowires. We study the n-type doping of InAs nanowires, using Si2H6 as a doping precursor, and clarify the predominant Si doping through Au catalyst particles. Using a series of segments in a single nanowire with a tapered shape and corresponding nanowire-channel field effect transistor characteristics, we show that the dopant atom incorporates predominantly via the Au-catalyzed vapor-liquid-solid mode, which accompanies the shell growth in the vapor-phase epitaxy mode. We determine the electrically active clasping concentrations via the Au-catalyzed vapor-liquid-solid and vapor-phase epitaxy modes in the InAs nanowire to be 1.37 x 10(18) and 1.57 x 10(17) cm(-3) under a In/Si source flow ratio of 600. This work developed a new method for the characterization of dopant distribution in semiconductor nanowires and the result provides more opportunities for the formation of modulation-doped core shell nanowires and novel nanostructures.

Few-layer graphene films were grown from spin-coated polystyrene films deposited on polycrystalline Ni substrates simply by heating in an Ar atmosphere. In this method, it is not necessary to use any reactive gas. Instead, the graphene film thickness is simply controlled by the initial thickness of the polystyrene film. This method of growing few-layer graphene film is safer and uses a much simpler apparatus than thermal chemical vapor deposition. (C) 2011 The Japan Society of Applied Physics

Single-wall carbon nanotubes were grown by thermal chemical vapor deposition using either boron-or nitrogen-containing feedstocks or both. Carrier doping was evidenced by hardenings of the G band in Raman spectra, and the estimated carrier concentration reached similar to 0.4%. In the G' and D band spectra, a doping-induced component was observed at the high- or low-energy side of the original one. However, the appearance of the new component did not always coincide with the carrier doping. The doped SWCNTs often show radial breathing mode peaks in the off-resonance region, indicating a defect-induced modification of absorption spectrum. (C) 2011 Elsevier Ltd. All rights reserved.

Graphene-like thin films were grown on patterned SiO2 substrates by simple thermal chemical vapor deposition using ethanol. The film growth occurred preferentially in the vicinity of pattern edges. Catalytic metal is not necessary for the substrate or the pattern. The films consist of graphitic nanocrystals of several nanometer scale. In the electric properties, the field effect is observed at room temperature. (c) 2010 Elsevier B.V. All rights reserved.

We analyzed defects created by low-energy irradiation in single-wall carbon nanotubes (SWCNTs) using Raman spectroscopy. The analysis is based on the recovery curves of the G/D ratio and there is no need to assume a specific functional form between the G/D ratio and the defect density. The obtained activation energies of defect healing are 0.7 or 1.4 eV, depending on the extent of the damage, which are close to the values for recombination barriers of vacancy-adatom defects. Calculated recovery curves of the G/D ratio at room temperature show that the recovery is so slow that almost no recovery is observed in a usual time scale, which is consistent with experimental results. (C) 2010 Elsevier Ltd. All rights reserved.

We synthesized suspended single-walled carbon nanotubes (SWNTs) by the thermal chemical vapor deposition method and functionalized them with Au nanoparticles (NPs). We used 3-(aminopropyl) triethoxysilane as a linker and controlled the Au NP density on the SWNT surface by changing the reaction time. In the Raman scattering spectra of the Au-functionalized SWNTs, an enhanced peak frequency and peak intensity were observed in the non-resonant region. A significant enhancement of the metallic character in the high frequency region was also observed, especially when we used a 633 nm laser. By measuring the electric properties using a standard field effect transistor configuration, we found that charge transfer occurred during the functionalization processes. It is expected that the charge transfer related optical enhancement may affect the observed change in the Raman profiles.

We have fabricated single-walled carbon nanotube (SWNT) quantum dot device with local multi-back gates, in which a SWNT is not surrounded by an insulator or gate electrodes. The charge states of multi-quantum dots, which are separated by an intrinsic defect of a SWNT, can be independently controlled by applying two local back gates. The charge stability diagram changes depending on the gate voltage range, which changes the interdot coupling between two dots. Furthermore, a honeycomb charge stability diagram corresponding to an intermediately coupled double-quantum dot is also observed. (C) 2009 The Japan Society of Applied Physics

Low-energy electron and photon irradiation cause damage in single-walled carbon nanotubes. In this work, irradiation effects of photons (hv < 20eV) in an ultra-high vacuum were systematically studied. The threshold energy of the low-energy irradiation damage was evaluated to be about 6 eV. Less damage was observed at 8 eV, which seems to be due to a small optical absorption coefficient at that energy.

To investigate the effect of the oxide layer on the oxide-mediated growth in solid phase epitaxy for the formation of a single-phase iron disilicides, the annealing processes were analyzed using spectroscopic photoemission and low-energy electron microscopy for a special surface where oxide areas and clean substrate areas (voids) coexist closely in a micrometer-order view. From the analysis of high-resolution X-ray adsorption near edge structure (XANES), we found that a pure alpha phase of FeSi2 was obtained in the oxide area after annealing at 720 degrees C, although a mixture of its alpha and beta phases was obtained in the void area. This indicates that the oxide layer effectively worked and a single-phase alpha-FeSi2 was successfully formed by the oxide-mediated growth. Copyright (C) 2008 John Wiley & Sons, Ltd.

A new ordered fullerene phase encapsulated by large-diameter CNTs is systematically investigated by combining a growth technique by chemical vapour deposition, high-resolution transmission electron microscopy and molecular-dynamics simulations. In contrast to fullerenes in smaller (1-2 nm) diameter CNTs, where fullerenes are packed in linear or helical chains, fullerenes form a nanoscale cylinder in double-walled CNTs with diameters of similar to 4 nm. The fullerenes were shown to form a nanocylinder with a side wall that resembled the (111) plane of solid C-60. This ordered phase is different from peapods or fullerene solids known so far, and a result of the interaction between the CNT wall and fullerenes. This finding will open up a new field of fullerene science.

We demonstrate that nanosized Au particles have carbon solubility. Au-catalyzed carbon material growth by chemical vapor deposition undergoes a structural change, either a carbon nanowire or a single-walled carbon nanotube, depending on the catalyst particle size. This carbon material growth from Au is derived by the formation of Au - C eutectic nianosized alloy.

Effects of local low-energy irradiation on the electric properties of metallic single-walled carbon nanotubes were studied. Defects formed by 20 keV-electron irradiation in an electron beam lithography system converted the room-temperature electric properties to p-type or ambipolar semiconducting. Coulomb oscillation was also observed at room temperature. The results also suggest that electric measurements are inconclusive for distinguishing whether a nanotube is metallic or semiconducting.

The authors report size control of catalytic nanoparticles by thermal annealing for diameter-controlled growth of single-walled carbon nanotubes (SWNTs). They found that Co nanoparticle-size gradually decreased through repetitive annealing at 1000 degrees C in Ar ambient. Results of x-ray photoelectron spectroscopy and secondary ion mass spectroscopy show that thermal evaporation is responsible for the decrease. After SWNT growth using this phenomenon, the authors found that thinner SWNTs with a narrower diameter distribution grew as the nanoparticles became smaller. Their results provide a rational and straightforward technique to prepare catalysts having a desirable size and uniformity toward diameter-controlled SWNT growth. (c) 2007 American Institute of Physics.

The experimental creation and annihilation of defects on single-walled carbon nanotubes (SWCNT) with the tip of a scanning tunneling microscope are reported. The technique used to manipulate the wall structure of a nanotube at the atomic scale consists of a voltage ramp applied at constant tunneling current between the tip and the nanotube adsorbed on a gold substrate. While topographic images show an interference pattern at the defect position, spatially resolved tunneling spectroscopy reveals the presence of localized states in the band gap of the nanotube. Removal of the defect by the same procedure demonstrates the reversibility of the process. Such a precise control in the local modification of the nanotube wall opens up new opportunities to tailor SWCNT electronic properties at will.

We succeeded in growing carbon nanotubes in a photoelectron spectroscopy analysis system using thermal chemical vapor deposition and analyzed the chemical states of the Co catalysts by in situ X-ray photoelectron spectroscopy before and after the growth. The carbon nanotubes were grown in an ethanol ambient at about 4 x 10(-2) Torr, a relatively low pressure that was used because we needed to subsequently measure photoelectrons under ultrahigh vacuum. We found that almost all of the Co particles are metallic after the growth. This shows that the metallic state is stable for Co under low-pressure ethanol ambient in our growth condition for carbon nanotubes.

Iron nanoparticles derived from DNA-binding proteins from starved cells (Dps) were used to grow single-walled carbon nanotubes (SWCNTs) with narrow diameter distribution. An atomic force microscopy, Raman spectroscopy, and photoluminescence were used for evaluation of diameter or chirality distribution of the SWCNTs. We found that thin SWCNTs (1.1 nm diameter) were grown from the large Dps-derived nanoparticles (2.4 nm diameter) on and above the substrates. From the size comparison with ferritins and Co-filled apoferritins, we also found that SWCNTs become thinner as the catalyst becomes smaller. The synthesis of smaller catalysts (ca. 1 nm diameter) and their use for growth becomes crucial for the control of SWCNT diameter. (c) 2007 Elsevier Ltd. All rights reserved.

Low-energy electron irradiation causes damage in single-walled carbon nanotubes and changes the electric behavior of a nanotube field-effect transistor from metallic to semiconducting at low temperature. The irradiation damage was found to form an energy barrier of several 10 meV in the nanotube channel. We show that the transition behavior can be reasonably explained by the barrier formation and gate-induced band bending. (c) 2007 American Institute of Physics.

We show that low-energy (20 eV-20 keV) electron or photon irradiation extinguishes the characteristic physical and chemical properties of single-walled carbon nanotubes, indicating that the irradiation damages the nanotubes. The irradiation-induced defects convert the electric properties of metallic SWNTs to semiconducting, and the nominal bandgap can be tuned simply by the irradiation dose. The defects also have the following interesting properties. The damage and recovery are reversible, indicating that the number of carbon atoms is preserved. The damage and recovery strongly depend on the diameter, suggesting that the damage is prominent in a rolled up graphene sheet, but not in a planar one. The activation energy of the defect healing is so small, depending on the diameter, that the defects can be healed even at room temperature or below.

Single-walled carbon nanotubes are damaged by low-energy electron and photon irradiation, depending strongly on the diameter. In this study, the formation and healing of the irradiation-induced defects were found to be in competition during irradiation even at room temperature or below. The diameter dependence of the damage can be mainly ascribed to a diameter dependence of the activation energy of the defect healing. The activation energy was estimated to be about 1 eV.

Nanoscale metal catalysts have been indispensable for carbon nanotube (CNT) synthesis by chemical vapor deposition (CVD). We show that even semiconductor nanoparicles of SiC, Ge, and Si produce single-walled and double-walled CNTs in CVD with ethanol. This implies that nanosize structures might act as a template for the formation of CNT caps composed of five- and six-membered rings. Providing a template for cap formation is the essential role of the catalysts.

We have fabricated field-effect transistors (FETs) with pristine and potassium-encapsulated single-walled carbon nanotube (SWNT) films, and the effects of potassium encapsulation are investigated. The transformation from a unipolar characteristic to an ambipolar characteristic by potassium encapsulation is observed from the measurement of the gate voltage dependence (V-gs)of the current (I) for SWNT-film FETs. This result indicates that the potassium encapsulation into SWNTs causes band gap narrowing. In addition, the n-type region of the I-V-gs curve is expanded during annealing of the devices; electron transfer from potassium to SWNTs occurs owing to the removal of the adsorbates. The adsorbate removal is confirmed by photoemission spectroscopy measurement. The FET with an individual potassium-encapsulated SWNT shows an ambipolar characteristic.

To clarify how Co reacts with a SiO2 layer on Si substrates, the annealing processes were observed using low energy electron microscopy (LEEM) and photoemission electron microscopy (PEEM). LEEM results using patterned substrate showed that Co silicide nano-dots form and that the Co diffusion length on the SiO2 layer is very small. From PEEM observations of a SiO2 layer with voids on Si substrate, we found that Co atoms in the void area, remain because of the formation of silicides, but those on the SiO2 surface disappear because metallic Co atoms easily diffuse. Co diffusion through the SiO2 layer to the Si substrate can compatibly explain these two results. © 2006 The Surface Science Society of Japan.

We demonstrate that any metal, even gold, silver, and copper, can act as a catalyst for SWCNT synthesis in chemical vapor deposition (CVD). Metal nanoparticles 3 nm or less in diameter, introduced into CVD ambience immediately after heat treatment at 800-950 degrees C in air, produce SWCNTs. The activation method is effective for copper and various noble metals as well as for iron-family elements. This implies that any metal particle may produce SWCNTs when its size becomes 1-3 nm. In other words, carbon atoms can form SWCNTs in a self-assembling fashion on nanoparticles without the specific functions of iron-family elements.

In this review we will focus our attention on the characterization of carbon nanotubes by X-ray photoelectron spectroscopy. In contrast to other spectroscopic techniques, photoelectron spectroscopy allows to obtain information on the overall electronic structure of the sample in a wide energy range, which makes it a unique technique. We will discuss the most recent and the most significant results on the intrinsic electronic properties of carbon nanotubes obtained with photoelectron spectroscopy. Furthermore, we will discuss in detail the application of photoelectron microscopy, i.e. photoelectron spectroscopy with high lateral resolution, to the study of carbon nanotubes. This technique allows to obtain laterally resolved spectroscopic information from ensembles of carbon nanotubes, but it even allows to perform photoelectron spectroscopy on single carbon nanotubes. (c) 2006 Elsevier Ltd. All rights reserved.

We have developed an in situ observation technique for chemical vapor deposition (CVD) of single-walled carbon nanotubes (SWNTs) using a scanning electron microscope (SEM). The growth of SWNTs in the SEM is achieved by employing low-pressure ethanol vapor as the carbon source and cobalt as the catalyst. The time evolution of SWNT growth is successfully traced by alternate CVD growth in low vacuum and SEM observation in high vacuum. By limiting the nanotube growth areas, individual SWNT growth could be observed. Copyright (C) 2006 John Wiley & Sons, Ltd.

Low-energy photon irradiation in an ultra-high vacuum, as well as low-energy electron irradiation, was found to damage single-walled carbon nanotubes, meaning that electronic excitations are solely responsible for the defect formation. The formation and healing of the defects were found to strongly depend on nanotube diameter; that is, thinner nanotubes are more easily damaged and hardly recovered. The results mean that the irradiation damage is prominent in a rolled up graphene sheet, but not in a planar one. The curvature-induced strain energy seems to be essentially important for the damage. (c) 2006 Elsevier B.V. All rights reserved.

The change in the size of iron nanoparticles before and after single-walled carbon nanotubes (SWNTs) growth was systematically investigated for the first time using atomic force microscopy and transmission electron microscopy. In contrast to the conventional growth model of SWNTs, we found that thin SWNTs grow from large catalytic nanoparticles and all particles are finally embedded in SiO2 substrates after thermal chemical vapor deposition. The present results provide useful information for achieving diameter-con trolled growth of SWNTs. (c) 2006 Elsevier B.V. All rights reserved.

Individually isolated single-walled carbon nanotubes (SWNTs) have been successfully grown vertically on a substrate by chemical vapor deposition with methane and Fe or Co catalysts. Vertical growth is obtained when the growth temperature was high, 900-1000 degrees C, and the tube diameter is large, 2-5 nm. Vertically grown SWNTs are short, ranging from several tens to 300 nm, which are useful for use as, tips of a field emission or probe microscope. Fullerene encapsulation has been achieved directly in vertical SWNTs on a substrate.

A simple method for spatially selective removal of single-walled carbon nanotubes is demonstrated. This method is based on low-acceleration-voltage electron irradiation damage and consists of local electron irradiation of nanotubes and annealing in air. The irradiation damage seems to follow excitation of valence electrons.

Individual suspended single-walled carbon nanotubes were observed by means of C Is and secondary electron photoemission electron microscopy. No band bending was observed in the images, suggesting that the depletion width near the catalytic Fe/nanotube or substrate Si/nanotube contact is comparable to the spatial resolution of 40 nm or less. Work function differences between the nanotubes were clearly observed in secondary electron images. The work functions of 93 SWNTs were found to range within 0.6 eV, but most distributed in a much narrower energy range of 0.2 eV. (c) 2005 Elsevier B.V. All rights reserved.

We report the effect of low-energy (1 keV) electron beam irradiation on gated, three-terminal devices constructed from metallic single-walled carbon nanotubes. Pristine devices, which exhibited negligible gate voltage response at room temperature and metallic single-electron transistor characteristics at low temperatures, when exposed to an electron beam, exhibited ambipolar field effect transistor (room temperature) and single-electron transistor (low temperature) characteristics. This metal-semiconductor transition is attributed to inhomogeneous electric fields arising from charging during electron irradiation.

Direct imaging of single-walled carbon nanotubes ( SWNTs) suspended on pillar-patterned Si or SiO2 substrates is investigated using transmission electron microscopy ( TEM) and scanning electron microscopy ( SEM). The suspended nanotubes are successfully observed by direct TEM imaging and it is seen that they have either individual or bundles of SWNTs. Low energy ( <= 2 keV) SEM produces high contrast images of suspended SWNTs. On the contrary, when SWNTs contact a SiO2 substrate, they are imaged using electron-beam induced current. The image brightness depends on the length of SWNTs.Direct imaging of single-walled carbon nanotubes (SWNTs) suspended on pillar-patterned Si or SiO2 substrates is investigated using transmission electron microscopy (TEM) and scanning electron microscopy (SEM). The suspended nanotubes are successfully observed by direct TEM imaging and it is seen that they have either individual or bundles of SWNTs. Low energy (<= 2 keV) SEM produces high contrast images of suspended SWNTs. On the contrary, when SWNTs contact a SiO2 substrate, they are imaged using electron-beam induced current. The image brightness depends on the length of SWNTs.

Single-walled carbon nanotubes (SWNTs) were synthesized by chemical vapor deposition (CVD) using catalytic nanoparticles both on the substrates and above the substrates in order to investigate the effect of nanoparticle density on diameter-controlled SWNT growth. As the density of the catalytic nanoparticles increased, tube-diameter distribution broadened and the diameter itself also increased. SWNTs observed in this study were grown by the base-growth mechanism and their diameters were much smaller than those of the nanoparticles. Based on elaborate diameter measurements, we reasonably conjecture that the time evolution of catalytic nanoparticles during CVD growth can explain these large size differences. (c) 2005 American Institute of Physics.

Effects of low-acceleration-voltage (20kV) electron irradiation damage in single-walled carbon nanotubes (SWNTs) were studied by in-situ electric measurements. The irradiation drastically decreased the conductivity for both metallic and semiconducting nanotubes due to the irradiation-induced damage. Intensive electron irradiation made the SWNTs almost insulating. This phenomenon could be utilized to fabricate nanotube-based electric networks.