avto-zao.ru/includes/como/1003.php With further increase of the N2 flow rate, branching nanowalls became noticeable and height uniformity deteriorated gradually, as shown in Fig. Hall measurements were carried out for the carbon nanowall films grown by the addition of N2 at various flow rates. The Hall coefficient of the undoped carbon nanowall film was positive, while that of the N-doped carbon nanowall film was negative.
On the other hand, in the case of the carbon nanowall film grown by the N2 addition, the N-doped carbon nanowall films exhibited n-type conduction, suggesting that nitrogen was included in the carbon nanowalls and acted as a donor. On the other hand, in the case of the Ndoped carbon nanowall films, the resistivity decreased drastically at first. The carrier concentration and Hall mobility of the undoped and N-doped carbon nanowall films were derived from the results of the Hall measurements.
For determining the carrier concentration, the scattering factor was assumed to be 1.
As a result of the N2 addition at a flow rate of 1 sccm, the conduction type of the carbon nanowall film changed to n type. In the case of n-type N-doped carbon nanowall films, the carrier concentration increased with an increase in the N2 flow rate. The carrier concentration behavior of the n-type N-doped carbon nanowall films is considerably similar to that of the N content in the carbon nanowalls.
The mobility of the undoped carbon nanowall film was low. As a result of the N2 addition at a flow rate of 1 sccm, the mobility of the N-doped carbon nanowall film increased twice as much as that of the undoped carbon nanowall film. The carrier transport in graphene occurs in the orbitals perpendicular to the surface. Meanwhile, an individual carbon nanowall is composed of nanodomains that are a few tens of nanometers in size . Moreover, each carbon nanowall has many edges and defects, which was indirectly confirmed by the Raman spectroscopy. In the case of the undoped carbon nanowalls, the major conduction carrier is presumably a positive hole because mobile p electrons would be easily trapped by the defect.
On the other hand, as reported by Shimoyama et al. In this case, undoped carbon nanowall film exhibits p-type conduction, which is characterized by high resistivity and low mobility probably due to the high 4. As a result of the N inclusion at a low flow rate of 1 sccm during the growth process, the conduction type of the carbon nanowalls changed to n-type and the carrier concentration decreased; however, the mobility increased.
This transition suggests that some N atoms substituted C atoms, and others infilled the defects, resulting in the compensation of holes as well as reduction in defects to improve the crystallinity.
With a further increase in the N inclusion, a less aligned and highly branching carbon nanowall film was obtained. Therefore, on the contrary, the Hall mobility decreased due to the slight deterioration in the crystallinity, while the carrier concentration increased. Since the electrical transport inside the complex network of carbon nanowalls has not been fully understood, the intrinsic property of carbon nanowall 10—30 layers of graphene sheets cannot be derived from the result of Hall measurement at present, while single graphene sheet reportedly possesses semiconductor characteristics.
The carrier concentration and mobility derived from the Hall measurements shown in Fig. Therefore, in order to fully exploit the potential of graphene and to develop nextgeneration electronic devices, reliable methods are required for fabricating graphene sheets with high crystallinity. In Sect. In contrast, the addition of small amount of CO2 or H2O to the hydrocarbon plasma is effective to etch disordered carbon species [23, 24]. Carbon—oxygen reactions on various kinds of carbon surfaces have been studied .
In some cases, the trigonal bond of carbon at the edges of basal planes of the crystallites and defects within the basal plane act as nucleation sites for undesirable nanostructured carbon resulting in the formation of blanched structure. These nucleation sites can be eliminated by the reaction with atomic oxygen to form volatile by-products such as CO and CO2. Thus, using carbon—oxygen reactions would enable us to synthesize carbon materials that have larger crystallite sizes, higher crystallite alignment, and higher purity.
The C2F6 and H2 flow rates were maintained at 50 and sccm, respectively.
Additional O2 gas was introduced into the capacitive coupled plasma CCP region at flow rates of 0—5 sccm, and the total pressure was maintained at Pa. The carbon nanowalls grown with O2 addition exhibit less branching than those produced without O2. The carbon nanowall film grown with O2 had larger plane sheets with wider interspaces than that grown without O2. The O2 addition is effective in improving the crystallinity of N-doped carbon nanowalls. However, by introducing small amount of O2, the crystallinity of N-doped carbon nanowalls was also improved and monolithic sheet structure could be obtained as shown in Fig.
Low-magnification cross-sectional TEM images of carbon nanowall films grown without and with O2 are shown in Fig. In contrast, monolithic self-sustaining graphene sheets larger than nm in size were clearly observed in the carbon nanowalls grown with O2 Fig. KGaA a b c d D Intensity arb. On the other hand, a highly orientated, smooth multi-layered graphene structure was clearly obtained in the carbon nanowall grown with O2 Fig. Carbon nanowalls grown without and with O2 gas addition were also characterized by Raman spectroscopy to investigate the influence of O2 gas addition on the structural property of carbon nanowalls.
Raman spectra for the deposits were measured at room temperature with a triple monochromator Jobin Yvon, Ramanor T using the Moreover, based on comparisons between Raman spectra b and d , it is confirmed that the O2 addition is effective for improving the crystallinity of N-doped carbon nanowalls. The crystallinity of carbon nanowalls was analyzed using synchrotron X-ray surface diffraction at grazing incidence and exit at the beamline BL13XU of SPring-8 .
KGaA 4. The average thickness of the carbon nanowalls was determined from the mean crystallite size Lc , which is calculated from the peak width using the Scherrer equation .
The values of Lc for carbon nanowalls grown without and with O2 were estimated to be 9. A quartz substrate was used to evaluate the electrical properties of the carbon nanowall films. After synthesizing the carbon nanowall films on a quartz substrate, four aluminum Al contacts were symmetrically positioned on the carbon nanowall film for Hall measurements by the van der Pauw method . The Hall coefficients of the carbon nanowall film samples grown with and without O2 were measured in the same manner as described in the last section. The resistivity of all the carbon nanowall films decreased with an increase in the measured temperature, indicating the semiconductor behavior of the carbon nanowall films.
The regions I and II in Fig.
Thus, the carrier generation mechanism for carbon nanowall films grown with and without O2 is considered to be the same. However, the carrier concentration of carbon nanowall film grown with O2 was slightly higher than that of carbon nanowall film grown without additional gas. The band gap can be obtained from the slope of the intrinsic range I in Fig. Thus, the band gap of carbon nanowall film was estimated from the region I at K, which was approximately 80 meV for both carbon nanowall films grown without and with O2 addition.
However, the intrinsic region was not clearly observed in Fig. The slope in region I for the measured curve in Fig. Therefore, the band gap of carbon nanowall film was expected larger than 80 meV. The electrical conduction of graphite with a fluorine junction CF n was found to be disturbed since the excess charge induced by the F atoms reduces the graphite p density .
Watari, T. Figure 2. Horaguchi, M. In addition, by probing water molecules at the oxygen K-edge, the different hydrogen bond networks among three different carbon dots dispersions have been compared. Several nanorods are connected to each other forming a macroporous network in submicron size, which is preserved through carbonization. Hence, to improve the exfoliating efficiency of graphene, we prepared graphene by solvent exfoliation of graphite powder in N-methylpyrrolidone NMP with the assistance of three sodium salts, sodium citrate Na3C6H5O7 , sodium phosphate Na3PO4 and sodium pyrophosphate Na4P2O7 , respectively. The ripple effects in the local economy are likely to be extremely significant given the sheer number of local industries that can take advantage of these innovative new materials.
The bent multi-layered graphene structures shown in Fig. These residual F atoms act as impurities and F-induced defects in carbon nanowalls References 79 will affect electrical conduction. This suggests that oxygen etch F atoms and small graphitic fragments, thereby contributing to the higher graphitization and improving the crystallinity and electrical conduction to form highly oriented monolithic graphene sheets. The crystallinity of vertical graphene sheets was improved by introducing O2 into the plasma used for carbon nanowall growth.
The resistivity of the carbon nanowall film decreased with an increase in the temperature, indicating that the carbon nanowall films exhibit semiconductor behavior. The current results demonstrate that carbon nanowalls consisting of vertical, monolithic self-sustaining nanographene sheets have great potential for application in next-generation electronic devices.
However, the resistivity and carrier concentration derived in this section reflect the electrical property of the bulk carbon nanowall film comprising the web of nanographene sheets with interspaces. These values could be useful for the design and evaluation of electronic devices using bulk carbon nanowall films. On the other hand, the carrier concentration and mobility as well as the band gap inside the individual monolithic carbon nanowall sheets are of great interest for realizing nanographene devices. Jpn J Appl Phys — 3. Takashima S, Hori M, Goto T, Kono A, Ito M, Yoneda K Vacuum ultraviolet absorption spectroscopy employing a microdiacharge hollow-cathode lamp for absolute density measurements of hydrogen atoms in reactive plasmas.
Appl Phys Lett — 4.
J Appl Phys — 5. Sugai H, Toyoda H Appearance mass spectrometry of neutral radicals in radio frequency plasmas. J Vac Sci Technol A — 6. J Appl Phys — 8. Thin Solid Films — — Diam Relat Mater 4: — Shimoyama I, Wu G, Sekiguchi T, Baba YJ Evidence for the existence of nitrogensubstituted graphite structure by polarization dependence of near-edge x-ray-absorption fine structure.
Phys Rev B R—R Van der Pauw LJ A method of measuring specific resistivity and Hall effect of discs of arbitrary shape. Philips Res Repts 1—9 Phys Rev B — Chapter 5 Growth Mechanism of Carbon Nanowalls In the case of film formation using plasma-enhanced chemical vapor deposition CVD , high performance can be achieved by 1 the selective production of specific reactive species crucial for the film growth and nucleation, 2 the efficient transport of important species onto the growing surface, and 3 the control of surface reaction for both nucleation and subsequent growth.
In the case of carbon nanostructure fabrication, it is important to elucidate the specific species such as carbon-containing radicals and hydrogen atoms that contribute to the growth and then determine the morphology of the nanostructures. Moreover, on the basis of the knowledge of the species, it is necessary to control the process plasma in order to obtain carbon nanostructures with structure and morphology customized for a specific application.
This chapter addresses issues on the growth mechanism of carbon nanowalls. Examples of radical density measurements in the plasma are described in the beginning. Then, the growth mechanisms of carbon nanowalls in the steady-state growth and nucleation stage are discussed. Furthermore, carbon nanowalls are synthesized using multi-beam CVD system consisting of ion, fluorocarbon radical, and H radical sources, and the role of ion bombardment for the nucleation of carbon nanowalls is discussed.
In this section, density measurements of carbon-containing species and H atoms in the plasmas used for the CVD for carbon nanostructure fabrication are presented. Figure 5. The flow rates of CH4 and H2 were 50 and 70 sccm, respectively. The total pressure was 70 Torr and microwave power was W. The C2 radical density at the lowest excited state a3Pu Fig. The optical emission intensity of the 0,0 bandhead of the C2 Swan system was also measured. Ha C2 Hb Fig. The absorption coefficient was obtained by dividing the negative of natural logarithm of the transmittance by an absorption pass length 33 cm , which is approximately equal to a three-fold length of inner diameter of the cylindrical stainless steel chamber.
Due to the overlapping of rotational lines at the bandhead at approximately nm as shown in Fig. The C2 radical density increased with increasing CH4 concentration. By further increasing the CH4 concentration 5. It is apparent that the optical emission intensity correlated linearly with C2 radical density at moderate concentrations of CH4.
This linear correlation is consistent with the result in hydrogendeficient plasma-enhanced CVD . The flow rates of C2F6 and H2 were 50 and sccm, respectively. The total pressure was Pa. The UV emission band between and nm represents CF2 radical transition, while the visible emission band around nm is due to the transitions of CF3 radical states .
Emissions from hydrogen atomic lines Ha and Hb and band from CH radical H atoms are considered to be important species for the abstraction of fluorine from CFx radicals migrating on the growing surface or edge of a graphene layer and removal of undesirable amorphous phases. In the case of the measurement of absolute H radical density, a H2 microdischarge hollow-cathode lamp H-MHCL was used as a vacuum ultraviolet VUV light source for absorption spectroscopy.
The transition line used for the absorption measurement was Lyman alpha La at At an ICP power of W, where definite typical carbon nanowalls were fabricated, the radical density ratio of H to CF3 increased by approximately five times compared with that of the case without H radical injection. Carbon C atoms can be produced in the gas phase by electron-impact dissociation of carbon-containing molecules or radicals and are considered to be extremely reactive in the gas phase, and their sticking probability at the surface is expected to be close to unity.
Therefore, C atoms likely play important roles in the mechanism of carbon nanostructure formation. Since the wavelength resolution of the VUV monochromator was 0. The height of carbon nanowall film was derived from cross-sectional SEM observations.
On the other hand, the height of the carbon nanowall films decreased with an 88 5 Growth Mechanism of Carbon Nanowalls Fig. At a low total pressure of On the other hand, the carbon nanowall film grown at a total pressure of 80 Pa had wide interspaces of 30—40 nm Fig. From the Raman spectroscopy, with an increase in the total pressure, the G band peak was clearly observed and its intensity increased, while the intensity of the D0 band decreased data not shown.
These results indicate that an excess amount of H atoms relative to the C atoms enhances the etching of undesirable amorphous phases, resulting in higher graphitization and the formation of well-defined carbon nanowalls at a reduced growth rate. Therefore, both CH3 and C2 radicals might act as precursors for the formation of carbon nanowalls. In contrast, it is believed that H radicals play an important role in diamond film growth by etching non-diamond phase preferentially due to the different etch rates of amorphous, sp2and sp3 hybridized carbons.
The importance of hydrogen as a process component is supported by the several process conditions associated with carbon nanowalls and related carbon nanostructures reported in the literature [16, 18—20]. These plasmas are useful to produce H radicals effectively. Meanwhile, the CF3 radical density in the C4F8 plasma is considered to be low, although large amounts of CF2 radicals would be generated in the C4F8 plasma by electron-impact dissociation from C4F8, due to the cyclic structure of the C4F8 molecule [21, 22].
Reactive CF2 radicals are believed to be the direct precursors for the formation of fluorocarbon polymers [23, 24]. The sticking coefficient of C atoms to the surface is reported to be about 0. The surface morphology of the carbon nanowalls was strongly influenced not only by the C atoms but also by the H atoms.
However, the morphology and spacing between carbon nanowalls are considered to be determined originally in the nucleation stage of carbon nanowalls; the radical densities of H and CFx, as well as ion densities and their energies arriving on the surface at the very early stages of nucleation, must be responsible for determining the morphology of carbon nanowalls.
The effects of ion bombardment on nucleation will be discussed later in this chapter. The SEM images of typical carbon nanowalls grown for 3 h are shown in Fig. As the growth time increased, spread vertical nanowalls met with one another, eventually resulting in the formation of linked nanowalls like a maze Fig.
With further increase of growth time, the height of carbon nanowalls increased, while the thickness of nanowalls and the spacing between nanowalls became almost saturated while keeping the morphology of carbon nanowalls, as shown in Fig. In addition, a variation of averaged maximum spacing between adjacent nanowalls was indicated by a broken line . A broken line indicates the variation of the averaged maximum spacing between adjacent nanowalls estimated from SEM observation  — reproduced with permission from American Institute of Physics Wall height and thickness nm 92 Height Average space 0 Thickness 0 2 4 6 8 Growth time hours 10 As shown in Fig.
On the other hand, the thickness of nanowalls indicated by closed circles increased gradually at first up to 3 h, thereafter became constant at about 50 nm. The spacing between nanowalls increased at first, then became almost constant as well. After the nucleation stage of carbon nanowalls, growth of less-aligned, isolated carbon sheets with a semicircular shape standing on the substrate is confirmed as shown in Fig.
As the growth time increased, density of isolated nanosheets increased and those standing almost vertically on the substrate continued preferably to spread faster. Then, spreading nanosheets met one another; eventually resulting in the formation of linked nanowalls as shown in Fig. With the further increase of growth time, the height of vertically aligned carbon nanowalls increased, while the thickness of nanowalls and the spacing between nanowalls became almost saturated with the morphology of carbon nanowalls, as shown in Fig. After the nucleation stage of carbon nanowalls, almost vertically aligned, isolated carbon nanosheets standing on the substrate met one another in 10 min, eventually resulting in the formation of linked nanowalls as shown in Fig.
In the case of carbon nanowalls grown for 30 min shown in Fig. On the other hand, the density of carbon nanowalls decreased; the spacing between nanowalls increased by 5—10 times, compared with the carbon nanowalls grown for 10 min. With the further increase of growth time, the height of vertical aligned carbon nanowalls increased almost linearly, while the spacing between nanowalls increased very gradually with the morphology of carbon nanowalls, as shown in Fig.
It is noted that stunted nanowalls at the base of carbon nanowall film were observed in the cross-sectional SEM image of Fig.
It was found that carbon nanowalls with a height of approximately nm remained in the ripped area and their interspaces were narrower. Most of the carbon nanowalls remaining in the ripped area would evidence the existence of the stunted ones that their growth had been terminated at the early stage of growth. In addition, a variation of averaged maximum spacing between adjacent nanowalls was indicated by a dotted line.
The average spacing between adjacent nanowalls increased at first, and then became almost saturated at approximately nm. A dotted line indicates the variation of the averaged maximum spacing between adjacent nanowalls estimated from SEM observation 5. SEM images of carbon nanowall films grown for 3 and 30 min are shown in Fig. On the other hand, vertically standing carbon nanowalls with uniform height were fabricated in 30 min, as seen from Fig. In contrast, shortly after the commencement of growth up to 5 min, the carbon nanowall growth rate was lower than that in steady-state conditions.
At the nucleation stage, carbon species would condense to form nanoislands with dangling bonds Fig. At these dangling bonds, disordered carbon nanosheets of smaller sizes would be nucleated, followed by the two-dimensional growth and subsequent formation of nanographene sheets. Among the nucleated graphene sheets with random orientations, those standing almost vertically on the substrate continued preferably to grow up faster to vertically standing nanosheets owing to the difference in the growth rates along the strongly bonded planes of graphene sheets expanding and in the weakly bonded stacking direction.
Reactive carbon species arriving at the edge of the graphene layer are easily bonded to the edge, and eventually the graphene layer would expand preferably along the direction of radical diffusion, perpendicular to the electrode plane. On the other hand, lowlying inclined graphene sheets were shadowed by the high-grown vertical graphene 5. As a result, the amounts of reactive carbon species arriving at the low-lying inclined graphene sheets decreased, resulting in the termination of growth for the inclined smaller nanowalls.
As growth time increased, spreading vertical nanowalls met one another, eventually resulting in the formation of linked nanowalls similar to a maze. Therefore, with the increase of growth time in the early stage, the spacing between nanowalls at their top increased gradually, and then became almost saturated, resulting in the formation of two-dimensional carbon sheets standing vertically on the substrate with high aspect ratio. As results of above-mentioned growth experiment using various plasmas and source gas mixtures, the height of nanowalls increased almost linearly with keeping their morphology, as the growth time increased in the steady-state growth condition.
In contrast, when the behavior of growth rate for the carbon nanowalls in the very early stage of growth was investigated carefully, it was found that the growth rate of carbon nanowalls shortly after the commencement of growth up to 5 min was lower than that in steady-state conditions, as shown in Fig. Nucleation of carbon nanowalls is considered to occur during this period. Significant interest in this section exists in clarifying the nucleation mechanism of carbon nanowalls at the very early stage and controlling the growth of carbon nanowalls for obtaining the self-sustaining carbon nanowall with good crystallinity.
Here, carbon nanowalls were deposited using a fluorocarbon by VHF plasmaenhanced CVD with H radical injection, and the morphology and structure of deposits formed in the nucleation stage were investigated in detail. During the very early stage of nucleation Fig. In 1 min, the Si substrate is almost completely covered with nanoislands, resulting in the formation of a thin layer, as shown in Fig. The thickness and the surface roughness of this first layer are approximately 10 and 3 nm, respectively. At this moment, some small two-dimensional nanoflakes have started to form at the aggregations of nanoislands forming the first layer.
Subsequently, disordered carbon nanoflakes form [Fig. The number density 98 5 Growth Mechanism of Carbon Nanowalls b 1 min a 30 sec nm nm d 3 min c 2 min nm nm Fig. In 3 min, the wall structures grow preferentially in a vertical direction, while their number density, which was estimated to be approximately 5 per 10, nm2, was less than that at 2 min. Thus, vertical carbon nanowalls, which are a percentage of nucleated nanoflakes, grow continuously. A cross-sectional TEM image of carbon nanowalls synthesized for 30 min is shown in Fig.
In Fig. The interface layer thickness remains unchanged relative to that of the first layer formed in the nucleation stage. The carbon nanowall film was characterized by secondary ion mass spectrometry SIMS to investigate the atomic composition of carbon nanowalls and interface 5. The depth distributions of the relative atomic compositions of C, Si, and F were measured, as shown in Fig. The F signal was detected at the interface between Si substrate and carbon first layer according to the SIMS depth profile, suggesting that F atoms exist on the Si substrate surface, and not in the carbon interface layer.
The flow rates of N2 and H2 were 80 and 20 sccm, respectively, and the total gas pressure was kept at 6. The height of the etched carbon nanowall film was evaluated by SEM images and spectroscopic ellipsometry. Etch rate curves for the carbon nanowall bulk film samples with different heights were obtained by measuring the thickness of the remaining carbon nanowall film for differing amounts of etching time 0—1 min. As a result of etching, even in short period most of carbon nanowalls were removed. Since it was difficult to measure the thickness of remaining very thin layer by the cross-sectional SEM observation, the thickness of remaining materials after etching was evaluated by spectroscopic ellipsometry.
After etching for 20 s, the height of all carbon nanowall film samples decreased to 10 nm regardless of the initial height of carbon nanowall films. On the other hand, with further increase of etching time, the etch rate decreased drastically, and the height after etching for 60 s was approximately 7 nm in all samples. The height of 10 nm for the remaining layer corresponded to the thickness of the amorphous carbon interface layer formed between carbon nanowall film and Si substrate. The results indicate that the carbon nanowalls were rapidly etched away for less than 20 s, while the interface layer was etched slowly after the removal of carbon nanowalls.
The morphology of the surface of the interface layer undergoing etching exhibited the aggregation of nanoislands, which was completely different from that of the carbon nanowall film. In the case of etching the amorphous interface layer, with the increase of etching time, removal of nanoislands would proceed gradually. It is noted that in estimating the thickness of remaining interface layer by spectroscopic ellipsometry the interface layer was assumed to be a simple monolayer without voids instead of bumpy surface consisting of the aggregation of nanoislands.
Accordingly, the height of the etched carbon nanowalls estimated by the spectroscopic ellipsometry was an averaged value, corresponding to the thickness of the equivalent squashed material without voids and roughness.
Therefore, in terms of the film density, the decrease of height from 10 nm in 20 s to 7 nm in 60 s in Fig. After the synthesis of carbon nanowall film on the Si substrate, the carbon nanowall film was taken out of the main chamber into the atmosphere, and tetrafluoroethylene-related polymer film was attached to the top of carbon nanowall film. Then, polymer-covered carbon nanowall film on Si is dunked in concentrated nitric acid solution, resulting in the exfoliation of carbon nanowall film from the Si substrate in a few minutes.
After the exfoliation, the interface layer remained on the underside of the detached carbon nanowall film, and the interfacial surface was smooth and free from pinholes, suggesting that the chemical bonding between the interface layer and the base of carbon nanowalls is stronger than that between the interface layers and the Si substrate. In terms of physical strength of carbon nanowalls, the interface layer is 5.
In addition, in the case of attaching carbon nanowalls to the different materials, the interface layer would serve as an adhesion intermediate between carbon nanowalls and another material to be attached. In contrast, the addition of a small amount of O2 or H2O to hydrocarbon 5 Growth Mechanism of Carbon Nanowalls plasma is effective for etching disordered carbon species . In the case of carbon nanowall growth, its nucleation is influenced by carbon—oxygen reactions, resulting in the improvement of crystallinity of carbon nanowalls.
The morphologies and structures of deposits formed with O2 gas addition in the nucleation stage were investigated in the same manner as those produced without O2 described in the above section. No deposits were observed on the substrate at 30 s, as shown in Fig. Nucleation of nanoislands during the growth with O2 gas addition took longer than that for the growth without O2. There are approximately 20 nanoislands in a 10, nm2 area in Fig.
The number density of nanoislands increased after 2 min growth, but the fractional surface coverage was low and a distinct interface layer was not formed, as shown in Fig. At this stage, some small two-dimensional nanoflakes have started to grow at isolated nanoislands. After 3 min growth, as shown in Fig. Ex situ X-ray photoelectron spectroscopy XPS analysis was carried out to measure the atomic compositions of the deposits in the nucleation stage.
The O signal originates from the native oxide on the surface of Si substrate as well as the contaminant due to the exposure of the samples to laboratory atmosphere up to 2 min growth. As results from Figs. It is noted that at the very early stage of the nucleation the nanoislands were dotted on the substrate as shown in Fig. This discrepancy can be explained by the first formation of very thin fluorocarbon layer covering the Si surface before the start of nanoisland formation. Unfortunately, this thin fluorocarbon layer has not been detected by the cross-sectional TEM observation.
This value was 20 times higher than that for the sample formed without O2 shown in Fig. Furthermore, even after the deposition mode moved from the nucleation to the steady-state growth, the fraction of surface coverage was low and the definite interface layer was not formed; carbon nanowalls originated on the isolated nanoislands. It should be noted that oxygen was not included in the carbon nanowalls in the case of carbon nanowall synthesis with O2 addition.
Raman spectra were measured for the deposits at the nucleation stage of growth. In the case of the carbon nanowalls synthesized without O2, neither D band nor G band were observed for the Raman spectra of nanoislands formed on the substrate for 1 min or less, in spite of the fact that carbon was detected on these samples by XPS analysis.
Similarly, the same holds for the case of the deposits synthesized with O2 gas addition for 2 min or less. Therefore, the nanoislands and the interface layer underlying two-dimensional nanographene were considered to be amorphous carbon. On the other hand, in the case of the carbon nanowalls synthesized with O2 gas addition, the formation of two-dimensional graphite has started after 3 min growth.
A model for the initial growth mechanism is as follows. Carbon species are generated from CFx radicals adsorbed on or migrating on the surface through the F abstraction reaction by H radicals and condensed to form nanoislands with dangling bonds. This is due to the different growth rates along the strongly bonded planes of expanding graphene sheets and along the weakly bonded stacking direction. Reactive carbon species arriving at the edge of the graphene layer bond easily to the edge, and eventually the graphene layer will expand preferentially in the radical diffusion direction, which is perpendicular to the electrode plane.
On the other hand, low-lying inclined nanoflakes are overshadowed by the fast-growing vertical graphene sheets, terminating the growth of the smaller inclined nanowalls. In the initial growth process of carbon nanowalls with O2 gas, the results in Figs. Oxygen cleans the surface of the Si substrate and etches the amorphous carbon, thereby suppressing the size and number density of carbon nanoislands.
Because oxygen etches the amorphous carbon and reduces the number of defects, carbon nanoflake nucleation was suppressed to some degree. Hence, it may be possible to control nucleation by varying the O radical injection in a controlled manner. Because carbon nanowalls originate on the nuclei, the morphology control and selective growth of carbon nanowalls will be possible, which will be useful for realizing electrical devices made from carbon nanowalls. In this section, carbon nanowalls fabricated using fluorocarbon plasma with H radical injection are the subject of discussion on nucleation and growth mechanism.
In contrast, several groups have synthesized carbon nanowalls using hydrocarbon gases such as CH4 in microwave plasma, inductively coupled plasma ICP , electron-beam-excited plasma, and dc plasma systems [16, 17, 20, 27, 32, 33]. We also synthesized carbon nanowalls using a On the other hand, Zhu et al. At the onset of carbon nanowall nucleation in the initial growth stage, ionic species and the local electric field near the substrate surface may contribute to the nucleation of carbon nanowalls. In the following section, the effect of ionic species on the nucleation of carbon nanowalls is discussed.
In addition to the dominant radicals, however, at the onset of carbon nanowall nucleation in the initial growth stage, ionic species may contribute to the nucleation of carbon nanowalls. The ionic species bombard the substrate or growing surface, resulting in the enhancement of deposition, etching, or surface modification of growing surface physically and chemically. To elucidate the role of ionic species in the nucleation of carbon nanowalls, multi-beam CVD system is constructed, where H and fluorocarbon radical sources and an ion source are installed and the heated substrate is exposed to specific radicals and ions at the same time in a controlled manner.
Carbon nanowalls were synthesized by the simultaneous irradiation of H and fluorocarbon radicals and ions. Two identical radical sources for H and fluorocarbon radicals, which were mounted on both sides of the chamber, consisted of radio frequency RF: C2F6 and H2 gases were introduced into each source separately.
The dominant radical from the dissociation of C2F6 gas was evaluated to be CF3. On the other hand, the ion source consisting of RF A metal mesh connecting to the ground was installed in the head of the ion source. In this experiment, Ar gas was used for the ion source. Generated Ar ions were accelerated between the ICP and the metal mesh on the head, and they irradiated vertically the substrate with an energy ranging from 0 to eV.
The energetic Ar ions dominantly donate physical momentum effects, resulting in the enhancement of chemical reaction under the H and fluorocarbon radical irradiations. Thereby, we focused on the effects of Ar ion irradiation with controlled energy and flux on the formation of carbon nanowalls under simultaneous irradiation of H radicals and fluorocarbon radicals. Growth experiment was carried out under simultaneous irradiations of H and fluorocarbon radicals and Ar ions. The flow rates of H2, C2F6, and Ar were 6, 10, and 5 sccm, respectively, and the total gas pressure was 2. The flux and energy of Ar ions irradiating the heated Si substrate were kept at 3.
X-ray photoelectron spectroscopy XPS showed that these nanoislands were mainly composed of carbon. On the other hand, carbon nanowalls were actually formed after the growth for 50 min as shown in Fig. Thus, it is confirmed that carbon nanowlls were grown by multi-beam CVD with the simultaneous irradiation of H and fluorocarbon radicals and Ar ions. In contrast, after min growth the refractive index of deposits decreased abruptly and then became saturated at approximately 1.
The result in Fig. The nucleation phase consisted of the formation stage of nanoislands of amorphous carbon on the plain substrate. The SEM image in Fig. After a certain duration of nucleation, vertically standing nanographene started to grow on the nanoislands of amorphous carbon, resulting in the transition to growth phase.
It is noted that the carbon nanowall was not formed by the multi-beam CVD without ion irradiation. In order to investigate the effects of Ar ion irradiation on the nucleation and growth of carbon nanowalls separately, growth process was divided into two steps, the first step for nucleation phase 0—15 min and the second step for growth phase 15—50 min , and growth experiment was carried out with and without Ar ion irradiation for each step.
As a result, carbon nanowalls were not grown without Ar ion irradiation in the first step, indicating that the ion irradiation is crucial for the nucleation of carbon nanowalls. To investigate the effect of ion irradiation in the nucleation phase, growth experiment was conducted by multi-beam CVD for 15 min with and without the Ar ion irradiation.
In this case, the flux and energy of Ar ions irradiating the heated Si substrate were kept at 1. Furthermore, the flow rates of H2, C2F6, and Ar were 6, 5, and 10 sccm, respectively, and the total gas pressure was 2. Tilted-view scanning tunneling microscopy STM images of the deposits formed for 15 min with and without the Ar ion irradiation are shown in Fig. Hiramatsu, Y. Tokuda, H. Kano, S. Kimura, O. Sakata, H. Tajiri, M. Applied Physics Express Volume : 3. Japanese Journal of Applied Physics Volume : Malinowski, M. Sekine, W.
Takeuchi, L. Lukasiak, A. Jakubowski, D. Journal of Nanoscience and Nanotechnology 10 Pages : Materials 3 Pages : Kawai, W. Yamakawa, S. Den, H. Sasaki, S.
Kato, S. Takashima, M.
Machino, W. Mori, M. Hiramatsu, K. Yamakawa, K. Ura, M. Kondo, K. B26 Pages : Mitsuguchi, H. Hiramatsu Invited. Kanda, K. Yamakawa, H. Yasuda, H. Shimoeda, H. Takeda, K. Ishikawa, M. Watanabe, K. Takeda, H. Hori Author s Y. Nihashi, M. Kino, H. Whereas, the method developed during my research allows a time-effective synthesis of these nanomaterials; moreover, the deposition of the CNWs directly onto conductive substrate permits for the first time the fabrication of carbon-based resistive switching memory devices.
This technique could be used for the development on a large scale of this type of devices, whose broad fabrication has been hindered due to the complex production mechanisms. Another advantage of the electrochemical processes is the possibility of modifying the chemical composition of the materials.
In this thesis, the anodic oxidation has been used for the first time to oxidize the carbon structures obtained by EPD in order to engineer their electrical performances. In literature, the anodic oxidation has been used to study the redox processes in electronic devices or to increase the electrochemical capacitance of carbon materials, but never as a specific technique to tailor the materials properties. ZnO rods are usually grown by hydrothermal processes, which can be time consuming.
In this thesis, the growth of the rods has been conducted directly on conductive substrates, which were then patterned for the fabrication of electronic devices. Collections Mechanical and Mechatronics Engineering Theses. Cite this version of the work Paola Russo Search UWSpace. This Collection.