<?xml version='1.0'?>
<!DOCTYPE art SYSTEM 'http://www.biomedcentral.com/xml/article.dtd'>
<art><ui>1556-276X-7-264</ui><ji>1556-276X</ji><fm><dochead>Nano Express</dochead><bibl><title><p>Relation between electrical properties of aerosol-deposited BaTiO<sub>3</sub> thin films and their mechanical hardness measured by nano-indentation</p></title><aug><au id="A1"><snm>Kim</snm><fnm>Hong-Ki</fnm><insr iid="I1"/><email>kimhk@kw.ac.kr</email></au><au id="A2"><snm>Oh</snm><fnm>Jong-Min</fnm><insr iid="I1"/><email>ojm@kw.ac.kr</email></au><au id="A3"><snm>In Kim</snm><fnm>Soo</fnm><insr iid="I2"/><email>kksooin@kookmin.ac.kr</email></au><au id="A4"><snm>Kim</snm><fnm>Hyung-Jun</fnm><insr iid="I1"/><email>hyungjun@kw.ac.kr</email></au><au id="A5"><snm>Lee</snm><mnm>Woo</mnm><fnm>Chang</fnm><insr iid="I2"/><email>cwlee@kookmin.ac.kr</email></au><au id="A6" ca="yes"><snm>Nam</snm><fnm>Song-Min</fnm><insr iid="I1"/><email>smnam@kw.ac.kr</email></au></aug><insg><ins id="I1"><p>Department of Electronic Materials Engineering, Kwangwoon University, 447-1, Wolgye-dong, Nowon-gu, Seoul, 139-701, South Korea</p></ins><ins id="I2"><p>Department of Nano and Electronic Physics, Kookmin University, 861-1, Jeongnung-dong, Seongbuk-gu, Seoul, 136-702, South Korea</p></ins></insg><source>Nanoscale Research Letters</source><section><title><p>Regular submissions</p></title></section><issn>1556-276X</issn><pubdate>2012</pubdate><volume>7</volume><issue>1</issue><fpage>264</fpage><url>http://www.nanoscalereslett.com/content/7/1/264</url><xrefbib><pubidlist><pubid idtype="doi">10.1186/1556-276X-7-264</pubid><pubid idtype="pmpid">22616759</pubid></pubidlist></xrefbib></bibl><history><rec><date><day>6</day><month>2</month><year>2012</year></date></rec><acc><date><day>22</day><month>5</month><year>2012</year></date></acc><pub><date><day>22</day><month>5</month><year>2012</year></date></pub></history><cpyrt><year>2012</year><collab>Kim et al.; licensee Springer.</collab><note>This is an Open Access article distributed under the terms of the Creative Commons Attribution License (<url>http://creativecommons.org/licenses/by/2.0</url>), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.</note></cpyrt><kwdg><kwd>BaTiO<sub>3</sub></kwd><kwd>thin film</kwd><kwd>aerosol deposition</kwd><kwd>nano-indentation</kwd><kwd>leakage current</kwd></kwdg><abs><sec><st><p><b>Abstract</b></p></st><p>To achieve a high capacitance density for embedded decoupling capacitor applications, the aerosol deposition (AD) process was applied as a thin film deposition process. BaTiO<sub>3</sub> films were fabricated on Cu substrates by the AD process at room temperature, and the film thickness was reduced to confirm the limit of the critical minimum thickness for dielectric properties. As a result, the BaTiO<sub>3</sub> thin films that were less than 1-&#956;m thick showed unstable electric properties owing to their high leakage currents. Therefore, to overcome this problem, the causes of the high leakage currents were investigated. In this study, it was confirmed that by comparing BaTiO<sub>3</sub> thin films on Cu substrates with those on stainless steels (SUS) substrates, macroscopic defects and rough interfaces between films and substrates influence the leakage currents. Moreover, based on the deposition mechanism of the AD process, it was considered that the BaTiO<sub>3</sub> thin films on Cu substrates with thicknesses of less than 1 &#956;m are formed with chinks and weak particle-to-particle bonding, giving rise to leakage currents. In order to confirm the relation between the above-mentioned surface morphologies and the dielectric behavior, the hardness of BaTiO<sub>3</sub> films on Cu and SUS substrates was investigated by nano-indentation. Consequently, we proposed that the chinks and weak particle-to-particle bonding in the BaTiO<sub>3</sub> thin films with thicknesses of less than 0.5 &#956;m on Cu substrates could be the main cause of the high leakage currents.</p></sec></abs></fm><bdy><sec><st><p><b>Background</b></p></st><p>The ever increasing functionality, portability, and switching frequencies of consumer electronics, which cause parasitic inductance effects such as electromagnetic interference (EMI) and simultaneous switching noise (SSN) <abbrgrp><abbr bid="B1">1</abbr><abbr bid="B2">2</abbr></abbrgrp>, are putting tremendous pressure on designers and manufacturers to pack more circuitry into a smaller space at the same time as suppressing SSN and EMI. Because of these demands, the three-dimensional integration of passive components should be realized instead of relying on existing surface-mounted passive components, i.e., two-dimensional integration, owing to the occupation of space by the passives over the entire printed circuit board area <abbrgrp><abbr bid="B3">3</abbr></abbrgrp>. Among the passive components requiring three-dimensional integration, the development of an embedded decoupling capacitor is particularly important for miniaturization and the suppression of EMI and SSN <abbrgrp><abbr bid="B4">4</abbr><abbr bid="B5">5</abbr></abbrgrp>.</p><p>As candidates for an embedded decoupling capacitor, there are low-temperature co-fired ceramics (LTCCs), polymer composites, films deposited by sputtering, and so on <abbrgrp><abbr bid="B6">6</abbr></abbrgrp>. However, LTCCs still require high-temperature processes around 850&#176;C, which pose a critical problem for the embedding of passive components <abbrgrp><abbr bid="B7">7</abbr></abbrgrp>, and polymer composites have poor dielectric properties <abbrgrp><abbr bid="B8">8</abbr></abbrgrp>. Also, ferroelectric films deposited by sputtering must be annealed at a high temperature in order to achieve the desired crystalline phase and orientation <abbrgrp><abbr bid="B6">6</abbr></abbrgrp>. Therefore, to overcome the processing problems, a new technological approach has been attempted using an aerosol deposition (AD) process that enables us to fabricate dense ceramic thick films at room temperature. Although the AD process was invented for the growth of ceramic thick films <abbrgrp><abbr bid="B9">9</abbr></abbrgrp>, we applied the AD process as a thin film fabrication process to achieve a high capacitance density of over 1,000 nF/cm<sup>2</sup><abbrgrp><abbr bid="B10">10</abbr></abbrgrp>.</p><p>In our previous research, to produce a high capacitance density, BaTiO<sub>3</sub> was deposited on Cu substrates by the AD process at room temperature, and the film thickness was reduced to confirm the critical minimum thickness for dielectric properties. Consequently, all BaTiO<sub>3</sub> thin films that were less than 0.5-&#956;m thick showed short-circuit characteristics owing to high leakage currents, while the BaTiO<sub>3</sub> thin films that were 0.5- to 1-&#956;m thick had only partial dielectric properties, which was an obstacle in achieving a satisfactory thin film process using the AD process <abbrgrp><abbr bid="B11">11</abbr><abbr bid="B12">12</abbr><abbr bid="B13">13</abbr></abbrgrp>. However, the BaTiO<sub>3</sub> films that were more than 1-&#956;m thick exhibited a dense morphology from which defects were markedly absent, which was considered to be due to the rigidity of the surfaces of the hard ceramic BaTiO<sub>3</sub> films. From the above consideration, in order to clarify the role of substrate hardness in growing the deposited films, BaTiO<sub>3</sub> films were fabricated on stainless steel (SUS) substrates which were harder than the Cu substrates. As a result, the BaTiO<sub>3</sub> films on SUS substrates had a lower critical minimum thickness of 0.2 &#956;m, a fewer macroscopic defects, and a lower interface roughness than the BaTiO<sub>3</sub> films on Cu substrates. Thus, we discussed that the high leakage currents were related to macroscopic defects and a rough interface <abbrgrp><abbr bid="B14">14</abbr></abbrgrp>.</p><p>In this study, in order to reveal the influence of the macroscopic defects and the rough interface on the leakage currents, a current image of the macroscopic defects was obtained by conductive atomic force microscopy (C-AFM) analysis, and the roughness of the interface between the films and the substrates was also observed by the focused ion beam (FIB) technique. Moreover, based on the deposition mechanism of the AD process, it was considered that BaTiO<sub>3</sub> thin films of less than 1 &#956;m in thickness are formed on Cu substrates with chinks and weak particle-to-particle bonding, and these surface morphologies are related to the presence of leakage currents. Therefore, the hardness, which depended on the surface morphologies of the BaTiO<sub>3</sub> films on Cu and SUS substrates, was measured by nano-indentation, and the relation between the hardness and the electrical properties of the BaTiO<sub>3</sub> films on Cu and SUS substrates was investigated to identify another cause of the high leakage currents.</p></sec><sec><st><p><b>Methods</b></p></st><p>The AD process is based on shock-loading solidification due to the impact of ceramic particles. During the AD process, fine ceramic particles are accelerated by gas flow in the nozzle up to a velocity of several hundred meters per second and are sprayed onto the substrates. The details of the AD process apparatus can be referred to elsewhere <abbrgrp><abbr bid="B15">15</abbr></abbrgrp>. BaTiO<sub>3</sub> films were deposited by the AD process using a commercial BaTiO<sub>3</sub> powder (BT-045 J, Samsung Fine Chemicals Co., Ltd., Ulsan, South Korea) with a particle size of 0.45 &#956;m as a starting powder. The particles were aerosolized in an aerosol chamber and transported into a deposition chamber using He gas at a flow rate of 5 L/min. The transported BaTiO<sub>3</sub> powder was continuously ejected through the nozzle and deposited onto the Cu and SUS substrates, respectively. The orifice size of the nozzle, the deposition area, the distance between the nozzle and the substrate, the working pressure, and the deposition time were 10&#8201;&#215;&#8201;0.4 mm<sup>2</sup>, 10&#8201;&#215;&#8201;10 mm<sup>2</sup>, 10 mm, 3.4 Torr, and 1 to 10 min, respectively.</p><p>The dielectric properties of the BaTiO<sub>3</sub> films with thicknesses of 0.1 to 2.2 &#956;m on Cu and SUS substrates were measured using an impedance analyzer (HP 4194A, Agilent Technologies, Inc., Santa Clara, CA, USA) and their dielectric behaviors were confirmed. The crystallinity and crystallite size of the BaTiO<sub>3</sub> films on the Cu and SUS substrates were analyzed using an X-ray diffractometer (XRD; X&#8217;Pert PRO, PANalytical, Almelo, The Netherlands). Moreover, in order to confirm the causes of high leakage currents, the roughness values of the interfaces between deposited films and substrates were observed through a FIB system (Helios 600i, FEI, Hillsboro, OR, USA), and the surface morphology and current images were observed by C-AFM (SPM 9600, Shimadzu Corporation, Kyoto, Japan). During C-AFM analysis, a voltage of 5 V was applied. In addition, a field-emission scanning electron microscopy (S-4700, Hitachi, Ltd., Chiyoda-ku, Japan) analysis was conducted, and the hardness of the BaTiO<sub>3</sub> films was investigated by nano-indentation (TriboIndenter, Hysitron, Inc., Eden Prairie, MN, USA) which has been established as a powerful method to characterize the near-surface mechanical properties of materials. For the measurement of the hardness of the BaTiO<sub>3</sub> films, a Berkovich tip was used, the diameter of which was about 100 nm, which can avoid the influence of the surface roughness of the BaTiO<sub>3</sub> films on the measured data. The details of the nano-indentation can be referred to elsewhere <abbrgrp><abbr bid="B16">16</abbr><abbr bid="B17">17</abbr><abbr bid="B18">18</abbr><abbr bid="B19">19</abbr><abbr bid="B20">20</abbr></abbrgrp>.</p></sec><sec><st><p><b>Results and discussions</b></p></st><sec><st><p><b>Influence of rough interface and macroscopic defects on leakage currents</b></p></st><p>In our previous research, BaTiO<sub>3</sub> films with thicknesses of 0.1 to 3.0 &#956;m were fabricated on Cu substrates by an AD process at room temperature and compared with BaTiO<sub>3</sub> films on SUS substrates. As a result, all BaTiO<sub>3</sub> thin films with a thickness of less than 0.5 &#956;m showed short-circuit characteristics owing to high leakage currents, and BaTiO<sub>3</sub> thin films of 0.5 to 1 &#956;m in thickness had partial dielectric properties <abbrgrp><abbr bid="B11">11</abbr><abbr bid="B12">12</abbr><abbr bid="B13">13</abbr></abbrgrp>. In addition, the BaTiO<sub>3</sub> thin films on Cu substrates exhibited rough interfaces between the films and the substrates as well as macroscopic defects such as pores and not-fully-crushed particles. On the other hand, the BaTiO<sub>3</sub> films on SUS substrates had a lower critical minimum thickness of 0.2 &#956;m, a lower interface roughness, and a fewer macroscopic defects than the BaTiO<sub>3</sub> films on Cu substrates. From these results, we proposed that the rough interfaces and the macroscopic defects in the BaTiO<sub>3</sub> thin films with thicknesses of less than 0.5 &#956;m on Cu substrates lead to high leakage currents <abbrgrp><abbr bid="B14">14</abbr></abbrgrp>.</p><p>In this study, in order to confirm the proposed causes of the high leakage currents, BaTiO<sub>3</sub> films of 0.1 to 2.2 &#956;m in thickness were fabricated on Cu and SUS substrates by the AD process, and their electrical properties were also confirmed and compared with those of the previous researches. The results for the electrical properties of the BaTiO<sub>3</sub> films on Cu and SUS substrates are summarized as shown in Table&#8201;<tblr tid="T1">1</tblr>. From Table&#8201;<tblr tid="T1">1</tblr>, the dielectric properties and the dielectric behaviors of the BaTiO<sub>3</sub> films on Cu and SUS substrates had almost the same tendencies as those found in previous researches. The critical minimum thicknesses of the BaTiO<sub>3</sub> films on Cu and SUS substrates were 0.5 and 0.2 &#956;m, respectively. When the thickness of a BaTiO<sub>3</sub> film was below the critical minimum thickness, the leakage currents abruptly increased, indicating short-circuit characteristics. In addition, BaTiO<sub>3</sub> films with thicknesses of 0.5 to 1 &#956;m on Cu substrates exhibited partial dielectric behavior, whereas BaTiO<sub>3</sub> films on SUS substrates exhibited stable dielectric behavior thicknesses of 0.2 &#956;m or more. At the critical minimum thickness of BaTiO<sub>3</sub> films on Cu substrates, their relative permittivity and loss tangent were 60 and 1.9%, respectively, at 1 MHz. In the case of SUS substrates, they were 59 and 2.1%, respectively, at 1 MHz.</p><table id="T1"><title><p>Table 1</p></title><caption><p><b>Relationship of dielectric properties with substrate type and thickness of BaTiO</b><sub><b>3</b></sub><b>films (at 1 MHz)</b></p></caption><tgroup align="left" cols="8"><colspec align="left" colname="c1" colnum="1" colwidth="1*"/><colspec align="left" colname="c2" colnum="2" colwidth="1*"/><colspec align="left" colname="c3" colnum="3" colwidth="1*"/><colspec align="left" colname="c4" colnum="4" colwidth="1*"/><colspec align="left" colname="c5" colnum="5" colwidth="1*"/><colspec align="left" colname="c6" colnum="6" colwidth="1*"/><colspec align="left" colname="c7" colnum="7" colwidth="1*"/><colspec align="left" colname="c8" colnum="8" colwidth="1*"/><thead valign="top"><row rowsep="1"><entry colname="c1" nameend="c2" namest="c1"><p><b>Film thickness (&#956;m)</b></p></entry><entry colname="c3"><p><b>~0.1</b></p></entry><entry colname="c4"><p><b>0.2</b></p></entry><entry colname="c5"><p><b>0.4</b></p></entry><entry colname="c6"><p><b>0.5</b></p></entry><entry colname="c7"><p><b>0.8</b></p></entry><entry colname="c8"><p><b>1.0~</b></p></entry></row></thead><tbody valign="top"><row rowsep="1"><entry colname="c1" morerows="3"><p>Cu substrate</p></entry><entry colname="c2"><p><it>&#1013;</it><sub>r</sub></p></entry><entry colname="c3"><p>N/A</p></entry><entry colname="c4"><p>N/A</p></entry><entry colname="c5"><p>N/A</p></entry><entry colname="c6"><p>60</p></entry><entry colname="c7"><p>69</p></entry><entry colname="c8"><p>&gt;71</p></entry></row><row rowsep="1"><entry colname="c2"><p>tan &#948; (%)</p></entry><entry colname="c3"><p>N/A</p></entry><entry colname="c4"><p>N/A</p></entry><entry colname="c5"><p>N/A</p></entry><entry colname="c6"><p>1.9</p></entry><entry colname="c7"><p>1.7</p></entry><entry colname="c8"><p>&lt;1</p></entry></row><row rowsep="1"><entry colname="c2"><p>Capacitance density (nF/cm<sup>2</sup>)</p></entry><entry colname="c3"><p>N/A</p></entry><entry colname="c4"><p>N/A</p></entry><entry colname="c5"><p>N/A</p></entry><entry colname="c6"><p>106</p></entry><entry colname="c7"><p>76</p></entry><entry colname="c8"><p>63&lt;</p></entry></row><row rowsep="1"><entry colname="c2"><p>Dielectric behavior (%)</p></entry><entry colname="c3"><p><it>0</it></p></entry><entry colname="c4"><p><it>0</it></p></entry><entry colname="c5"><p><it>0</it></p></entry><entry colname="c6"><p><it>50 to 80</it></p></entry><entry colname="c7"><p><it>50 to 80</it></p></entry><entry colname="c8"><p><it>90 to 100</it></p></entry></row><row rowsep="1"><entry colname="c1" morerows="3"><p>SUS substrate</p></entry><entry colname="c2"><p><it>&#1013;</it><sub>r</sub></p></entry><entry colname="c3"><p>N/A</p></entry><entry colname="c4"><p>59</p></entry><entry colname="c5"><p>63</p></entry><entry colname="c6"><p>61</p></entry><entry colname="c7"><p>60</p></entry><entry colname="c8"><p>&gt;65</p></entry></row><row rowsep="1"><entry colname="c2"><p>tan &#948; (%)</p></entry><entry colname="c3"><p>N/A</p></entry><entry colname="c4"><p>2.1</p></entry><entry colname="c5"><p>1.9</p></entry><entry colname="c6"><p>1.7</p></entry><entry colname="c7"><p>2.3</p></entry><entry colname="c8"><p>&lt;1.5</p></entry></row><row rowsep="1"><entry colname="c2"><p>Capacitance density (nF/cm<sup>2</sup>)</p></entry><entry colname="c3"><p>N/A</p></entry><entry colname="c4"><p>261</p></entry><entry colname="c5"><p>139</p></entry><entry colname="c6"><p>108</p></entry><entry colname="c7"><p>66</p></entry><entry colname="c8"><p>&lt;50</p></entry></row><row rowsep="1"><entry colname="c2"><p>Dielectric behavior (%)</p></entry><entry colname="c3"><p><it>0</it></p></entry><entry colname="c4"><p><it>90 to 100</it></p></entry><entry colname="c5"><p><it>90 to 100</it></p></entry><entry colname="c6"><p><it>90 to 100</it></p></entry><entry colname="c7"><p><it>90 to 100</it></p></entry><entry colname="c8"><p><it>90 to 100</it></p></entry></row></tbody></tgroup></table><p>In order to determine the proposed causes of high leakage currents in the previous researches, BaTiO<sub>3</sub> thin films with a thickness of 0.2 &#956;m on Cu and SUS substrates were chosen. This choice was made because, although the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films on Cu substrates have short-circuit characteristics, those on SUS substrates have a dielectric behavior. The fabricated 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films on Cu and SUS substrates successively became crystalline as shown in Figure&#8201;<figr fid="F1">1</figr>a,b. From the XRD patterns shown in Figure&#8201;<figr fid="F1">1</figr>a,b, it was confirmed that the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films on Cu and SUS substrates have a cubic crystal system with a single perovskite phase. The reason for this cubic crystal system of the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films on Cu and SUS substrates is their small crystallite size and nonuniform distortion during impaction in the AD process, and it brings about a decrease in the relative permittivity of the BaTiO<sub>3</sub><abbrgrp><abbr bid="B21">21</abbr></abbrgrp>. However, the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films on Cu and SUS substrates exhibited different peak shapes from each other, so that XRD analysis was additionally conducted between 30.5&#176; and 32.5&#176;, which is the region of the 101 main diffraction peaks of the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films on Cu and SUS substrates, as shown in Figure&#8201;<figr fid="F1">1</figr>c. As a result, the XRD patterns confirm that the diffraction peak of the BaTiO<sub>3</sub> thin films on SUS substrates broadened in comparison with the diffraction peak of those on Cu substrates. This peak broadening was caused by a smaller crystallite size. To support this conclusion, the crystallite sizes were calculated by Scherrer's method. The calculated crystallite sizes of the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films on Cu and SUS substrates were 12.6 and 11.3 nm, and the difference in the full width at half maximum was 10.3%. From this result, it was considered that the crystallite size of the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films on Cu and SUS substrates was affected by the hardness of the substrates.</p><fig id="F1"><title><p>Figure 1</p></title><caption><p>XRD patterns of the 0.2-&#956;m-thick BaTiO<sub>3</sub>thin films on Cu and SUS substrates.(a) Cu and (b) SUS substrates.</p></caption><text>
   <p><b>XRD patterns of the 0.2-&#956;m-thick BaTiO</b><sub><b>3</b></sub><b>thin films on Cu and SUS substrates.</b> (<b>a</b>) Cu and (<b>b</b>) SUS substrates. (<b>c</b>) Cu and SUS substrates between 30.5&#176; and 32.5&#176;.</p>
</text><graphic file="1556-276X-7-264-1"/></fig><p>In our previous researches, the roughness of the interfaces between the films and the substrates was observed by peeling BaTiO<sub>3</sub> thick films of 2 &#956;m in thickness from Cu and SUS substrates to confirm the cause of high leakage currents <abbrgrp><abbr bid="B14">14</abbr></abbrgrp>. As a result, the roughness difference between BaTiO<sub>3</sub> thick films on Cu substrates and BaTiO<sub>3</sub> thick films on SUS substrates was confirmed, but the compared films had a higher thickness than the critical minimum thickness. Therefore, in this study, the roughness of the interface between the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films and the Cu or SUS substrates was observed by a FIB technique. Figure&#8201;<figr fid="F2">2</figr> shows the interfaces of the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films with the Cu and SUS substrates. The BaTiO<sub>3</sub> thin films on Cu substrates had high interface roughness values of 0.15 to 0.3 &#956;m compared with the thickness of the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films, as shown in Figure&#8201;<figr fid="F2">2</figr>a,c. On the other hand, a decreased interface roughness of 0.05 to 0.1 &#956;m was observed for the interfaces between the BaTiO<sub>3</sub> thin films and SUS substrates, as shown in Figure&#8201;<figr fid="F2">2</figr>b,d. From these results, we consider that the interface roughness is generated by the impact of the accelerated BaTiO<sub>3</sub> particles onto the substrate, and thus it is intensified on ductile substrates such as the Cu substrate. Therefore, the BaTiO<sub>3</sub> thin films on Cu substrates had a higher interface roughness than the BaTiO<sub>3</sub> thin films on SUS substrates, and an area with a shortened distance between the top of the BaTiO<sub>3</sub> thin film and the substrate was largely formed. This area can cause a high field concentration, giving rise to high leakage currents, and higher leakage currents are generated for BaTiO<sub>3</sub> thin films on Cu substrates than for those on SUS substrates <abbrgrp><abbr bid="B12">12</abbr></abbrgrp>. In addition, in order to observe the leakage currents at the macroscopic defects, which are the proposed cause of the leakage currents in our previous researches, the surface morphologies of the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films on Cu and SUS substrates were observed by C-AFM analyses. Figure&#8201;<figr fid="F3">3</figr>a,c shows the AFM and C-AFM images of a 0.2-&#956;m-thick BaTiO<sub>3</sub> thin film on a Cu substrate, and Figure&#8201;<figr fid="F3">3</figr>b,d shows the AFM and C-AFM images of a 0.2-&#956;m-thick BaTiO<sub>3</sub> thin film on an SUS substrate. In the AFM image shown in Figure&#8201;<figr fid="F3">3</figr>a, it can be seen that macroscopic defects existed on the surfaces of BaTiO<sub>3</sub> thin films on Cu substrates. Also, high leakage currents were observed in the macroscopic defect region as shown in Figure&#8201;<figr fid="F3">3</figr>c, whereas the BaTiO<sub>3</sub> thin films on SUS substrates had no macroscopic defects and no leakage currents, as shown in Figure&#8201;<figr fid="F3">3</figr>b,d. This result revealed that the leakage currents largely flowed through the macroscopic defects, and thus the macroscopic defects are another cause of the high leakage currents.</p><fig id="F2"><title><p>Figure 2</p></title><caption><p><b>Cross-sectional SEM micrographs of 0.2-&#956;m-thick BaTiO</b><sub><b>3</b></sub><b>thin films on (a) Cu and (b) SUS substrates.</b> Enlarged cross-sectional SEM micrographs of the films on (<b>c</b>) Cu and (<b>d</b>) SUS substrates.</p></caption><text>
   <p><b>Cross-sectional SEM micrographs of 0.2-&#956;m-thick BaTiO</b><sub><b>3</b></sub><b>thin films on (a) Cu and (b) SUS substrates.</b> Enlarged cross-sectional SEM micrographs of the films on (<b>c</b>) Cu and (<b>d</b>) SUS substrates.</p>
</text><graphic file="1556-276X-7-264-2"/></fig><fig id="F3"><title><p>Figure 3</p></title><caption><p><b>AFM topographies of 0.2-&#956;m-thick BaTiO</b><sub><b>3</b></sub><b>thin films on (a) Cu and (b) SUS substrates.</b> Current images of the films on (<b>c</b>) Cu and (<b>d</b>) SUS substrates.</p></caption><text>
   <p><b>AFM topographies of 0.2-&#956;m-thick BaTiO</b><sub><b>3</b></sub><b>thin films on (a) Cu and (b) SUS substrates.</b> Current images of the films on (<b>c</b>) Cu and (<b>d</b>) SUS substrates.</p>
</text><graphic file="1556-276X-7-264-3"/></fig></sec><sec><st><p><b>Surface morphology of BaTiO</b><sub><b>3</b></sub><b>films without macroscopic defect areas</b></p></st><p>From the above results, it was confirmed that the roughness of the interface and the macroscopic defects have a direct influence on the leakage currents. However, we consider that the causes of the leakage currents are not only the rough interface and the macroscopic defects, but also the chinks and weak particle-to-particle bonding that arise due to the deposition mechanism of the AD process. The bonding of ceramic particles in the AD process is based on shock-loading solidification, resulting from the impact of ceramic particles. The densification of the ceramic films is formed by the continuous impaction of ceramic particles onto the pre-impacted particles or substrate. Therefore, it is considered that the impaction of ceramic particles is not sufficient for the densification of the BaTiO<sub>3</sub> thin films on Cu substrates, and therefore, these BaTiO<sub>3</sub> thin films have chinks and weak particle-to-particle bonding. On the other hand, in the case of the BaTiO<sub>3</sub> thin films on SUS substrates, it is considered that even if the impaction of ceramic particles is not high enough, the ceramic particles are sufficiently fractured to form dense BaTiO<sub>3</sub> thin films due to the use of harder substrates than Cu substrates. From the calculated crystallite sizes based on the XRD data, the smaller crystallite sizes of the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films on SUS substrates compared to those of the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films on Cu substrates support the above consideration.</p><p>Therefore, scanning electron microscopy (SEM) analysis was conducted on the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films on Cu and SUS substrates to observe their surface morphologies. In addition, in order to confirm the results of the above-mentioned densification procedure of the BaTiO<sub>3</sub> thick films on Cu and SUS substrates, SEM analysis was additionally conducted on 2-&#956;m-thick BaTiO<sub>3</sub> thick films on Cu and SUS substrates. Figure&#8201;<figr fid="F4">4</figr>a,b shows the SEM images of the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films on Cu and SUS substrates, and Figure&#8201;<figr fid="F4">4</figr>c,d shows the SEM images of the 2-&#956;m-thick BaTiO<sub>3</sub> thick films on Cu and SUS substrates. From the SEM images, the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films on Cu substrates showed incompact surface morphologies and even chinks. On the other hand, the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films on SUS substrates showed dense surface morphologies. In addition, the 2-&#956;m-thick BaTiO<sub>3</sub> thick films on Cu and SUS also had dense surface morphologies, and the incompact and chink morphologies of the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films on Cu substrates disappeared in the 2-&#956;m-thick BaTiO<sub>3</sub> thick films on Cu substrates. From the above results, it is considered that the weak particle-to-particle bonding due to incompact surface morphologies and chinks in the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films on Cu substrates can cause high leakage currents, and the 2-&#956;m-thick BaTiO<sub>3</sub> thick films on Cu substrates become dense due to the above-mentioned densification procedure of the AD process.</p><fig id="F4"><title><p>Figure 4</p></title><caption><p><b>Surface SEM micrographs.</b> Surface SEM micrographs of 0.2-&#956;m-thick BaTiO<sub>3</sub> thin film on (<b>a</b>) Cu and (<b>b</b>) SUS substrates and 2-&#956;m-thick BaTiO<sub>3</sub> thick films on (<b>c</b>) Cu and (<b>d</b>) SUS substrates.</p></caption><text>
   <p><b>Surface SEM micrographs.</b> Surface SEM micrographs of 0.2-&#956;m-thick BaTiO<sub>3</sub> thin film on (<b>a</b>) Cu and (<b>b</b>) SUS substrates and 2-&#956;m-thick BaTiO<sub>3</sub> thick films on (<b>c</b>) Cu and (<b>d</b>) SUS substrates.</p>
</text><graphic file="1556-276X-7-264-4"/></fig></sec><sec><st><p><b>Relation between electrical properties and film hardness</b></p></st><p>From the surface morphologies of the BaTiO<sub>3</sub> films on Cu and SUS substrates, it was considered that the chinks and weak particle-to-particle bonding of the BaTiO<sub>3</sub> thin films bring about an increase in the leakage currents. However, it was difficult to reveal the relation between the surface morphologies and the leakage currents by using only the AFM and SEM analyses. Therefore, in order to exhibit the above-mentioned relation, the hardness of the BaTiO<sub>3</sub> films with thicknesses of 0.2 to 2.2 &#956;m on Cu and SUS substrates was measured by nano-indentation, and then the tendencies of the hardness and the dielectric behavior of the BaTiO<sub>3</sub> films were compared.</p><p>Figure&#8201;<figr fid="F5">5</figr>a,b shows the load versus depth of penetration (<it>P</it><it>h</it>) data plots of the BaTiO<sub>3</sub> films on Cu and SUS substrates according to their thickness values, and Figure&#8201;<figr fid="F5">5</figr>c,d shows calculated hardness data from the <it>P</it><it>h</it> data plot. In order to properly measure the hardness data of the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films on Cu and SUS substrates excluding the effect of the substrate hardness, a load force of 900 &#956;N was used and applied to the BaTiO<sub>3</sub> films with thicknesses over 0.2 &#956;m on Cu and SUS substrates. From the hardness data, although all the BaTiO<sub>3</sub> films with thicknesses of less than 0.5 &#956;m on Cu substrates had hardness values of less than 3 GPa, the hardness values of the BaTiO<sub>3</sub> films with thicknesses of more than 0.5 &#956;m on Cu substrates were increased to more than 3 GPa. On the other hand, the BaTiO<sub>3</sub> films on SUS substrates had hardness values of more than 3 GPa even for their initial deposition thickness of 0.2 &#956;m. From the above results, it is confirmed that the initial hardness and the surface morphologies of the BaTiO<sub>3</sub> films are affected by substrate hardness, and the BaTiO<sub>3</sub> films on Cu substrates become hard and develop dense morphologies as they grow thicker. In the case of the electrical properties of the BaTiO<sub>3</sub> films on Cu and SUS substrates, the BaTiO<sub>3</sub> films on Cu substrates exhibited a critical minimum thickness of 0.5 &#956;m, whereas the BaTiO<sub>3</sub> films on SUS substrates exhibited a dielectric behavior for thicknesses of at least 0.2 &#956;m. To sum up the above results, it was confirmed that the tendency of the hardness of the BaTiO<sub>3</sub> films on Cu and SUS substrates matched that of the electrical properties. It means that the BaTiO<sub>3</sub> thin films of less than 0.5 &#956;m in thickness on Cu substrates with chinks and incompact surface morphologies have low hardness values of less than 3 GPa and short-circuit characteristics, whereas the BaTiO<sub>3</sub> thin films on SUS substrates with dense surface morphologies have high hardness values of more than 3 GPa and a dielectric behavior for thicknesses of 0.2 &#956;m or more. Therefore, we propose that the leakage currents flow through the chinks and the weak particle-to-particle bonding areas of the BaTiO<sub>3</sub> thin films with thicknesses of less than 0.5 &#956;m on Cu substrates and that this is the main cause of the high leakage currents. In addition, from the increased hardness and the dense surface morphologies of the BaTiO<sub>3</sub> films with thicknesses of more than 0.5 &#956;m on Cu substrates that develop as the films grow, it is considered that the hammering effect <abbrgrp><abbr bid="B22">22</abbr></abbrgrp>, which is a densification procedure that works by continuous impaction of ceramic particles, is an important factor to overcome the causes of high leakage currents. Therefore, to achieve a thin film deposition process based on the AD process, further study of the densification mechanism of the BaTiO<sub>3</sub> films should be conducted. If the densification mechanism of the BaTiO<sub>3</sub> films is clarified and the causes of high leakage currents are overcome, the AD process could be successfully applied to embedded decoupling capacitor applications with high capacitance density as a thin film process.</p><fig id="F5"><title><p>Figure 5</p></title><caption><p><b>Load&#8211;displacement curve and thickness dependence of surface hardness of BaTiO</b><sub><b>3</b></sub><b>films.</b><it>P</it>-<it>h</it> curves of BaTiO<sub>3</sub> films on (<b>a</b>) Cu and (<b>b</b>) SUS substrates for a load force of 900 &#956;N and the thickness dependence of surface hardness of the films on (<b>c</b>) Cu and (<b>d</b>) SUS substrates.</p></caption><text>
   <p><b>Load&#8211;displacement curve and thickness dependence of surface hardness of BaTiO</b><sub><b>3</b></sub><b>films.</b><it>P</it>-<it>h</it> curves of BaTiO<sub>3</sub> films on (<b>a</b>) Cu and (<b>b</b>) SUS substrates for a load force of 900 &#956;N and the thickness dependence of surface hardness of the films on (<b>c</b>) Cu and (<b>d</b>) SUS substrates.</p>
</text><graphic file="1556-276X-7-264-5"/></fig></sec></sec><sec><st><p><b>Conclusions</b></p></st><p>In order to clarify the causes of high leakage currents, BaTiO<sub>3</sub> films with thicknesses of 0.1 to 2.2 &#956;m were fabricated on Cu and SUS substrates by an AD process at room temperature. The critical minimum thicknesses of the BaTiO<sub>3</sub> films on Cu and SUS substrates were 0.5- and 0.2-&#956;m thick. From the FIB technique, it was exhibited that the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films on Cu substrates had higher interface roughness than the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films on SUS substrates, which can cause high field concentrations and therefore give rise to high leakage currents. In addition, it was confirmed from the C-AFM analysis that the leakage currents largely flow through the macroscopic defects. Chinks and incompact surface morphologies of the 0.2-&#956;m-thick BaTiO<sub>3</sub> thin films on Cu substrates were observed, and the hardness of BaTiO<sub>3</sub> thin films with thicknesses of less than 0.5 &#956;m on Cu substrates was below 3 GPa. On the other hand, the BaTiO<sub>3</sub> films with thicknesses over 0.5 &#956;m on Cu substrates and the BaTiO<sub>3</sub> films of 0.1 to 2.2 &#956;m in thickness exhibited dense surface morphologies and hardness values of more than 3 GPa. It was confirmed that the tendency of the hardness of the BaTiO<sub>3</sub> films on Cu and SUS substrates matched that of their electrical properties. We explained the above results by concluding that the chinks and weak particle-to-particle bonding bring about leakage currents based on the relation between hardness and the electrical properties.</p></sec><sec><st><p><b>Competing interests</b></p></st><p>The authors declare that they have no competing interests.</p></sec><sec><st><p><b>Authors' contributions</b></p></st><p>HKK, JMO, SIK, and HJK carried out the aerosol-deposited sample fabrication, measurements, and interpretation of the results. CWL and SMN initiated the idea of working on the present topic and analyzed all experiments. All authors read and approved the final manuscript.</p></sec></bdy><bm><ack><sec><st><p><b>Acknowledgments</b></p></st><p>This research was supported by a grant from the Consumer-connected Components and Materials Technology Development program funded by the Ministry of Knowledge Economy, Republic of Korea. The present research has been conducted by the research grant of Kwangwoon University in 2012.</p></sec></ack><refgrp><bibl id="B1"><title><p>Electromagnetic interference (EMI) reduction from printed circuit boards (PCB) using electromagnetic bandgap structures</p></title><aug><au><snm>Shahparnia</snm><fnm>S</fnm></au><au><snm>Ramahi</snm><fnm>OM</fnm></au></aug><source>IEEE Trans Electromagn Compat</source><pubdate>2004</pubdate><volume>46</volume><issue>4</issue><fpage>580</fpage><lpage>587</lpage><xrefbib><pubid idtype="doi">10.1109/TEMC.2004.837671</pubid></xrefbib></bibl><bibl id="B2"><title><p>Simultaneous switching noise suppression using hydrothermal barium titanate thin film capacitors</p></title><aug><au><snm>Balaraman</snm><fnm>D</fnm></au><au><snm>Choi</snm><fnm>J</fnm></au><au><snm>Patel</snm><fnm>V</fnm></au><au><snm>Raj</snm><fnm>PM</fnm></au><au><snm>Abothu</snm><fnm>IR</fnm></au><au><snm>Bhattacharya</snm><fnm>S</fnm></au><au><snm>Wan</snm><fnm>L</fnm></au><au><snm>Swaminathan</snm><fnm>M</fnm></au><au><snm>Tummala</snm><fnm>R</fnm></au></aug><source>Electronic Components and Technology Conference</source><pubdate>2004</pubdate><volume>4</volume><fpage>282</fpage><lpage>288</lpage></bibl><bibl id="B3"><title><p>High dielectric properties of Bi1.5Zn1.0Nb1.5O7thin films fabricated at room temperature</p></title><aug><au><snm>Ryu</snm><fnm>JH</fnm></au><au><snm>Kim</snm><fnm>KY</fnm></au><au><snm>Choi</snm><fnm>JJ</fnm></au><au><snm>Hahn</snm><fnm>BD</fnm></au><au><snm>Yoon</snm><fnm>WH</fnm></au><au><snm>Park</snm><fnm>DS</fnm></au><au><snm>Park</snm><fnm>C</fnm></au></aug><source>J Am Ceram Soc</source><pubdate>2008</pubdate><volume>91</volume><issue>10</issue><fpage>3399</fpage><lpage>3401</lpage><xrefbib><pubid idtype="doi">10.1111/j.1551-2916.2008.02539.x</pubid></xrefbib></bibl><bibl id="B4"><title><p>Room-temperature preparation of Al2O3thick films by aerosol deposition method for integrated RF modules</p></title><aug><au><snm>Tsurumi</snm><fnm>T</fnm></au><au><snm>Nam</snm><fnm>SM</fnm></au><au><snm>Mori</snm><fnm>N</fnm></au><au><snm>Kakemoto</snm><fnm>H</fnm></au><au><snm>Wada</snm><fnm>S</fnm></au><au><snm>Akedo</snm><fnm>J</fnm></au></aug><source>J Kor Ceram Soc</source><pubdate>2003</pubdate><volume>40</volume><issue>8</issue><fpage>715</fpage><lpage>719</lpage></bibl><bibl id="B5"><title><p>Alumina thick films as integral substrates using aerosol deposition method</p></title><aug><au><snm>Nam</snm><fnm>SM</fnm></au><au><snm>Mori</snm><fnm>N</fnm></au><au><snm>Kakemoto</snm><fnm>H</fnm></au><au><snm>Wada</snm><fnm>S</fnm></au><au><snm>Akedo</snm><fnm>J</fnm></au><au><snm>Tsurumi</snm><fnm>T</fnm></au></aug><source>Jpn J Appl Phys</source><pubdate>2004</pubdate><volume>43</volume><issue>8A</issue><fpage>5414</fpage><lpage>5418</lpage><xrefbib><pubid idtype="doi">10.1143/JJAP.43.5414</pubid></xrefbib></bibl><bibl id="B6"><aug><au><snm>Ulrich</snm><fnm>RK</fnm></au><au><snm>Schaper</snm><fnm>LW</fnm></au></aug><source>Integrated Passive Component Technology</source><publisher>Wiley-IEEE Press, Hoboken</publisher><pubdate>2003</pubdate></bibl><bibl id="B7"><title><p>Advanced heat management methods in LTCC technology</p></title><aug><au><snm>Lahti</snm><fnm>M</fnm></au><au><snm>Kautio</snm><fnm>K</fnm></au><au><snm>Juntunen</snm><fnm>E</fnm></au><au><snm>Pet&#228;j&#228;</snm><fnm>J</fnm></au><au><snm>Karioja</snm><fnm>P</fnm></au></aug><source>Electro Magnetic Remote Sensing Defence Technology Centre Technical Conference</source><pubdate>2005</pubdate><volume>2</volume><fpage>A25</fpage></bibl><bibl id="B8"><title><p>Electrical properties of ceramic/polymer composites</p></title><aug><au><snm>Abdullah</snm><fnm>MJ</fnm></au><au><snm>Das-Gupta</snm><fnm>DK</fnm></au></aug><source>IEEE Trans Elec Ins</source><pubdate>1990</pubdate><volume>25</volume><issue>3</issue><fpage>605</fpage><xrefbib><pubid idtype="doi">10.1109/14.55739</pubid></xrefbib></bibl><bibl id="B9"><title><p>Aerosol deposition method (ADM) for nano-crystal ceramics coating without firing</p></title><aug><au><snm>Akedo</snm><fnm>J</fnm></au><au><snm>Lebedev</snm><fnm>M</fnm></au><au><snm>Iwata</snm><fnm>A</fnm></au><au><snm>Ogiso</snm><fnm>H</fnm></au><au><snm>Nakano</snm><fnm>S</fnm></au></aug><source>Mat Res Soc Symp Proc</source><pubdate>2003</pubdate><volume>778</volume><fpage>289</fpage><lpage>299</lpage></bibl><bibl id="B10"><title><p>International Technology Roadmap for Semiconductors</p></title><note>[<url>http://public.itrs.net</url>]</note></bibl><bibl id="B11"><title><p>Thickness dependence of dielectric properties in BaTiO3films fabricated by aerosol deposition method</p></title><aug><au><snm>Oh</snm><fnm>J</fnm></au><au><snm>Kim</snm><fnm>NH</fnm></au><au><snm>Choi</snm><fnm>SC</fnm></au><au><snm>Nam</snm><fnm>SM</fnm></au></aug><source>Mat Sci Eng B</source><pubdate>2009</pubdate><volume>161</volume><fpage>80</fpage><lpage>84</lpage><xrefbib><pubid idtype="doi">10.1016/j.mseb.2009.01.028</pubid></xrefbib></bibl><bibl id="B12"><title><p>Causes of high leakage currents in thin BaTiO3films prepared by aerosol deposition method</p></title><aug><au><snm>Oh</snm><fnm>JM</fnm></au><au><snm>Nam</snm><fnm>SM</fnm></au></aug><source>J Kor Ceram Soc</source><pubdate>2010</pubdate><volume>56</volume><issue>1</issue><fpage>1</fpage><lpage>4</lpage></bibl><bibl id="B13"><title><p>Characterization of leakage current mechanisms for aerosol-deposited BaTiO3thin films at room temperature</p></title><aug><au><snm>Oh</snm><fnm>JM</fnm></au><au><snm>Kim</snm><fnm>HJ</fnm></au><au><snm>Nam</snm><fnm>SM</fnm></au></aug><source>J Kor Ceram Soc</source><pubdate>2010</pubdate><volume>57</volume><issue>4</issue><fpage>1096</fpage><lpage>1101</lpage></bibl><bibl id="B14"><title><p>Role of surface hardness of substrates in growing BaTiO3thin films by aerosol deposition method</p></title><aug><au><snm>Oh</snm><fnm>JM</fnm></au><au><snm>Nam</snm><fnm>SM</fnm></au></aug><source>Jpn J Appl Phys</source><pubdate>2009</pubdate><volume>48</volume><fpage>09KA07-1</fpage></bibl><bibl id="B15"><title><p>Aerosol deposition of ceramic thick films at room temperature: densification mechanism of ceramic layers</p></title><aug><au><snm>Akedo</snm><fnm>J</fnm></au></aug><source>J Am Ceram Soc</source><pubdate>2006</pubdate><volume>89</volume><issue>6</issue><fpage>1834</fpage><xrefbib><pubid idtype="doi">10.1111/j.1551-2916.2006.01030.x</pubid></xrefbib></bibl><bibl id="B16"><title><p>Physical properties and collapse force of according to the z-position of poly-Si pattern using nano-tribology</p></title><aug><au><snm>Kim</snm><fnm>SI</fnm></au><au><snm>Lee</snm><fnm>CW</fnm></au></aug><source>J Nanosci Nanotechnol</source><pubdate>2011</pubdate><volume>11</volume><issue>2</issue><fpage>1401</fpage><lpage>1404</lpage><xrefbib><pubidlist><pubid idtype="doi">10.1166/jnn.2011.3392</pubid><pubid idtype="pmpid">21456198</pubid></pubidlist></xrefbib></bibl><bibl id="B17"><title><p>Physical properties of a HDI-PR after dipping it in a plasma-induced liquid&#8211;vapor-activated (PLVA) PR stripper as measured by using nano-indentation</p></title><aug><au><snm>Kim</snm><fnm>SI</fnm></au><au><snm>Lee</snm><fnm>CW</fnm></au></aug><source>J Kor Phys Soc</source><pubdate>2009</pubdate><volume>55</volume><issue>3</issue><fpage>995</fpage><lpage>998</lpage><xrefbib><pubid idtype="doi">10.3938/jkps.55.995</pubid></xrefbib></bibl><bibl id="B18"><title><p>Surface potential behaviors of UV treated of Ag anode for high-performance T-OLED by nanotribology</p></title><aug><au><snm>Kim</snm><fnm>SI</fnm></au><au><snm>Oh</snm><fnm>HW</fnm></au><au><snm>Huh</snm><fnm>JW</fnm></au><au><snm>Ju</snm><fnm>BK</fnm></au><au><snm>Lee</snm><fnm>CW</fnm></au></aug><source>Thin Solid Films</source><pubdate>2011</pubdate><volume>519</volume><fpage>6872</fpage><lpage>6875</lpage><xrefbib><pubid idtype="doi">10.1016/j.tsf.2011.04.046</pubid></xrefbib></bibl><bibl id="B19"><title><p>Atomic force microscopy measurement of the Young's modulus and hardness of single LaB6nanowires</p></title><aug><au><snm>Zhang</snm><fnm>H</fnm></au><au><snm>Tang</snm><fnm>J</fnm></au><au><snm>Zhang</snm><fnm>L</fnm></au><au><snm>An</snm><fnm>B</fnm></au><au><snm>Qin</snm><fnm>L</fnm></au></aug><source>Appl Phys Lett</source><pubdate>2008</pubdate><volume>92</volume><fpage>173121</fpage><xrefbib><pubid idtype="doi">10.1063/1.2919718</pubid></xrefbib></bibl><bibl id="B20"><title><p>Nanoscale mechanical behavior of vanadium doped ZnO piezoelectric nanofiber by nanoindentation technique</p></title><aug><au><snm>Chen</snm><fnm>YQ</fnm></au><au><snm>Zheng</snm><fnm>XJ</fnm></au><au><snm>Mao</snm><fnm>SX</fnm></au><au><snm>Li</snm><fnm>W</fnm></au></aug><source>Appl Phys Lett</source><pubdate>2010</pubdate><volume>107</volume><fpage>094302</fpage></bibl><bibl id="B21"><title><p>Room temperature impact consolidation (RTIC) of fine ceramic powder by aerosol deposition method and applications to microdevices</p></title><aug><au><snm>Akedo</snm><fnm>J</fnm></au></aug><source>J Thermal Spray Tech</source><pubdate>2007</pubdate><volume>17</volume><issue>2</issue><fpage>181</fpage></bibl><bibl id="B22"><title><p>Growth process of &#945;-Al2O3ceramic films on metal substrates fabrication at room temperature by aerosol deposition</p></title><aug><au><snm>Lee</snm><fnm>DW</fnm></au><au><snm>Kim</snm><fnm>HJ</fnm></au><au><snm>Kim</snm><fnm>YH</fnm></au><au><snm>Yun</snm><fnm>YH</fnm></au><au><snm>Nam</snm><fnm>SM</fnm></au></aug><source>J Am Ceram Soc</source><pubdate>2011</pubdate><volume>94</volume><issue>9</issue><fpage>3131</fpage><lpage>3138</lpage><xrefbib><pubid idtype="doi">10.1111/j.1551-2916.2011.04493.x</pubid></xrefbib></bibl></refgrp></bm></art>