1: % Converted from Microsoft Word to LaTeX

2: % by Chikrii Softlab Word2TeX converter (version 4.0)

3: % Copyright (C) 1999-2007 Chikrii Softlab. All rights reserved.

4: % http://www.chikrii.com

5: % mailto: info@chikrii.com

6: % License: CSL#003FD7

7: 

8: \documentclass[rmp, twocolumn, groupedaddress, superscriptaddress, showkeys]{revtex4}

9: \usepackage{latexsym}

10: \usepackage{graphicx}

11: 

12: \begin{document}

13: \title{Reverse Hall-Petch effect in ultra nanocrystalline diamond}

14: 

15: \author{Ioannis N. Remediakis}

16: \thanks{To whom correspondence should be addressed}

17: \email{remed@materials.uoc.gr}

18: \affiliation{Department of Materials Science and Technology, University of

19: Crete, 71003 Heraklion, Greece}

20: \author{Georgios Kopidakis}

21: \affiliation{Department of Materials Science and Technology, University of

22: Crete, 71003 Heraklion, Greece}

23: \author{Pantelis C. Kelires}

24: \affiliation{Department of Physics, University of Crete, 71003 Heraklion,

25: Greece}

26: \affiliation{Department of Mechanical Engineering and Materials Science and

27: Engineering, Cyprus University of Technology, 3603 Limassol, Cyprus}

28: 

29: \begin{abstract}

30: We present atomistic simulations for the mechanical response of ultra

31: nanocrystalline diamond, a polycrystalline form of diamond with grain

32: diameters of the order of a few nm. We consider fully three-dimensional model

33: structures, having several grains of random sizes and orientations, and employ

34: state-of-the-art Monte Carlo simulations. We calculate structural properties,

35: elastic constants and the hardness of the material; our results compare well

36: with experimental observations for this material. Moreover, we verify that

37: this material becomes softer at small grain sizes, in analogy to the observed

38: reversal of the Hall-Petch effect in various nanocrystalline metals. The

39: effect is attributed to the large concentration of grain boundary atoms at

40: smaller grain sizes. Our analysis yields scaling relations for the elastic

41: constants as a function of the average grain size.

42: \end{abstract}

43: 

44: \date{May 21, 2008}

45: 

46: \keywords{nanocrystalline materials; carbon; hardness \vspace{6mm} \\

47: Proceedings of the IUTAM Symposium on Modelling Nanomaterials and Nanosystems,

48: Aalborg, Denmark, May 19-22 2008; to be published in the IUTAM Bookseries by

49: Springer.

50: \vspace{6mm}}

51: 

52: \maketitle

53: 

54: \section{Introduction}

55: 

56: Most solids are polycrystalline, having grains in the micrometer to millimeter

57: range. As the percentage of atoms residing on grain boundaries is negligible,

58: the polycrystallicity only marginally affects the properties of the

59: material. In particular, mechanical properties of such solids usually depend

60: on bulk properties of the ideal material and the concentration of various

61: defects, such as cracks, dislocations, vacancies and interstitials; usually

62: the grain size is of minor importance. There are, however, properties where

63: grain size plays a key role, hardness being one of them.  The Hall-Petch law

64: states that the hardness ($H$) of polycrystalline metals increases with

65: decreasing average grain size ($d$), being a linear function of $d^{-n}$,

66: where $n>$0. The effect is attributed to the impedance of dislocation motion

67: due to grain boundaries (Bata and Pereloma 2004).

68: 

69: Modern nanotechnology makes it possible to synthesize nanocrystalline 

70: solids, i.e. polycrystalline solids with average grain sizes of a few nm. 

71: Such materials offer new possibilities for technological applications, 

72: mainly due to their unique mechanical properties (Meyers et al. 2006). The 

73: Hall-Petch law dictates that a nanocrystalline solid should have huge 

74: hardness, usually much higher than its usual polycrystalline phase. While 

75: this is true in most cases, Cu was found to become softer with decreasing 

76: grain size in the range between 3 and 7 nm (Schi{\o}tz et al. 1998). This 

77: was called the ``reverse Hall-Petch effect''. Later, it was found that many 

78: materials posses a ``strongest size'' (Yip 1998), which turned out to be 

79: around 15 nm for Cu (Schi{\o}tz and Jacobsen 2003). A similar effect has 

80: been observed recently for BN (Dubrovinskaia 2007).

81: 

82: The existence of a ``strongest size'' suggests that the mechanism of

83: undertaking mechanical load should be different in the nano-world. In this

84: region, the presence of dislocations no longer governs the mechanical response

85: on the material: as dislocations are extended defects, they cannot reside in

86: the limited space of nano-grains. Any external mechanical load will be

87: primarily undertaken by sliding along grain boundaries (Schi{\o}tz et

88: al. 1998, van Swygenhoven et al. 1999, Yamakov et al. 2004). This, in turn, is

89: a direct consequence of the enormous concentration of grain boundary atoms in

90: a nanocrystalline material. Imagine for example a cube of side $d$, containing

91: $N\times N \times N$ atoms. The fraction of surface atoms is roughly

92: proportional to 6$N^{2}$/$N^{3}$ = 6/$N$ $\approx $ 1 nm/$d$, ranging from 1

93: per million when $d$ is of the order of mm, to 30{\%} when $d$ is around 3 nm.

94: 

95: Contrary to a large number of studies for metals, very few studies have

96: addressed the mechanical properties of semiconductors and insulators as a

97: function of their grain size, although several pioneer workers have examined

98: the mechanical response of nanocrystalline ceramics (Demkowicz et al. 2007;

99: Szlufarska et al. 2005). Covalent solids are characterized by their open

100: structures and the strong directionality of their bonds. Such bonds should not

101: allow easy sliding along grain boundaries. At the same time, bonds on grain

102: boundaries will be considerably weaker than those in the bulk of grains, due

103: to the loss of the ideal local bonding geometry (Keblinski et

104: al. 1999). Recent studies in BN (Dubrovinskaia et al. 2007), together with the

105: well-established results for various metals, suggest that the effect might be

106: universal. To check whether this is the case, we study diamond, which

107: comprises the ideal test suite for this purpose. C atoms form perhaps the most

108: directional bonds known, as indicated by the supreme hardness and shear

109: modulus of diamond. In addition to being strong, C-C might in some cases break

110: and form a more stable structure, as threefold $sp^{2}$ C atoms are more

111: stable than fourfold $sp^{3}$ ones, the later being responsible for mechanical

112: failure in carbon-based materials (Fyta et al. 2006; Remediakis et al. 2007).

113: 

114: Ultra nanocrystalline diamond (UNCD) is a polycrystalline carbon-based 

115: material, with grain diameters mostly between 2 and 5 nm (Gruen 1999). It is 

116: a low-cost material with a potential for a wide range of applications due to 

117: its unique mechanical and electronic properties (Krauss et al. 2001). 

118: Despite the strong directional C-C bonds, resulting in inhomogeneity at the 

119: atomic scale, the material can be considered as isotropic at larger scales, 

120: as no particular orientation for the grain boundaries in UNCD seems to be 

121: favoured in the experiment (Gruen 1999) or simulations (Zapol et al. 2001, 

122: Kopidakis et al. 2007). 

123: 

124: 

125: \section{Computational method}

126: 

127: We use fully three-dimensional atomistic models for UNCD, having grains of 

128: different random size and orientations separated by random grain boundaries. 

129: The simulations were performed using a continuous-space Monte Carlo method. 

130: We employ the many-body potential of Tersoff (Tersoff 1988), which provides 

131: a very good description of the structure and energetics for a wide range of 

132: carbon-based materials (Kelires 1994; Kelires 2000). This method, although 

133: considerably demanding computationally, allows for great statistical 

134: accuracy, as it is possible to have samples at full thermodynamic 

135: equilibrium. Such accuracy is necessary in order to capture all 

136: hybridizations of C. 

137: 

138: We model UNCD by a periodic repetition of cubic supercells that consist of 

139: eight different regions (grains). The number of grains in the unit cell 

140: guarantees the absence of artificial interactions between a grain and its 

141: periodic images. The grains have random shapes and sizes, and are filled 

142: with atoms in a randomly oriented diamond structure. The method we use is 

143: identical to the method used by Schi{\o}tz et al. (Schi{\o}tz et al. 1998; 

144: Schi{\o}tz and Jacobsen 2003). To achieve a fully equilibrated structure for 

145: each grain size, we perform four steps: first, the structure is compressed 

146: and equilibrated at constant volume at 300 K, in order to eliminate large 

147: void regions near some grain boundaries, that are an artifact of the 

148: randomly generated structure. In the second step, we anneal the system at 

149: 800 K allowing volume relaxation and quench down to 300 K. Third, we anneal 

150: once more, at 1200 K this time, in order to ensure full equilibration. 

151: Fourth, we fully relax the structure at 300 K allowing for changes in both 

152: volume and shape of the unit cell.

153: 

154: 

155: 

156: \begin{table}

157: \caption{Properties of some characteristic UNCD samples at 300K: average grain

158: size ($d$, in nm); number of atoms in the simulation cell ($N$);

159: percentage of three-fold atoms in the cell ($N_{3}$, at {\%}); mass

160: density ($\rho$, in g/cc); cohesive energy ($E_{c}$, in eV

161: per atom); bulk ($B$), Young's ($E$) and shear ($G$)

162: moduli (all in GPa); Vickers hardness ($H$), as estimated from the

163: theory of Gao et al. (Gao et al. 2003). For comparison, the corresponding

164: values for single-crystal diamond, calculated with the same method (‰elires

165: 1994), are shown in the last line.}

166: 

167: \begin{center}

168: \begin{tabular}{rrrrrrrrr}

169: \hline

170: $d$& 

171: $N$& 

172: $N_{3}$& 

173: $\rho $& 

174: $E_{c}$& 

175: $B$& 

176: $E$& 

177: $G$& 

178: $H$ \\

179: \hline

180: 2.4& 

181: 18,528& 

182: 26& 

183: 3.22& 

184: -7.06& 

185: 323& 

186: 808& 

187: 373& 

188: 89.8 \\

189: \hline

190: 3.4& 

191: 53,494& 

192: 12& 

193: 3.30& 

194: -7.10& 

195: 363& 

196: 939& 

197: 439& 

198: 90.9 \\

199: \hline

200: 4.4& 

201: 116,941& 

202: 9& 

203: 3.40& 

204: -7.15& 

205: 384& 

206: 987& 

207: 461& 

208: 91.6 \\

209: \hline

210: $\infty $& 

211: $\infty $& 

212: 0& 

213: 3.51& 

214: -7.33& 

215: 443& 

216: 1066& 

217: 485& 

218: 94.2 \\

219: \hline

220: \end{tabular}

221: \label{tab1}

222: \end{center}

223: \end{table}

224: 

225: \begin{figure}

226: \centerline{\includegraphics[width=\columnwidth]{rem1.eps}}

227: \label{fig1}

228: \caption{ Relaxed model structure for Ultra-Nanocrystalline Diamond, with an

229: average grain size of 4.4 nm. Atoms are coloured according to their number of

230: neighbours ($z$) and average angle between their bonds ($\theta $). Atoms in

231: the diamond structure ($z$=4, $\cos\theta $=-1/3) are coloured gray; all other

232: atoms are coloured blue.}

233: \end{figure}

234: 

235: 

236: \section{Structure and elastic moduli}

237: 

238: The relaxed structure for a typical sample is shown in Fig. 1. The grain 

239: boundaries are a few atomic diameters across, in accordance with experiments 

240: showing widths of 0.2-0.5 nm (Gruen 1999). Atoms at grain boundaries are 

241: either three-fold coordinated or form bonds at different lengths or angles 

242: from those observed in diamond. The structural and elastic properties for 

243: characteristic samples are summarized in Table I. The fraction of the 

244: three-fold atoms in the samples is about 1/10 for grain sizes between 3.5 

245: and 4.5 nm; in experiment, it was observed that the fraction of atoms 

246: residing at grain boundaries is close to 10{\%} for similar crystallite 

247: sizes (Gruen 1999). The density of UNCD increases with increasing 

248: 

249: grain size, as both the percentage of \textit{sp}$^{2}$ atoms and the 

250: concentration of voids is decreased. The cohesive energy decreases with 

251: increasing grain size, suggesting that most grain boundary atoms should be 

252: considered as defective ones.

253: 

254: The bulk modulus of UNCD decreases with decreasing grain size, despite the 

255: increase in the fraction of energetically favorable \textit{sp}$^{2}$ atoms. 

256: Although they have lower energy, \textit{sp}$^{2}$ C atoms are actually 

257: easier to deform compared to \textit{sp}$^{3}$ ones. This can be 

258: demonstrated by employing the concept of local bulk moduli (Kelires 2000). A 

259: similar analysis of our samples yields the average local bulk modulus of 

260: \textit{sp}$^{2 }$atoms to be around 250 GPa, while the average bulk modulus 

261: of \textit{sp}$^{3 }$atoms is around 420 GPa. This agrees very well with 

262: experimental observations for UNCD where the grain boundaries have been 

263: found to have much lower local bulk moduli than the bulk of grains (Pantea 

264: et al. 2006).

265: 

266: \begin{figure}

267: \centerline{\includegraphics[width=\columnwidth]{rem2.eps}}

268: \label{fig2}

269: \caption{Cohesive energy ($E_{c})$ , bulk modulus ($B$), Young's modulus ($E$)

270: and shear modulus ($G$) of UNCD as a function of the average grain size. Solid

271: lines in the upper panels are fits to the simulations. The triangle in the

272: graph for $E$ represents an experimental measurement (Espinosa et al. 2006). }

273: \end{figure}

274: 

275: 

276: \section{Hardness}

277: 

278: Decrease of all elastic moduli with decreasing average grain size suggests 

279: that the hardness of the material should also drop with decreasing grain 

280: size, as the hardness of many materials is proportional to the Young's or 

281: shear modulus (Brazhkin et al. 2002); in particular, the hardness of all 

282: known carbon-based materials has been found to be between 10{\%} and 16{\%} 

283: of the Young's modulus (Robertson 2002). Elastic moduli are reliable probes 

284: of hardness for nanocrystalline solids, as the later cannot contain extended 

285: defects, such as cracks or dislocations, that have characteristic lengths 

286: exceeding the size of the grains. 

287: 

288: To get a quantitative description of hardness, we use the theory of Gao et 

289: al., who correlate the Vickers hardness of covalent crystals with the 

290: electron density per bond and the energy gap of the material (Gao et al. 

291: 2003). The hardness of a complex material is the geometrical mean of the 

292: values of hardness for each subsystem. Here, we consider each individual 

293: pair of neighbouring C atoms as a subsystem. The density of valence 

294: electrons in a particular bond can be obtained from the bond length and the 

295: coordination numbers of the two atoms that participate in this bond. The 

296: calculated hardness of UNCD is shown in Fig. 3, demonstrating the existence 

297: of the reverse Hall-Petch effect for this material.

298: 

299: \begin{figure}

300: \centerline{\includegraphics[width=\columnwidth]{rem3.eps}}

301: \label{fig3}

302: \caption{Estimated hardness of UNCD as a function of the average 

303: grain size. Hardness is calculated based on the theory of F. Gao et al. for 

304: covalent solids (Gao et al. 2003).}

305: \end{figure}

306: 

307: 

308: \section{Scaling laws for the properties of UNCD}

309: 

310: To understand the mechanism behind softening of UNCD, as well as other 

311: materials, at low grain sizes, we consider the different types of atoms and 

312: local bonding geometries that exist in a nanocrystalline material. The main 

313: dramatic change that takes place in a polycrystalline material as its grain 

314: size enters the nm regime is the increase in the fraction of atoms residing 

315: near grain boundaries. Atoms at grain faces, edges or vertexes, as well as 

316: atoms near other discontinuities, will naturally form bonds that are weaker 

317: than those formed by atoms in the bulk. Such weaker bonds will then bend or 

318: stretch with greater ease, compared to the bonds in the crystalline region. 

319: This explains the softening of polycrystalline solids when the grain size is 

320: at the nanometer range. For much larger grain sizes, the number of 

321: grain-boundary atoms will be negligible compared to the number of bulk 

322: atoms; in this regime, the behavior of the material under mechanical load 

323: will be mostly determined by bulk defects, such as dislocations. 

324: 

325: To make this picture quantitative, let us divide the atoms in the 

326: polycrystalline material into three categories:

327: 

328: \begin{enumerate}

329: \item Atoms deep inside the grains, forming bonds that are similar to those in

330: the single-crystal material. The number of such atoms is proportional to

331: \textit{d}$^{3}$, where \textit{d} is the average grain size.

332: \item Atoms near the grain boundary; these behave similarly to surface or

333: interface atoms. The number of such atoms is proportional to \textit{d}$^{2}$.

334: \item Atoms near grain boundary edges; these are similar to kink surface

335: atoms, or atoms near dislocation cores. The number of such atoms is

336: proportional to \textit{d}.

337: \end{enumerate}

338: Of course, there will be other types of atoms, such as vertex atoms or atoms 

339: near topologic defects, but their number will be much smaller than the 

340: numbers of atoms falling in one of the aforementioned categories. The 

341: cohesive energy of the solid will be the sum of the energies of the three 

342: different atom types, multiplied by their respective numbers, and divided by 

343: the total number of atoms, which is proportional to \textit{d}$^{3}$. 

344: Therefore, the cohesive energy should be described by a function of the form

345: 

346: \begin{center}

347: \textit{E}$_{c}$ = \textit{E}$_{0}$ +\textit{a/d}+\textit{b/d}$^{2}$,

348: \end{center}

349: 

350: where \textit{a} and \textit{b} are constants, and \textit{E}$_{0}$ is the 

351: cohesive energy of the single crystal.

352: 

353: Indeed, such a function fits our data perfectly, the rms error being less 

354: than 0.5{\%}. Moreover, \textit{E}$_{0}$ is found to be -7.31 eV, very close 

355: to the calculated cohesive free energy of diamond at 300 K, which is -7.33 

356: eV. As B is proportional to the second derivative of the total energy with 

357: respect to the system volume, it can also be decomposed into contributions 

358: from bulk, interface and vertex atoms. As shown in Fig. 2, a quadratic 

359: function of 1/\textit{d} fits the results of the simulation very nicely. The 

360: constant value, 467 GPa, corresponding to the ideal monocrystalline solid, 

361: is only 5{\%} off the calculated value for diamond (see Table I). Such a 

362: decomposition of the total bulk modulus to a sum of atomic-level moduli has 

363: been used previously, in order to investigate the rigidity of amorphous 

364: carbon (Kelires 2000). 

365: 

366: A similar scaling law should also hold for the mass density of UNCD as a 

367: function of grain size, assuming that the volume per atom is different for 

368: atoms in grain boundaries and atoms in the bulk of grains. Fitting our data 

369: to a quadratic form of 1/\textit{d} gives \textit{$\rho 

370: $=}3.6-1.2/\textit{d}+0.4/\textit{d}$^{2}$. Again, the agreement of the 

371: constant value with the calculation for ideal diamond (Table I) is very good 

372: (3{\%}). 

373: 

374: Hardness is related to the electron density according to Gao et al.; the 

375: local electron density is proportional to the local mass density, as all C 

376: atoms have the same number of electrons. Therefore, hardness should also be 

377: decomposed into contributions from different kinds of atoms. As shown in 

378: Fig. 3, the hardness of UNCD can be fitted to a quadratic form of 

379: 1/\textit{d. }Moreover, the constant term, showing the limit of hardness as 

380: \textit{d} goes to infinity, coincides with the hardness of diamond at 300 

381: K, calculated using the same method (see Table I).

382: 

383: The Young's and shear modulus of UNCD will not necessarily follow the same 

384: scaling law. As both moduli are related to bond bending, the nature of the 

385: inter-atomic bonds is perhaps equally important to their number. We tried to 

386: fit our simulation data to a quadratic form of 1/\textit{d}. Although the 

387: fit does not look disappointing, the rms error in the fits were 

388: significantly higher than those for the fits of \textit{E}$_{c}$, 

389: \textit{B}, \textit{$\rho $} or \textit{H}; moreover, the constant values 

390: deviate from the properties of diamond by more than 20{\%}. However, even 

391: such a poor agreement between model and simulation provides extra evidence 

392: that our model has some solid basis.

393: 

394: 

395: \section{Conclusions}

396: 

397: Using ultra-nanocrystalline diamond (UNCD) as a prototype for a 

398: polycrystalline covalent solid with grains at the nanometer region, we have 

399: observed softening of the material as the grain size decreases, in analogy 

400: with the reverse Hall-Petch effect observed in nanocrystalline metals. The 

401: effect is attributed to the increasing fraction of grain-boundary atoms as 

402: the grain size decreasing. A simple quadratic form in 1/\textit{d}, where 

403: \textit{d} is the average grain size, suffices to provide excellent fit of 

404: our results for cohesive energy, mass density, bulk modulus and estimated 

405: hardness, while it yields the correct values for bulk diamond. The measured 

406: Young's modulus of UNCD is reproduced well by the simulations. Our results 

407: provide further evidence that softening at low grain sizes might be a 

408: universal property for nanocrystalline solids.

409: 

410: \acknowledgments{The authors are grateful to Prof. Jacob Schi{\o}tz who shared

411: his programs for generating models of nanocrystalline solids, and acknowledge

412: inspiring discussions with Dr. Maria Fyta. This work was supported by a grant

413: from the Ministry of National Education and Religious Affairs in Greece

414: through the action ``$\mathrm{E}\Pi\mathrm{EAEK}$'' (programme ``$\Pi \Upsilon

415: \Theta$A$\Gamma$OPA$\Sigma$'').}

416: 

417: \begin{thebibliography}{99}

418: 

419: 

420: \bibitem{1} Bata V, Pereloma EV (2004) An alternative physical explanation of

421: the Hall-Petch relation. Acta Mater 52:657-665

422: 

423: \bibitem{2} Brazhkin VV, Lyapin AG, Hemley RJ (2002) Harder than diamond:

424: dreams and reality. Philos Mag A 82:231-253

425: 

426: \bibitem{3} Demkowicz MJ, Argonz AS, Farkas D, Frary M (2007) Simulation of

427: plasticity in nanocrystalline silicon. Philos Mag 87:4253-4271

428: 

429: \bibitem{4} Dubrovinskaia N, Solozhenko VL, Miyajima N, Dmitriev V, Kurakevych

430: OO, Dubrovinsky L (2007) Superhard nanocomposite of dense polymorphs of boron

431: nitride: Noncarbon material has reached diamond hardness. Appl Phys Lett

432: 90:101912

433: 

434: \bibitem{5} Espinosa HD, Peng B, Moldovan N, Friedmann TA, Xiao X, Mancini DC,

435: Auciello O, Carlisle J, Zorman CA, Merhegany M (2006) Elasticity, strength and

436: toughness of single crystal silicon carbide, ultrananocrystalline diamond and

437: hydrogen-free tetrahedral amorphous carbon. Appl Phys Lett 89:073111.

438: 

439: \bibitem{6} Fyta MG, Remediakis IN, Kelires PC, Papaconstantopoulos DA (2006)

440: Insights into the fracture mechanisms and strength of amorphous and

441: nanocomposite carbon. Phys Rev Lett 96:185503

442: 

443: \bibitem{7} Gao FM, He JL, Wu ED, Liu SM, Yu DL, Li DC, Zhang SY, Tian YJ

444: (2003) Hardness of covalent crystals. Phys Rev Lett 91:015502

445: 

446: \bibitem{8} Gruen DM (1999) Nanocrystalline diamond films. Annu Rev Mater Sci

447: 29:211-259

448: 

449: \bibitem{9} Kaner RB, Gilman JJ, Tolbert SH (2005) Materials science -

450: Designing superhard materials. Science 308:1268-1269

451: 

452: \bibitem{a} Keblinski P, Phillpot SR, Wolf D, Gleiter H (1999) On the nature of

453: grain boundaries in nanocrystalline diamond. Nanostruct Mater 12:339-344

454: 

455: \bibitem{b} Kelires PC (1994) Elastic Properties Of Amorphous-Carbon

456: Networks. Phys Rev Lett 73:2460-2463

457: 

458: \bibitem{c} Kelires PC (2000) Intrinsic stress and local rigidity in

459: tetrahedral amorphous carbon. Phys Rev B 62:15686-15694

460: 

461: \bibitem{d} Kopidakis G, Remediakis IN, Fyta MG, Kelires PC (2007) Atomic and

462: electronic structure of crystalline-amorphous carbon interfaces. Diam Relat

463: Mater 16:1875-1881

464: 

465: \bibitem{e} Krauss AR, Auciello O, Gruen DM, Jayatissa A, Sumant A, Tucek J,

466: Mancini DC, Moldovan N, Erdemir A, Ersoy D, Gardos MN, Busmann HG, Meyer EM,

467: Ding MQ (2001) Ultrananocrystalline diamond thin films for MEMS and moving

468: mechanical assembly devices. Diam Relat Mater 10:1952-1961

469: 

470: \bibitem{f} Meyers MA, Mishra A, Benson DJ (2006) Mechanical properties of

471: nanocrystalline materials. Progr Mater Sci 51:427-556

472: 

473: \bibitem{g} Pantea C, Zhang J, Qian J, Zhao Y, Migliori A, Grzanka E, Palosz B,

474: Wang Y, Zerda TW, Liu H, Ding Y, Stephens PW, Botez CE (2006) Nano-Diamond

475: compressibility at pressures up to 85 GPa, \textit{2006 NSTI Nanotechnology

476: Conference and Trade Show}, pp 823-826.

477: 

478: \bibitem{h} Remediakis IN, Fyta MG, Mathioudakis C, Kopidakis G, Kelires PC

479: (2007) Structure, elastic properties and strength of amorphous and

480: nanocomposite carbon. Diam Relat Mater 16:1835-1840.

481: 

482: \bibitem{i} Robertson J (2002) Diamond-like amorphous carbon. Mat Sci Eng R

483: 37:129-281

484: 

485: \bibitem{j} Schiotz J, Di Tolla FD, Jacobsen KW (1998) Softening of

486: nanocrystalline metals at very small grain sizes. Nature 391:561-563.

487: 

488: \bibitem{k} Schiotz J, Jacobsen KW (2003) A maximum in the strength of

489: nanocrystalline copper. Science 301:1357-1359

490: 

491: \bibitem{l} Szlufarska I, Nakano A, Vashishta P (2005) A crossover in the

492: mechanical response of nanocrystalline ceramics. Science 309:911-914.

493: 

494: \bibitem{m} Tersoff J (1988) Empirical interatomic potential for carbon, with

495: applications to amorphous-carbon. Phys Rev Lett 61:2879-2882

496: 

497: \bibitem{n} Van Swygenhoven H, Spaczer M, Caro A, Farkas D (1999) Competing

498: plastic deformation mechanisms in nanophase metals. Phys Rev B 60:22-25

499: 

500: \bibitem{o} Yamakov V, Wolf D, Phillpot SR, Mukherjee AK, Gleiter H (2004)

501: Deformation-mechanism map for nanocrystalline metals by molecular-dynamics

502: simulation. Nat Mater 3:43-47

503: 

504: \bibitem{p} Yip S (1998) Nanocrystals - The strongest size. Nature 391:532-533

505: 

506: \bibitem{q} Zapol P, Sternberg M, Curtiss LA, Frauenheim T, Gruen DM (2002)

507: Tight-binding molecular-dynamics simulation of impurities in

508: ultrananocrystalline diamond grain boundaries. Phys Rev B 65:045403

509: 

510: \end{thebibliography}

511: 

512: \end{document}

513: