Physica B 406 (2011) 4170–4174

Contents lists available at ScienceDirect

Physica B

0921-45

doi:10.1

n Corr

E-m

journal homepage: www.elsevier.com/locate/physb

Structure of diamond nanoparticles grown by chemical vapor deposition

Rajarshi Chakraborty, Suresh C. Sharma n

Department of Physics, University of Texas at Arlington, Arlington TX 76019 USA

a r t i c l e i n f o

Article history:

Received 8 March 2011

Received in revised form

12 July 2011

Accepted 14 July 2011Available online 29 July 2011

Keywords:

Diamond

Carbon

Hexagonal diamond

Nanoparticles

26/$ - see front matter & 2011 Elsevier B.V. A

016/j.physb.2011.07.025

esponding author. Tel.: þ1 18172722470; fa

ail address: sharma@uta.edu (S.C. Sharma).

a b s t r a c t

We studied the structure of diamond nanoparticles grown by chemical vapor deposition. SEM images

show that the material contains cubic, hexagonal, and possibly icosahedral structures ranging in size

from 10 to 200 nm. Raman spectroscopy shows bands, which are characteristic of crystalline diamond,

E2g mode of hexagonal diamond, a-C, and graphite.

& 2011 Elsevier B.V. All rights reserved.

1. Introduction

Because of the unique mechanical, electrical, and electronicproperties of diamond and added effects due to the quantumconfinement of optical phonons in nanoparticles (NPs), nan-ometer-size diamond particles have attracted significant interestin recent years [1–11]. Natural diamond exhibits the highesthardness and thermal conductivity at room temperature; it isoptically transparent starting from IR, through visible, and downto �250 nm into the UV spectral range. It is extremely resistive tochemical erosion, electrically a very good insulator (band gap�5.5 eV), and useful as a passivating coating for electronicapplications. The combination of these extraordinary materialproperties with the quantum confinement effects has catalyzednumerous already-in-use and potential applications of diamondNPs. Noteworthy, in particular, are as follows:(1) single photongeneration and individual electron spin manipulation at roomtemperature utilizing the nitrogen-vacancy centers in diamondNPs[12–16], (2) potential applications of biocompatible diamondNPs for targeted drug delivery[17,18], (3) fabrication of field-effect-transistors[19–21], and (4) improved mechanical propertiesof diamond NPs-reinforced polymer matrix composites[22,23].Because of recent developments in the growth of the diamondNPs, there is renewed interest in other possible structures, whichdiffer from the expected cubic diamond structure. An energeticallyunfavorable hexagonal diamond structure is an example of such astructure. The hexagonal diamond structure has been of interest,

ll rights reserved.

x: þ1 18172723637.

particularly since the first report of its synthesis from crystallinegraphite at high pressure and temperature (Z 130 kbar and1000 1C) [24]. As summarized in Table 1, the properties ofhexagonal diamond are significantly different from those of thecrystalline (cubic) diamond. Although, the C–C bonding in bothcubic and hexagonal structures is sp3, the stacking sequences aredifferent. Whereas it is ‘‘ABCABCy’’ in the commonly observedcubic structure, it is ‘‘ABABy’’ in hexagonal diamond. Thesestructures are further characterized by: (i) bond lengtha¼1.545 A for cubic diamond and a¼2.52 A and c¼4.12 A forhexagonal diamond, (ii) calculated band gaps of 5.6 and 4.5 eV forthe cubic and hexagonal structures, respectively [25], and (iii)relative stability (hexagonal being less stable), hardness (hexago-nal harder than cubic diamond), and different vibrational spectra[26,27]. Besides the high-pressure/high-temperature technique,certain variants of chemical vapor deposition (CVD) have beenattempted, albeit scarcely to grow hexagonal diamond. Among thelimited successful attempts using CVD related techniques, are theworks by: (1) Lu and Chang,[28] who used microwave plasmaCVD, (2) Zhu et al.,[29] who used thermal evaporation of C60 withsimultaneous bombardment with 1.5 keV Neþ beam, (3) Silvaet al.,[30]who used 13.56 MHz radio frequency assisted CVD ,(4) Maruyama et al.[31], who utilized hydrogen plasma jet , and(5) Bhargava et al., who used CVD and microwave plasma to growE1 mm thick films [32]. However, to our knowledge hexagonaldiamond NPs (10odiametero200 nm) have been seen onlyrarely in the material grown by the simplest of the techniques,i.e., hot-filament-assisted CVD (HFCVD), which we have utilized togrow diamond NPs [5]. HFCVD offers obvious advantages ofsimplicity and safety over the detonation and shock wave techni-ques. Additionally, it provides reasonably good control over the

Table 1Selected properties of crystalline (cubic) and hexagonal diamond.

Property c-Diamond h-Diamond

Structure Cubic Hexagonal

Stacking ABCABC ABABAB

Lattice constant (A) 3.567 a¼2.522, c¼4.119

Band gap (eV) 5.5 [48] 4.5 [25]

Raman active modes (cm�1) 1332

[34,42]

A1g 1312 E1g 1305 E2g 1193,

1199 [26,29]

XRD peak positions, d-spacing (A),

and crystallographic planes (for

CuKa X-ray source)

43.934,

2.0592

(111) [39]

41.37, 2.1824 (100) 43.95,

2.0600 (111) 47.12, 1.9285

(101)[39,24]

R. Chakraborty, S.C. Sharma / Physica B 406 (2011) 4170–4174 4171

size of the NPs without need for post-growth chemical cleanup ofthe material. However, the growth rates are relatively low (onlyabout 200 nm/h realized in our experiments). In this recentlypublished work, our focus was on the quality of the material(crystalline diamond vs. other allotropes of carbon), NPs sizedistribution and growth rate.[5]. We were interested primarily indemonstrating that the HFCVD technique can be used to growdiamond NPs, whose overall size is controlled reasonably well bythe growth period. We have carried out additional SEM, XRD, andRaman spectroscopy measurements to better understand thestructure of the grown material. These measurements haveprovided new results on the structure of the NPs and they showthat the HFCVD grown material contains not only the cubicdiamond, but also hexagonal diamond and possibly other icosahe-dral carbon structures ranging in size from 10 to 200 nm.

Fig. 1. The surface morphology of sample SH77-4 measured by AFM with

resolution {5 nm.

2. Experimental details

Since our HFCVD system and characterization techniques havealready been discussed elsewhere [5,33], we provide only a briefdescription of the most relevant points. Diamond NPs anddiamond thin films were grown using HFCVD and characterizedby utilizing several different techniques [34]. As discussed else-where, we have recently reported the HFCVD growth of diamondNPs, ranging in size from 10 to 200 nm, in samples grown forperiods of 15, 30, 43, and 105 min.[5] There are essentially foursteps involved in the growth process: (1) preparing the feed gasmixture, (2) carbonization of the filament, (3) substrate prepara-tion, and (4) materials’ growth. Among these, the carbonization ofthe filament, as well as substrate preparation are specific enoughto be included in this description. For carbonization of the Tafilament, CH4/H2 mixture (r 2% CH4) was passed through theCVD chamber at a flow rate �550 sccm at �30 Torr, while thefilament was maintained at approximately 2100 K. In about15 min, the filament is coated with carbon and it changes fromsilver/gray to golden appearance. The pretreatment of the sub-strate involves scratching the substrate with commercially avail-able diamond powder (99.9% pure). The distribution of the sizesof the particles in the powder can be described by a Lorentzianfrequency distribution function given by

FðdÞ ¼ ðf 20 =g

2Þg2=4

fðd�d0Þ2þg2=4g

ð1Þ

where F is the frequency of the occurrence for a particular size d,the maximum value of the frequency (f 2

0 =g2) occurs at d¼d0, andthe full-width at half-maximum (FWHM) of the distribution is g.The mean particle size in the powder was about 0.5370.25 mmand there were no particles of sizes less than about 0.17 mm. Thesubstrate was rubbed with this powder to create nucleation sites,washed in deionized water, cleaned with acetone in ultrasonicbath, rinsed with methanol, and thereafter cleaned with

deionized water in the ultrasonic bath. This process ensured thatwhile the nucleation sites were created, no residual diamondparticles were left on the substrate. The diamond growth on thusprepared substrate was realized by continuously flowing CH4/H2

mixture (flow rate �250 sccm at �25 Torr) over carburized Tafilament running at �2000 K. The substrate was maintained atabout 1100 K. The surface topography and structure of the grownmaterial were examined using Veeco Multimode V-SPM atomicforce microscopy system (resolution �5 nm in tapping mode)and Zeiss Supra-55-VP SEM instrument (resolution of 1 nm at15 kV). The crystal structure was studied by making XRD mea-surements with a Bruker D8 Diffractometer using Cu-ka radiationsource (l¼ 1:5418 A). The Raman spectroscopy measurementswere made using Horiba Jobin Yvon LABRAM system equippedwith 633 nm He–Ne laser and CCD camera. The resolution of thespectrometer is characterized by the FWHM (3.48 cm�1) of the1332 cm�1 diamond line. The performance of the spectrometerwas carefully evaluated by making measurements on severalsamples with well-known Raman lines, for example, the1332 cm-1 line of diamond, 521 cm�1 line of single crystal silicon,and several prominent bands between 1300 and 1700 cm�1 ofsingle-walled carbon nanotubes, etc.[35].

3. Results and discussion

Fig. 1 shows an AFM scan illustrating the surface morphology ofone of the samples (SH77-4, growth period �105 min). As statedabove, we had examined four different samples, which were grownfor 15, 30, 43, and 105 min. The common feature among thesesamples is the presence of diamond NPs, whose size varies withgrowth period [5]. Here, we focus on the 105 min SH77-4 sample,which contains the largest NPs and thereby shows with relativeease that the material contains cubic, as well as hexagonal diamondNPs. At the resolution of the AFM used (r5 nm), diamond particlesappear to be uniformly spread with heights ranging from approxi-mately 10 to 75 nm. Additional details on the structure of the NPsare seen in the higher resolution SEM image of Fig. 2. Usingmagnified SEM images, we have examined sizes and shapes ofapproximately 200 particles. In these images, not only can indivi-dual particles be identified, different shapes (cubic and hexagonal)can also be recognized. For example in Fig. 3, we show magnifiedimage of one of the hexagonal particles in a sample grown for 30minutes, which is about 50 nm in diameter. Using magnified SEMimages, we also obtain information on the sizes of the particles.Fig. 4 shows size distributions for both the cubic and hexagonalNPs. The cubic and hexagonal NPs are represented by the dark andlightly shaded hatched bars, respectively. The Lorentzian function ofEq. (1) provides good representation with an average size of about(190725) nm for the cubic NPs. In comparison, the average size ofthe hexagonal NPs is estimated to be much smaller (r150 nm).

Fig. 2. SEM image of sample SH77-4 measured at 10 kV with a magnification of 25,000� ; cubic, hexagonal, and icosahedral NPs can be identified.

Fig. 3. A hexagonal particle is clearly seen in this magnified SEM image. As shown by the pair of vertical lines, the hexagonal particle is about 50 nm across.

R. Chakraborty, S.C. Sharma / Physica B 406 (2011) 4170–41744172

The sample contains relatively large number of hexagonalNPs (Z10%) of diameters 50rDr400 nm. Although, we do notobserve hexagonal NPs smaller than 50 nm, their presence cannotbe ruled out because of the resolution of the optical micrographs.The SEM micrographs also contains what appears to be one or twoparticles with five-fold symmetry. Such particles have beenobserved and interpreted to arise as a result of multiple twinningof cubic crystals. For example, Shevchenko and Madison[36–38]have shown that icosahedral carbon NPs can be formed inwhich the local environment of the carbon atoms is virtuallyidentical to that of carbon atoms in diamond. Breza et al. also haveobserved icosahedral and decahedral diamond NPs in samplesgrown on Tin-coated steel substrates[38].

In order to further discern the crystal structures, we havecarried out XRD measurements. A typical XRD spectrum is shownin Fig. 5, and it shows prominent peaks centered at 2y¼44.0 and44.4. Several sets of XRD scans were made for extended periodson different days (i.e., each scan made for 4 h). Because of therelatively small amount of the NPs on the substrate, the signal tonoise ratios in the XRD scans were much lower than thoseexpected and typically observed in the cases of the bulk samples.Nonetheless the two diffraction peaks seen at 2y¼44.0 and 44.4were present in each of several scans made with varying resolu-tion and collection times. Based on theoretical calculations(summarized in Table 1), it is known that the (111) reflectionsin cubic diamond (d-spacing¼2.0592 A) and (002) reflections in

Fig. 4. Particle size distribution for about 200 particles resolved in magnified SEM

image. The cubic and hexagonal NPs are represented by the dark and light-shaded

bars, respectively. The size distributions are well-represented by the Lorentzian

distribution function of Eq. (1) with mean size �190 nm for the cubic and

r150 nm for hexagonal particles.

44.0

44.4

Rel

ativ

e In

tens

ity (A

U)

50

45

40

35

30

25

20

15

10

5

0

2 Theta40 41 42 43 44 45 46 47 48 49 50

Fig. 5. Typical XRD spectrum for sample SH77-4 showing peaks at 2y¼44.01 and

44.41. As discussed in the text, the 2y¼44.0 peak has contributions from both the

cubic and hexagonal diamond NPs. The possible origin of the peak at 2y¼ 44.31 is

discussed in the text.

35

30

25

20

15

10

5

0

Rel

ativ

e In

tens

ity

Wavenumber (cm–1)

1200 16501150 1250 1300 1350 1400 1450 1500 1550 1600

Fig. 6. Raman spectrum of sample SH77-4. The spectrum is resolved into

components centered at approximately 1205, 1336, 1337, 1436, and 1538 cm�1.

The assignments of these bands are discussed in the text. In particular, it is

discussed that the 1205 cm�1 and the 1336 cm�1 bands originate from the E2g

mode of hexagonal diamond and crystalline diamond having defects, respectively.

R. Chakraborty, S.C. Sharma / Physica B 406 (2011) 4170–4174 4173

hexagonal diamond (d-spacing¼2.0593 A) appear at essentiallythe same 2y values, i.e., 2y¼43.934 and 43.932, respectively [39].Based on our observations from the SEM images that the materialcontains cubic, hexagonal, and possibly icosahedral NPs, it is notunreasonable to believe that the XRD peak at 2y¼44.0 representscontributions from cubic, as well as hexagonal diamond NPs.Furthermore, it is known theoretically that the (011) reflectionsfrom 4H-diamond, and the (010) and (011) planes in hexagonaldiamond give rise to peaks at 2y¼42.798 (d-spacing¼2.1112 A),2y¼41.301 (d-spacing¼2.1842 A), and 2y¼47.056 (d-spacing¼1.9296 A), respectively [39]. There is also experimental evidencethat the peak at 2y¼44.4 originates from hexagonal graphite [40].Relatively low intensity of the X-ray beam encountering rathersmall number of diamond NPs (even smaller number of hexagonalNPs) makes it difficult to discern more definitive information onthe crystal structure of the sample. However, in the light of theobservations of the hexagonal and icosahedral structures in SEMimages and further confirmation by Raman spectroscopy of

contributions from the E2g mode of hexagonal diamond (dis-cussed below), it is almost certain that the sample contains cubicand hexagonal diamond NPs. As stated above, in addition to the2y¼44.0 peak, the XRD spectrum also shows a weak, butreproducible peak at 2y¼44.4. We do not know the precise originof this peak. Although, the calculations by Ownby et al. [39] forXRD patterns of different diamond polytypes predict peaks at2y¼44.6 for 8H-diamond and at 2y¼44.7 for 21R-diamond. .Notonly is the observed peak weak, the difference in the observedand calculated 2y values for the 8H and 21R diamond polytypes istoo large to reach any definitive conclusions on the presence ofthese other polytypes in our sample.

In order to characterize different allotropes of carbon and inparticular to re-affirm that the sample contains hexagonal diamondNPs, we have carried out Raman spectroscopy measurements. Fig. 6shows the Raman spectrum measured for sample SH77-4. Thisspectrum shows several bands centered at 1205 cm�1 (FWHM�46 cm�1), 1336 cm�1 (FWHM �8.9 cm�1), 1337 cm�1 (FWHM�67 cm�1), 1436 cm�1 (FWHM �95 cm�1), and 1538 cm-1 (FWHM�99 cm�1). In our earlier works, we have shown that Raman spectraof CVD grown diamond films exhibit several prominent features,which are sensitive to the growth parameters [34]. These prominentfeatures include: (1) a sharp peak at 1332 cm�1, which is character-istic of the crystalline diamond, (2) a broad band near 1550 cm�1,which identifies diamond-like carbon (DLC), and (3) two bands at�1350 and 1580 cm�1 representing polycrystalline graphite oramorphous carbon (G- and D-bands, respectively) [34,41,42,10]. It isalso known that in samples containing microcrystallites smaller insize than the phonon decay lengths ({1 mm), an increase in thephonon scattering off grain boundaries gives rise to lifetime-broa-dened peaks, which are shifted from the 1332 cm�1 line [43,44]. Inthis context, both Sun et al. [44] and Ager et al. [43] have observeddiamond peaks shifted from the characteristic 1332 cm�1 line by asmuch as 13 cm�1 and broadened to about 80 cm�1. From themeasurements of the positron diffusion length, which is shorteneddue to the positron trapping in lattice defects, we have shown thatthe CVD grown diamond samples contain lattice vacancies andmicrovoids with relative concentrations ranging from 10–5 to 0.1[45]. Additionally, it is known that: (1) the quantum confinement of

R. Chakraborty, S.C. Sharma / Physica B 406 (2011) 4170–41744174

the phonons within NPs causes a shift and broadening in the Ramanline [46] and (2) internal stresses due to a lattice mismatch betweenthe film and the substrate can also cause small wavenumber shifts.For example, in the case of the SiO2 substrate (the Si substrate used inour experiments probably had SiO2 over layers), the diamond line hasbeen observed to shift from 1332 cm�1 to 1337 cm�1 [47]. It cantherefore be argued that the Raman spectrum of Fig. 6 representscontributions from: (1) crystalline diamond (shifted to 1336 cm�1

because of the above-discussed effects), (2) hexagonal nanodiamond(1205 cm�1) [6], (3) nanodiamond (1337 and 1436 cm�1) [6], and(4) graphitic carbon (G-band, 1538 cm�1). The following discussionprovides additional justifications for this understanding. Using a first-principles method based on density functional theory, Wu and Xu[26] have calculated vibrational frequencies at the G point for bothcubic and hexagonal diamond. As far as the first-order zone-centermode of cubic diamond is concerned, there is exceedingly goodagreement between the calculated (1334 cm�1) and observed fre-quencies (1332 cm�1). For hexagonal diamond NPs, there are nineoptical modes; three nondegenerate modes, A1g, B2g, and B1u andthree doubly degenerate modes, E1g, E2g, and E1u. Among these, onlyA1g, E1g, and E2g are Raman active modes. The theoretically calculatedfrequencies of the A1g, E1g, and E2g Raman active modes in hexagonaldiamond are 1312, 1305, and 1193 cm�1, respectively [26]. Although,we do not see Raman bands at 1305 and 1312 cm�1, we observe arelatively strong band centered at 1205 cm�1, which as discussedabove, represents Raman active E2g mode of hexagonal diamond NPs.Our observation of the E2g mode of hexagonal diamond NPs at1205 cm�1 is in reasonably good agreement with the results ofZhu et al.,[29] who have observed a strong line at 1199 cm-1 inthe Raman spectrum of hexagonal nanodiamond grown by ion-beam-assisted CVD.

4. Conclusions

In this work, we have extended our investigations of the structureof the diamond NPs grown by chemical vapor deposition. After about105 min of growth, numerous diamond NPs of different sizes areobserved on the substrate. A Lorentzian function, peaked at around190 nm, can be used to describe the distribution of the particle sizes.The visual inspection of SEM images reveals the growth of cubic,hexagonal, and 5-fold symmetry particles. Using an SEM image, inwhich about 200 NPs are identified, we estimateZ10% hexagonalNPs of diameters ranging from 50 to 400 nm. We present supportingevidence for the growth of hexagonal diamond NPs from XRD andRaman spectroscopy measurements.

Acknowledgments

The authors acknowledge the Center for NanostructuredMaterials and Dr. Muhammed Yousufuddin for assistance withsome of the measurements.

References

[1] V.V. Danilenko, Phys Solid State 46 (2004) 595.

[2] G.P. Bogatyreva, M.N. Voloshin, V. Padalko, Diamond Relat. Mater. 17 (2008)213.

[3] O. Shenderova, C. Jones, V. Borjanovic, S. Hens, G. Cunningham, S. Moseenkov,

V. Kuznetsov, G. McGuire, Phys. Status Solidi A 205 (2008) 2245.[4] W.S. Yeap, S.M. Chen, K.P. Loh, Langmuir 25 (2009) 185.[5] R. Chakraborty, S.C. Sharma, J.K. LaRoque, J. of Nano Res. 12 (2010) 123.[6] J.J. Gracio, Q.H. Fan, J.C. Madaleno, J. Phys. D :Appl. Phys. 43 (2010).[7] Y.B. Zhou, Y. Yang, W.M. Liu, Q. Ye, B. He, Y.S. Zou, P.F. Wang, X.J. Pan,

W.J. Zhang, I. Bello, S.T. Lee, Appl. Phys. Lett. 97 (2010).[8] E. Osawa, Chemistry of single-nano diamond particles, in: T. Akasaka, F. Wudl

S. Nagase (Eds.), Chemistry of Nanocarbons John Wiley, 2010.[9] F. Fendrych, A. Taylor, L. Peksa, I. Kratochvilova, J. Vlcek, V. Rezacova,

V. Petrak, Z. Kluiber, L. Fekete, M. Liehr, M. Nesladek, J. Phys. D: Appl. Phys.

43 (2010).[10] F. Klauser, D. Steinmuller-Nethl, R. Kaindl, E. Bertel, N. Memmel, Chem. Vapor

Deposition 16 (2010) 127.[11] O.A. Williams, M. Nesladek, M. Daenen, S. Michaelson, A. Hoffman, E. Osawa,

K. Haenen, R.B. Jackman, Diamond Relat. Mater. 17 (2008) 1080.[12] B. Lounis, M. Orrit, Rep. Prog. Phys. 68 (2005) 1129.[13] S. Kagami, Y. Shikano, K. Asahi, Physica E 43 (2011) 761.[14] B.J. Maertz, A.P. Wijnheijmer, G.D. Fuchs, M.E. Nowakowski, D.D. Awschalom,

Appl. Phys. Lett. 96 (2010).[15] J. Tisler, G. Balasubramanian, B. Naydenov, R. Kolesov, B. Grotz, R. Reuter,

J.P. Boudou, P.A. Curmi, M. Sennour, A. Thorel, M. Borsch, K. Aulenbacher,R. Erdmann, P.R. Hemmer, F. Jelezko, J. Wrachtrup, ACS Nano 3 (2009) 1959.

[16] S. Schietinger, M. Barth, T. Alchele, O. Benson, Nano Lett. 9 (2009) 1694.[17] A. Krueger, Adv. Mater. 20 (2008) 2445.[18] A. Krueger, Chem.Eur. J. 14 (2008) 1382.[19] R.K. Ahmad, A.C. Parada, R.B. Jackman, Appl. Phys. Lett. 98 (2011).[20] S. Myung, S. Woo, J. Im, H. Lee, Y.S. Min, Y.K. Kwon, S. Hong, Nanotechnology

21 (2010).[21] R.J. Narayan, R.D. Boehm, A.V. Sumant, Mater. Today 14 (2011) 154.[22] U. Maitra, K.E. Prasad, U. Ramamurty, C.N.R. Rao, Solid State Commun. 149

(2009) 1693.[23] K.D. Behler, A. Stravato, V. Mochalin, G. Korneva, G. Yushin, Y. Gogotsi, ACS

Nano 3 (2009) 363.[24] F.P. Bundy, J.S. Kasper, J. Chem. Phys. 46 (1967) 3437.[25] M.R. Salehpour, S. Satpathy, Phys. Rev. B 41 (1990) 3048.[26] B.R. Wu, J. Xu, Phys. Rev. B 57 (1998) 13355.[27] Z.C. Pan, H. Sun, Y. Zhang, C.F. Chen, Phys. Rev. Lett. 102 (2009).[28] C.A. Lu, L. Chang, Mater. Chem. Phys. 92 (2005) 48.[29] X.D. Zhu, Y.H. Xu, H. Naramoto, K. Narumi, A. Miyashita, K. Miyashita, J. Phys.

Condens. Matter 15 (2003) 2899.[30] S.R.P. Silva, G.A.J. Amaratunga, E.K.H. Salje, K.M. Knowles, J. Mater. Sci. 29

(1994) 4962.[31] K. Maruyama, M. Makino, N. Kikukawa, M. Shiraishi, J Mater. Sci. Lett. 11

(1992) 116.[32] S. Bhargava, H.D. Bist, S. Sahli, M. Aslam, H.B. Tripathi, Appl. Phys. Lett. 67

(1995) 1706.[33] M. Green, , Deposition of diamond films and comparative characterization

by Raman, X-ray photoelectron and positron annihilation spectroscopies,University of Texas, Arlington, 1991.

[34] S.C. Sharma, M. Green, R.C. Hyer, C.A. Dark, T.D. Black, A.R. Chourasia,D.R. Chopra, K.K. Mishra, J. Mater. Res. 5 (1990) 2424.

[35] S.C. Sharma, D. Singh, Y. Li, J. Raman Spectrosc. 36 (2005) 755.[36] V.Y. Shevchenko, A.E. Madison, Glass Phys. Chem. 32 (2006) 118.[37] V.Y. Shevchenko, A.E. Madison, A.L. Mackay, Acta Crystallogr. Sect. A 63

(2007) 172.[38] J. Breza, M. Kadlecikova, M. Vojs, M. Michalka, M. Vesely, T. Danis, Micro-

electron. J. 35 (2004) 709.[39] P.D. Ownby, X. Yang, J. Liu, J. Am. Ceram. Soc. 75 (1992) 1876.[40] V. Amornkitbamrung, I. Burinprakhone, W. Jarernboon, Surf. Coat. Technol.

203 (2009) 1645.[41] W. Kiefer, J. Raman Spectrosc. 38 (2007) 1538.[42] A.C. Ferrari, J. Robertson, Philos. Trans. R. Soc. A 362 (2004) 2477.[43] J.W. Ager, D.K. Veirs, G.M. Rosenblatt, Phys. Rev. B 43 (1991) 6491.[44] K.W. Sun, J.Y. Wang, T.Y. Ko, J. Nanopart. Res. 10 (2008) 115.[45] R.C. Hyer, M. Green, S.C. Sharma, Phys. Rev. B 49 (1994) 14573.[46] H. Richter, Z.P. Wang, L. Ley, Solid State Commun. 39 (1981) 625.[47] N. Ali, Q.H. Fan, J. Gracio, W. Ahmed, Surf. Eng. 18 (2002) 260.[48] F.J. Himpsel, J.A. Knapp, J.A. Vanvechten, D.E. Eastman, Phys. Rev. B (1979)

624.