Sn

Va

b

a

ARRA

KNSCPX

1

nspsghottpwWs[os6sa

m

1d

Spectrochimica Acta Part A 92 (2012) 64– 66

Contents lists available at SciVerse ScienceDirect

Spectrochimica Acta Part A: Molecular andBiomolecular Spectroscopy

j our na l ho me p age: www.elsev ier .com/ locate /saa

tudy of pressure induced phase transformation in CTAB capped CdSanoparticles

ineet Kumar Singha,∗, Pratima Chauhana, K.K. Pandeyb

Department of Physics, University of Allahabad, Allahabad 211002, IndiaHigh Pressure & Synchrotron Radiation Physics Division, Bhabha Atomic Research Center, Mumbai, India

r t i c l e i n f o

rticle history:eceived 12 December 2011eceived in revised form 8 February 2012ccepted 9 February 2012

a b s t r a c t

Effect of high pressure on as prepared 20 mM CTAB capped CdS nanoparticles (size ∼4 nm) has beenanalyzed in this paper. Raman scattering has been used to observe the phase transition pressure. X-raydiffraction pattern is used for structural characterization. Raman scattering predicts the phase transitionoccur from mixed cubical phase to rock salt phase above 6.6 GPa. One of the representative XRD pattern

eywords:anostructured materialsemiconductorshemical synthesishase transformation-ray diffraction

at 9.7 GPa confirms the existence of rock salt phase above 6.6 GPa.© 2012 Elsevier B.V. All rights reserved.

. Introduction

CdS occurs in nature with three different crystal structuresamely as hexagonal wurtzite, cubical zinc blende and high pres-ure rock salt phase. Hexagonal wurtzite phase occurs only in bulkhase while cubical zinc blend structure and high pressure rockalt phase exist in the nanocrystalline phase [1–5]. CdS under-oes a transition from the four-coordinate cubical zinc blende orexagonal wurtzite structures to the six-coordinate rock salt phasen the application of moderate pressure [6]. Earlier reports revealhat bulk CdS wurtzite crystals transform to the rock salt struc-ure at 2.7–3.0 GPa [6–8]. Corll has experimentally predicted thathase transition pressure is very similar for zinc blende (3.1 GPa),urtzite (2.5 GPa) and zinc blende-wurtzite mixtures (2.8 GPa) [8].hile in the case of CdS nanocrystals the zinc blende to rock salt

olid–solid phase transition takes place at much higher pressure6–11]. Several literature reports have confirmed the dependencef transition pressure on the particle size as well as on the matrixurroundings the particles [6–11]. The CdS nanoparticles having

.9 nm sizes in SiO matrix show the phase change at transition pres-ure of 4.51 GPa while particles of 3.4 nm show the phase transitiont 4.92 GPa [9,10]. However in the case of GeO2 matrix CdS particles

∗ Corresponding author. Tel.: +91 9451369833; fax: +91 532 2460993.E-mail addresses: 1vineetkumarsingh@gmail.com (V.K. Singh),

angu167@yahoo.co.in (P. Chauhan).

386-1425/$ – see front matter © 2012 Elsevier B.V. All rights reserved.oi:10.1016/j.saa.2012.02.024

of 50 nm particle size show phase transition at 6.35 GPa while thatof 6.9 nm show phase transition at 6.98 GPa [9,10]. This shows thatthe phase transition pressure of CdS particles embedded in GeO2glasses does not depend strongly on the particle size. This dras-tic change in behavior of phase transition pressure is assumed dueto the difference in nature of SiO and GeO2 matrix. SiO matrix ismore flexible than GeO2 glasses because of polymeric nature of SiO[9,10]. Makino et al. also reported that GeO2 glass is much harderthan the SiO. So the difference in bulk modulus between micro-crystals and matrix is one of the important factors for elevation ofphase transition pressure [10].

Alivisatos et al. reported that the surface energy of an inorganicnanocrystal can play a dominant role in the relative stability ofthe phases of the system. In their experiment they have synthe-sized the 4 nm CdS nanocrystals by either sodium polyphosphate ortetrabutylammonium-EDTA (tba-EDTA) as the surface stabilizingagent. High pressure Raman measurements show that the upstrokezinc blende to rock salt phase transition pressure is 8.0 GPa with thepolyphosphate stabilization, and 6.5 GPa with the tba-EDTA stabi-lization [6]. The increase in solid–solid transition pressure may beassigned to an increase in surface energy in the high-pressure phasenanocrystals [6–20]. Arai and co-workers reported the blue shiftin absorption edge on increasing the pressure. They also quantita-

tively point out that wurtzite structure has direct optical transitionhowever rock salt structure has indirect transition [6,9,10]. Inthe present work we report the effect of CTAB capping on thephase transition pressure of CdS nanoparticles. To the best of our

V.K. Singh et al. / Spectrochimica Acta Part A 92 (2012) 64– 66 65

Fs

ks

2

atMiathmtssoYs

3

acttduhww

Table 1Shift in 1LO peak position of CTAB capped CdS nanoparticles with pressure.

Pressure (GPa) Position of 1LO (cm−1)

Ambient pressure (0.0001) 303.491.3 307.345.2 326.586.6 332.828.4 305.7015.1 332.16Completely released 302.35

position at different pressures. In order to know the structure ofnew phase we carried out high pressure X-ray diffraction mea-surements. Fig. 3 shows the XRD pattern of CdS nanoparticles at

ig. 1. Raman spectra of CdS nanoparticles at different upstroke and released pres-ures.

nowledge this is the first report on pressure induced phase tran-ition effect of CTAB capped CdS.

. Materials and methods

The synthesis of 20 mM CTAB capped CdS nanoparticles withverage size ∼4 nm is given in Ref. [3]. High pressure X-ray diffrac-ion and Raman scattering measurements were carried out using

ao-Bell kind of diamond anvil cell (DAC). The sample was loadedn ∼100 �m sample chamber hole of pre-indented steel gasketlong with 4:1::methanol:ethanol mixture as pressure transmit-ing medium and ∼20 �m Ruby chip as pressure marker [20]. Theigh pressure XRD images were recorded using rotating anodeolybdenum X-ray source and MAR345 imaging plate. 2D diffrac-

ion images were converted to 1D diffraction pattern using FIT2Doftware [21] which were then analyzed with the help of GSASoftware [22]. Raman spectra at high pressures were recordedn con-focal micro Raman setup configured around HORIBA Jobinvon spectrograph. 532 nm frequency doubled diode pumped solidtate ND:YAG laser was used as excitation source.

. Results and discussion

Under ambient conditions, 20 mM CTAB capped CdS sample has mixture of cubical and hexagonal phases as has been briefly dis-ussed by Singh et al. [3]. Similar to earlier reports, it has been foundhat CTAB capped CdS nanoparticles also undergo structural phaseransformation at high pressures. Fig. 1 shows the Raman spectra atifferent pressures. 1LO Raman mode of mixed phase CdS is visible

p to 6.6 GPa. On the upstroke, 1LO mode frequency shifted towardsigher side. At ambient pressure, 1LO is observed at 303.49 cm−1

hile at 6.6 GPa the position of 1LO is found to be at 332.82 cm−1

hich is nearly 29 cm−1 higher than ambient pressure position as

Fig. 2. Raman peak position at different pressures.

shown in Table 1. Least square fitting of straight line yields a slopeof dω/dP = 4.6 ± 0.2 cm−1/GPa for 1LO mode. Above this pressureRaman mode of mixed phase disappears. Similar results are alsoreported for CdS nanocrystal by Alivisatos et al. for the polyphos-phate and tba-EDTA surface stabilizing reagent, however dω/dPwas reported to be 5 ± 0.2 cm−1/GPa which is comparable withbulk CdS even though the particle size was 4 nm [6]. At 8.4 GPa,a new mode appears at 305 cm−1, indicating the solid–solid phasetransformation of mixed phase. Fig. 2 represents the Raman peak

Fig. 3. XRD spectrum of CdS nanoparticles at 9 7 GPa confirms the rock-salt phaseof CdS. An unknown peak at 2� = 18.79◦ is also observed in the XRD pattern.

6 imica

ar(1agiamtbipistwr

psrtpr

shttmctfsctritWetttnft

ifisTlt

[

[[[

[[[[[[

[

6 V.K. Singh et al. / Spectroch

pressure of 9.7 GPa. The XRD pattern confirms the existence ofock salt phase due to presence of (1 1 1), (0 0 2), (0 2 2), (1 1 3) and2 2 2) diffraction planes corresponding to the 2� values of 12.9,4.9, 21.2, 24.9 and 26.1◦. The diffraction planes (1 1 0) and (2 0 0)t 2� values of 20.1 and 28.5 belong to the Fe due to presence of steelasket. Since first order Raman scattering of the rock salt phase isnactive [6,7,9,10], the emergence of new mode in Raman Spectrat and above 8.4 GPa could be due to the disorder activated, sym-etry forbidden, scattering from the rocksalt phase [23]. Wurtzite

o rocksalt phase transition is known to proceed via an orthorhom-ic path. One of the fundamental aspects of this phase transition

n general is the large reduction in volume that occurs. Since thisath has arbitrariness about the resulting relationship between the

nitial wurtzite a/b plane crystallographic axes and the final rock-alt ones. One would therefore expect a high degree of disorder inhe resulting (0 0 1) plane. This poor long range order in the latticeould allow relaxation of k-selection rules in the Raman scattering;

esulting in disorder activated Raman scattering.On releasing pressure the rock salt phase is sustained till the

ressure of 2 GPa. However below 2 GPa pressure 1LO active Ramancattering mode of mixed phase reappears which indicates theecovery of the original phase of CdS nanoparticles. It is interestingo note that earlier reports have shown poor recovery of ambienthase for smaller nanoparticles and preservation of high pressureock salt phase even at atmospheric pressure [9].

Our measurements show mixed cubical zinc blende to rock-alt transformation in CTAB capped CdS nanoparticles at muchigher pressure compared to bulk system. This can be explained inerms of surface energies. Surface tension plays an important role inhe elevation of transformation pressure for nano-particles. Nano-

aterials possess a large fraction of surface atoms per unit volumeompared to bulk material therefore, the internal energy is alteredo incorporate surface contribution. The phase transition pressureor nanocrystals of a given size will depends on the difference in theurface tension between the parent and daughter phases, in thisase mixed phase and rock salt [6,7,12–15]. In general, the surfaceension of a solid is difficult to define. Unlike liquids, where equilib-ium shape is spherical, the equilibrium shape of a solid crystallites the one in which each face is present in the inverse proportiono the magnitude of its surface tension. This shape is given by the

ulff construction. Based on these arguments it has been shownarlier that rock salt phase crystallites will have a larger surfaceension than the zinc blende phase as a larger fraction of bonds athe surface will be broken in the rock salt phase [6]. Other thanhe difference in surface energies of pure phases; capping of CdSanoparticles with CTAB will have additional contribution to sur-

ace energy. All these effects would lead to the elevation of phaseransformation pressure.

Besides surface tension effect, crystalline defects and the strainnside the nanocrystal also play a vital role in the elevation of trans-ormation pressure. It has been shown earlier that due to largernhomogeneous strain in parent phase of CdS, the transition pres-

ure elevates to ∼4.4 GPa even for particles as large as 75 nm [9].he CdS nanoparticles used in the present study have several defectevels inside the crystal as is clear from the difference in absorp-ion edge and photoluminescence spectra compared to that of bulk

[

[

[

Acta Part A 92 (2012) 64– 66

CdS system [3]. Bulk CdS has absorption edge at 515 nm whereas20 mM CTAB capped CdS has at 450 nm [3]. Further, the photolu-minescence spectra of 20 mM CTAB capped CdS has major greenemission band at 535 nm and a minor emission shoulder at 570 nmat 400 nm and 430 nm excitation wavelengths, discussed in Ref. [3].These observations indicate the presence of several defect levelsinside the crystal. So, the phase transition pressure of CTAB cappedCdS nanoparticles are not only affected by surface tension but alsoby crystalline defects and the strain inside the nanocrystals [9,10].

4. Conclusion

We have studied the pressure induced structure phase transi-tion of CTAB capped CdS nanoparticles by Raman scattering andX-ray diffraction pattern. Structural phase transition from mixedcubical phase to rock salt phase is observed above 6.6 GPa whichis far above the bulk CdS transition pressure of 2.7 GPa. The cap-ping reagent (CTAB) changes the value of the particle size as wellas surface tension of the bare CdS nanocrystal and introduces addi-tional defect levels inside crystal which accounts for the increasedphase transition pressure in the stabilized nanocrystals. On releas-ing the pressure the rock salt phase is stable only till 2 GPa. Belowthis pressure initial mixed phase reemerges.

Acknowledgment

One of the authors Vineet Kumar Singh is thankful to councilof Scientific and Industrial Research, India, for providing SeniorResearch Fellowship (CSIR-SRF).

References

[1] G. Cao, Nanostructures and Nanomaterials, Synthesis, Properties and Applica-tions, Imperial College press, London, 2004.

[2] R. Banerjee, R. Jaykrishnan, P. Ayyub, J. Phys. Condens. Matter 12 (2000)10647–10654.

[3] V. Singh, P.K. Sharma, P. Chauhan, Mater. Chem. Phys. 121 (2010) 202–207.[4] V. Singh, P. Chauhan, J. Phys. Chem. Solids 70 (2009) 1074–1079.[5] V. Singh, P.K. Sharma, P. Chauhan, Mater. Charact. 62 (2011) 43–52.[6] M. Haase, A.P. Alivisatos, J. Phys. Chem. 96 (1992) 6756–6762.[7] S.H. Tolbert, A.P. Alivisatos, Annu. Rev. Phys. Chem. 46 (1995) 595–625.[8] J.A. Corll, J. Appl. Phys. 35 (1964) 3032–3033.[9] T. Arai, T. Makino, M. Arai, K.K. Matsuishi, S. Onari, J. Phys. Chem. Solids 56

(1995) 491–494.10] T. Makino, K. Matsuishi, S. Onari, T. Arai, J. Phys. Condens. Matter 10 (1998)

10919–10930.11] M. Wilson, P.A. Madden, J. Phys. Condens. Matter 14 (2002) 4629–4643.12] S.H. Tolbert, A.P. Alivisatos, Science 265 (1994) 373–376.13] C.C. Chen, A.B. Herhold, C.S. Johnson, A.P. Alivisatos, Science 276 (1997)

398–400.14] J.N. Wickham, A.B. Herhold, A.P. Alivisatos, Phys. Rev. Lett. 84 (2000) 923–926.15] K. Jacobs, J. Wickham, A.P. Alivisatos, J. Phys. Chem. B 106 (2002) 3759–3762.16] Z.P. Tang, Y.M. Gupta, J. Appl. Phys. 81 (1997) 7203–7212.17] S.M. Sharma, Y.M. Gupta, Phys. Rev. B 58 (1998) 5964–5971.18] M.D. Knudson, Y.M. Gupta, A.B. Kunz, Phys. Rev. B 59 (1999) 11704–11715.19] C. Dameron, R. Reese, R. Mehra, A. Kortan, P. Carroll, M. Steigerwald, L. Brus, D.

Winge, Nature 338 (1989) 546–547.20] S.K. Sikka, S.M. Sharma, PINSA 68 (2002) 293–313.

21] A.P. Hammersley, ESRF Internal Report, ESRF97HA02T, FIT2D: an introduction

and overview, 1997.22] B.H. Toby, EXPGUI, a graphical user interface for GSAS, J. Appl. Cryst. 34 (2001)

210–213.23] M.P. Halsall, P. Harmer, P.J. Parbrook, S.J. Henley, Phys. Rev. B 69 (2004) 235207.