Structural, optical and thermal characterization of nanostructured CdSe obtained by mechanical alloying

Journal of Molecular Structure 1074 (2014) 511–515

Contents lists available at ScienceDirect

Journal of Molecular Structure journal homepage: www.elsevier.com/locate/molstruc

Structural, optical and thermal characterization of nanostructured CdSe obtained by mechanical alloying Gleison A. da Silva a, Daniela M. Trichês a, Edgar A. Sanches a,⇑, Kleber D. Machado b, Claudio M. Poffo a, João C. de Lima c, Sérgio M. de Souza a a b c

Departamento de Física, Universidade Federal do Amazonas, 3000 Japiim, 69077-000 Manaus, AM, Brazil Departamento de Física, Centro Politécnico, Universidade Federal do Paraná, 81531-990 Curitiba, Paraná, Brazil Departamento de Física, Universidade Federal de Santa Catarina, Campus Trindade, 88040-900 Florianópolis, SC, Brazil

h i g h l i g h t s  Mechanical alloying of Cd and Se promoted the nucleation of CdSe with few hours of milling.  A CdSe polymorphism was quantified by XRD using the Rietveld Method.  20 h of milling promoted the nucleation of a metastable Se phase.

a r t i c l e

i n f o

Article history: Received 7 April 2014 Received in revised form 4 June 2014 Accepted 6 June 2014 Available online 16 June 2014 Keywords: X-ray diffraction Mechanical alloying Zinc blend Wurtzite

a b s t r a c t A CdSe nanostructured alloy was produced by high energy Mechanical Alloying (MA). The existence of three crystalline phases, two CdSe polymorphs and a small fraction of CdO phase, was revealed through X-ray powder diffraction. With continued MA process, the CdO phase was dispersed between the nanostructures of zinc blende (ZB) and wurtzite (WZ). Differential Scanning Calorimetry (DSC) measurements indicated sublimation at 122 °C confirmed by thermal treatment. Raman spectroscopy revealed that the sublimated material was composed by a trigonal Se. After the thermal treatment, a partial phase transition from ZB to WZ was observed, which is in agreement with that observed by DSC, as well as by nucleation of the CdSeO3. Ó 2014 Elsevier B.V. All rights reserved.

Introduction Nanostructured CdSe semiconductors have been extensively studied in both applied and basic researches [1–5]. For applied researches they have been used as opto-electronic devices [6,7], solar cells [1,2,8] and photochemical biosensors [9], whereas the basic researches concentrates mainly on the structural characterization of these materials [4,5,10–12]. CdSe can be synthesized in two different structures, depending on the crystallite size. While in the bulk form exist some preference to form a hexagonal structure, S.G. P63MC (186), better known as wurtzite (WZ), in the nanometer form the trends is to form a face-centered cubic structure, S.G. F4-3 m (216), known as zinc blende (ZB), at any preparation temperature [13,14]. A wide variety of unique CdSe nanostructures have been synthesized using CdSe, such as nanotubes and nanowires [15] ⇑ Corresponding author. Tel.: +55 9284167887. E-mail address: [email protected] (E.A. Sanches). http://dx.doi.org/10.1016/j.molstruc.2014.06.023 0022-2860/Ó 2014 Elsevier B.V. All rights reserved.

nanosheets [16], nanorods and nanofractals [17], nanobelts [18] and quantum dots [19,20], utilizing a variety of techniques such as thermal [21], chemical [22], thermochemical [23,24] and mechanical [25,26]. It is already known that mechanical alloying (MA) can produce simultaneously at least two CdSe structures. A quantitative structural characterization is sometimes difficult due to the structural similarities between WZ and ZB [10] and their stacking faults disorders. In this work we present a structural study of nanostructured CdSe produced by MA using X-ray Diffraction (XRD), Differential Scanning Calorimetry (DSC) followed by annealing and microRaman spectroscopy. Experimental methods CdSe nanostructured semiconductor was produced by MA using a Spex Model 8000 Mixer/Mill (USA). Cd (Alfa Aesar, 99.9%) and Se (Alfa Aesar, 99.9%) were weighed and placed together inside the steel mill. The ratio of the mass of the balls to the powder was

512

G.A. da Silva et al. / Journal of Molecular Structure 1074 (2014) 511–515

previously chosen (5:1). The container was sealed under argon atmosphere and the milling was carried out for 2 and 20 h. After each milling time, sample was collected and characterized by X-ray Diffraction (XRD) using a Philips diffractometer, model X’Pert (Netherlands) equipped with Cu radiation (0.154056 nm). Data acquisition of XRD was 1 s per point. Rietveld method [28] was performed using the GSAS [27] package to determine the structural parameters related to the XRD patterns following the recommended IUCr guidelines [29]. Thermal behavior was verified through DSC between 25 and 600 °C with a heating rate of 10 °C/ min on a TA equipment, model DSC 2010. Raman spectroscopy was carried out to investigate vibrational behavior of the atoms in crystalline lattice. MicroRaman spectra were obtained with a spectrometer Jobin–Yvon T64000 triple Raman spectrometer equipped with a liquid-nitrogen-cooled charge coupled device multichannel detector. An excitation line of Ar laser (k = 514.5 nm) was focused down to 5 lm, 100 mW and acquisition time of 5  100 s. Spectrometer calibration was performed using a silicon wafer and setting the peak at 521.6 cm1. Raman frequencies were determined from the peak fit using a Lorentzian profile. Results and discussion X-ray diffraction and DSC Fig. 1 shows the XRD pattern of Cd and Se mixture at the milling times of 2 and 20 h. In the XRD pattern corresponding to 2 h of milling we could verify the nucleation of both structures, WZ and ZB, together with a cadmium oxide of cubic structure, ICSD n°.29290 [30] (indicated by stars in the figure), without any trace of the starting materials, as observed in reference [25]. The filled triangles correspond to the most intense peaks of ZB structure, ICSD n° 41528, while the inverted triangles correspond to WZ structure, ICSD n° 415786. From 2 to 20 h of milling, diffractograms presented only one significant difference related to the disappearance of CdO phase, which can indicate, among many others possibilities, the phase amorphization, dispersion or solubilization on the crystalline matrix [31]. The similarity between both XRD patterns indicates that, if the CdO phase formation is avoided, a milling time of two hours is enough for the process stabilization. It was noted that from 2h  60° no peaks of WZ phase the peaks were observed, but only this phase can explain the first peak at 2h  24° and the shoulder of the most intense peak observed at 2h  27°. The difficulty increases since the major peaks of ZB phase are overlapped with the main peaks of the WZ phase. We also see

Fig. 1. Structural evolution of the CdSe alloy during the milling process.

in Fig. 1 that the WZ structure should have peaks at 2h  35° and 2h  46°, corresponding to the planes (1 0 1) and (1 0 3), respectively. The reason why these peaks were not observed can be explained through the structure defects, or stacking faults, which can produce different features in diffraction patterns, such as shift peak, anisotropic enlargement or even their extinction [32,33]. In fact, both structures are based on the stacking faults of twodimensional planar units, coordinated tetrahedrally with four first neighbors. From this point of view, the main difference is that one is translated relative to each other, generating layers ABABAB along [1 1 1] for ZB structure and ABCABC along [0 0 1] for WZ structure [10]. Given these characteristics, it was necessary to perform a structural simulation using the Rietveld refinement [28]. Fig. 2 shows the XRD pattern of CdSe milled for 2 h, which was overlapped by their respective Rietveld refinement, together with the deconvolution of its three phases. The refinement results are summarized in Table 1. From this deconvolution it was verified that the relationship between the peaks intensities of the ZB phase does not change significantly from that published in the ICSD and JCPD databases [34]. This indicates that only the WZ phase has stacking faults, highlighted by the difficulty of adjusting the peaks at 2h  35° and 2h  46°. Crystallite size shown in Table 1 was calculated by Scherrer formula [35] using the more well-defined peaks average for each phase. Since this stacking fault causes anisotropic features in XRD patterns, it was applied the Stephens anisotropic model [36] to reach the presented refinements. This model proposes to simulate microdeformations dependent on the (h k l) planes. This model is particularly applicable for different enlargements for each family of planes [31]. The variance of the peak widths are modeled by Eq. (1) [36]: 2

½SðhklÞ ¼

X

H K L

SHKL h k l

ð1Þ

H;K;L

where h, k and l are Miller indices and SHKL are coefficients defined by crystal symmetry and can be refined in GSAS program. The function S (h k l) for hexagonal symmetry, which is the case of WZ, is given by the following equation:

SðhklÞ ¼

pd2hkl 18; 000 4

4

4

2 2

3

3

½S400 ðh þ k þ 3h k þ 2h k þ 2hk Þ 2 2

2 2

2

þS004 l þ 3S202 ðh l þ k l þ hkl Þ1=2

ð2Þ

where dhkl is the distance between the planes and H + K + L = 4 [37]. The three SHKL parameters obtained through Rietveld refinement of the sample milled for 20 h were S400 = 300 ± 15, S004 = 0.02 ± 0.08

Fig. 2. Experimental XRD pattern of the CdSe sample milled for 2 h overlapped by the best fitting achieved by the Rietveld method using the GSAS package and its discriminated phases.

G.A. da Silva et al. / Journal of Molecular Structure 1074 (2014) 511–515

513

Table 1 Parameters from ICSD and refined parameters for the sample milled for 20 h and its respective annealing. Phase

wRp

2h 0.0546

20 h 0.0544

Annealed 0.0966

%wt hdi (Å) a (Å) c (Å)

36.5 35 4.311(1) 7.083(4)

33.2 33 4.304(1) 7.084(4)

69.2 1081 4.3018(5) 7.0114(2)

%wt hdi(Å) a (Å)

61.8 8 6.103(1)

%wt hdi(Å) a (Å)

1.7 1442 4.730(2)

66.8 8 6.105(1) – – – –

8.2 1235 6.0805(2) – – – –

%wt hdi(Å) a (Å) b (Å) c (Å) b

– – – –

– – – –

22.6 1081 5.7025(4) 12.8428(9) 8.5940(6) 101.033(5)

WZ

P 63 MC (186) ZB F4–3 m (216) CdO FM-3 M (225) Cd(SeO3) P 1 21/c 1 (14)

and S202 = 462 ± 23 resulting in micro deformations of S (1 0 2) = 8.7% and S (1 0 3) = 7.8% for the planes (1 0 2) and (1 0 3), respectively. It is interesting to note that the low value of S004 associated with the family of planes along the direction [0 0 1] leads to S (0 0 2) = 0.1%, an order of magnitude lower than obtained for the planes (1 0 2) and (1 0 3). Considering this refinement as a structural simulation, this result indicates that the stacking of the planes perpendicular to the c axis produce an increase of structural order, as highlighted through the narrow peak (0 0 2) at 2h  25°. If SHKL is interpreted as a component of microstructural strain energy [38], this result represents the stacking situation as a low energy equilibrium state. In order to investigate the thermal stability of the sample a DSC measurement was carried out using a rate of 10 °C/min as shown in Fig. 3. An endothermic peak was observed between 90 and 150 °C with a minimum at 122 °C followed by two exothermic peaks at 382 °C and 445 °C. Since the melting points of crystalline Se and Cd are around 217 °C [39] and 320.9 °C [40], respectively, we have no traces of the precursor materials. Thus, the sample was subjected to annealing at 450 °C in argon atmosphere for 11 h and

Fig. 3. DSC curve for the sample milled for 20 h. Temperature increase rate of 10 °C/ min.

Fig. 4. Optical visualization of the pure condensed Se during the thermal treatment.

allowed to cool slowly inside the oven off. It was observed that considerable amount of material was condensed on the quartz tube walls. Fig. 4 shows a droplet with a 20-fold increase. This condensate was analyzed by micro Raman spectroscopy and corresponds to pure Se. It will be discussed later. This phenomenon explains the endothermic peak at 122 °C as a metastable Se phase, not confined [41], which sublimated at this temperature. Comparing with XRD measurements it was observed that this phase does not produce coherent X-ray scattering, neither can be considered amorphous, since Cardoso et al. [41] have produced amorphous selenium by MA and they did not observe any sublimation effect at this temperature through DSC technique. Considering these two exothermic peaks, Kotkata et al. [42] have produced chemically the CdSe nanoparticles and attributed the observed exothermic processes at approximately 325–450 °C to the morphological transition from ZB to WZ. These temperatures are in agreement with those observed in this work. However, when comparing the XRD patterns of the samples milled for 20 h and annealed sample, as shown in Fig. 5, we see that annealing promoted the disappearance of CdO phase, giving rise to a new structure, identified as Cd(SeO3), ICSD n° 75274, of monoclinic structure, as well as a considerable increase in the WZ structure at the expense of the ZB structure. The observed narrow peaks indicate considerable improvement in crystallinity, highlighted mainly by the disappearance of the stacking of the (1 0 2) and (1 0 3) planes of the WZ structure. In this case, due to the structural similarity, by visual inspection, it is not easy to determine the remnants of the ZB structure. However it was only possible to perform a good fitting through the Rietveld refinement when the WZ phase was introduced.

Fig. 5. Experimental XRD pattern of the CdSe sample milled for 20 h.

514

G.A. da Silva et al. / Journal of Molecular Structure 1074 (2014) 511–515

Fig. 6 shows the XRD pattern of the treated sample overlapped with its Rietveld refinement together with the deconvolution of its phases. The introduction of the ZB phase in this refinement was necessary to satisfy the ratio of intensities between the first two peaks of WZ phase, which corresponds to the planes (1 0 0) and (0 0 2), respectively. This ratio of intensities is around 60% and could not be solved with preferred orientation parameters. This result suggests that the ZB ? WZ transition does not occur abruptly, as proposed by DSC exothermic peak observed in Fig. 3. Actually, it occurs gradually, since the baseline is continuously exothermic until 600 °C. Thus, the annealing performed at 450 °C is not enough for the complete the ZB ? WZ transition.

Table 2 Raman results of the 20 h, annealed and droplet samples. E2 TO 20 h

LO

LO + LA ZB

TO + LO 2LO ZB

ZB

3LO

x

179 207 234

258 376

412 458 505 622

(cm1) Dx (cm1)

49

22

23

33

173 207 238

377

414

618

29

20

20

15

Annealed x (cm1) Dx (cm1) x Droplet (cm1) Dx (cm1)

13

10

32

45

9

143

237

453

9

7

33

48

113

Micro-Raman spectroscopy Fig. 7 shows three Raman spectra which correspond to the sublimated material (droplet) followed by milled sample for 20 h and its respective annealing at 450 °C. In this figure we also show the droplet spectrum, fitted with Lorentzian profile, at 143, 237 and 453 cm1. Se phase can solidify to amorphous structure in at least three different crystalline allotropes [43], one in trigonal symmetry and two in monoclinic symmetry. The different forms of Se involve

different molecular vibrational modes. The peaks, furthermore, are attributed E, A1 and second harmonic of Se trigonal (t-Se) [44]. All spectra were measured under the same experimental conditions as described in section ‘Experimental methods’. All spectra were fitted by Lorentzian functions and the resulting position and width peak are shown in Table 2. The similarity between the vibrational spectra of the sample with predominantly ZB phase and another with WZ phase reveals structural similarities at the molecular level. Furthermore, we could observe, in the sample milled for 20 h additional three peaks located at 258, 458 and 505 cm1. The one at 258 cm1, however, is much more pronounced than the two others. When compared to the Raman peaks of cubic CdO we observed no peaks in this region and it was disregarded. This exclusion, however, leads us to conclude that these modes might belong to the ZB structure, corroborating with the XRD results. Conclusions

Fig. 6. Experimental XRD pattern of the annealed sample overlapped by the best fitting achieved by the Rietveld method using the GSAS package.

Mechanical alloying of Cd and Se promoted the nucleation of CdSe with few hours of milling, with a predominance of ZB phase in relation with the WZ phase. By DSC and thermal treatment reveals the existence of pure selenium without chemical order in short, medium or long range. The existence of this phase plays an important role in the chemical environment of the sample, since it is not obtained by grinding of pure Se [45]. It was noted that even at 600 °C the transformation ZB ? WZ was not fully complete. Acknowledgements This research was financially supported by the Brazilian agency CNPq. We are also indebted to Dr. P. Chaudhuri for careful reading of the manuscript. References

Fig. 7. Raman Spectroscopy for the alloy milled for 20 h (red), thermally treated (blue), and for the condensate (black). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

[1] A.P. Kokate, U.B. Suryavanshi, C.H. Bhosale, Sol. Energy 80 (2006) 156. [2] D.L. Klein, R. Roth, A.K.L. Lim, A.P. Alivisatos, P. McEuen, Nature 386 (1997) 699. [3] P.J. Wu, Y.P. Stetsko, Appl. Phys. Lett. 90 (2007) 161911. [4] A.S. Masadeh, E.S. Bozˇin, C.L. Farrow, G. Paglia, P. Juhas, S.J.L. Billinge, M.G. Kanatzidis, Phys. Rev. B 76 (2007) 115413. [5] G. Dalba, P. Fornasini, R. Grisenti, D. Pasqualini, D. Diop, F. Monti, Phys. Rev. B 58 (1998) 4793. [6] G.L. Tan, X.F. Yu, J. Phys. Chem. C 113 (2009) 8724. [7] V.L. Colvin, M.C. Schlamp, A.P. Alivisatos, Nature 370 (1994) 354. [8] L. Han, D. Qin, X. Jiang, Y. Liu, L. Wang, J. Chen, Y. Cao, Nanotechnology 17 (2006) 4736. [9] K. Yan, R. Wang, J. Zhang, Biosens. Bioelectron. 53 (2014) 301. [10] C.Y. Yeh, Z.W. Lu, S. Froyen, A. Zunger, Phys. Rev. B 46 (1992) 10086. [11] A.C. Carter, C.E. Bouldin, K.M. Kemner, M.I. Bell, J.C. Woicik, S.A. Majetich, Phys. Rev. B 55 (1997) 13822.

G.A. da Silva et al. / Journal of Molecular Structure 1074 (2014) 511–515 [12] F. Widulle, S. Kramp, N.M. Pyka, A. Göbel, T. Ruf, A. Debernardi, M. Cardona, Phys. B: Condens. Matter 263 (1999) 448. [13] C. Murray, D.J. Norris, M.G. Bawendi, J. Am. Chem. Soc. 115 (1993) 8706. [14] R.J. Bandaranayake, G.W. Wen, J.Y. Lin, H.X. Jiang, C.M. Sorensen, Appl. Phys. Lett. 67 (1995) 831. [15] X. Jiang, B. Mayers, T. Herricks, Y. Xia, Adv. Mater. 15 (2003) 1740. [16] J.S. Son, X.D. Wen, J. Joo, J. Chae, S.I. Baek, K. Park, T. Hyeon, Angew. Chem. 121 (2009) 6993. [17] Q. Peng, Y. Dong, Z. Deng, Y. Li, Inorg. Chem. 41 (2002) 5249. [18] R. Venugopal, P.I. Lin, C.C. Liu, T.Y. Chen, J. Am. Chem. Soc. 127 (2005) 11262. [19] T.A. El-Brolossy, S. Abdallah, T. Abdallah, H. Awad, M.B. Mohamed, S. Negm, H. Talaat, Eur. Phys. J. Special Topics 153 (2008) 369. [20] T.A. El-Brolossy, S. Abdallah, T. Abdallah, M.B. Mohamed, S. Negm, H. Talaat, Eur. Phys. J. Special Topics 153 (2008) 365. [21] S.A. Al’fer, V.F. Skums, Inorg. Mater. 37 (2001) 1237. [22] S.H. Tolbert, A.B. Herhold, C.S. Johnson, A.P. Alivisatos, Phys. Rev. Lett. 73 (1994) 3266. [23] J.W. Williams, C.N. Adams, N.A. Kotov, P.E. Savage, Ind. Eng. Chem. Res. 46 (2007) 4358. [24] R. Seoudi, M.M. Elokr, A.A. Shabaka, A. Sobhi, Phys. B: Condens. Matter 403 (2008) 152. [25] G.L. Tan, J.H. Du, Q.J. Zhang, J. Alloys Comp. 468 (2009) 421. [26] A.C. Larson, R.B. Von Dreele, GSAS Manual, Rep. LAUR 86–748, Los Alamos Natl. Lab., Los Alamos, 1988. [27] H. Rietveld, J. Appl. Crystallogr. 2 (2) (1969) 65. [28] L.B. McCusker, R.B. Von Dreele, D.E. Cox, D. Louer, P. Scardi, J. Appl. Crystallogr. 32 (1999) 36. [29] ICSD, Inorganic Crystal Structure Database Gmchin-Institute fur AnorganisheChemie and Fachinformationszentrum FIZ, Karlsruhe, Germany, 1995.

515

[30] Q.H.F. Rebelo, E.A. Cotta, S.M. de Souza, D.M. Trichês, K.D. Machado, J.C. de Lima, L. Manzato, J. Alloys Comp. 575 (2013) 80. [31] J.D. Makinson, J.S. Lee, S.H. Magner, R.J. De Angelis, W.N. Weins, A.S. Hieronymus, Adv. X-Ray Anal. 42 (2000) 407. [32] B. Palosz, W. Palosz, S. Gierlotka, Acta Crystallogr. C 41 (1985) 807. [33] File, Powder Diffraction, JCPDS-ICDD, 12 Campus Boulevard, Newtown Square, PA, 247 2001 19073. [34] S. Vives, E. Gaffet, C. Meunier, Mater. Sci. Eng. A 366 (2004) 229–238. [35] P.W. Stephens, J. Appl. Crystallogr. 32 (1999) 281. [36] J. Frantti, S. Eriksson, S. Hull, S. Ivanov, V. Lantto, J. Lappalainen, M. Kakihana, J. Eur. Ceram. Soc. 24 (2004) 1141. [37] M.E. Manley, B. Fultz, D.W. Brown, B. Clausen, A.C. Lawson, J.C. Cooley, D.J. Thoma, Phys. Rev. B 66 (2002) 024117. [38] F.B. Mainier, L.P. Monteiro, L.H. Fernandes, M.A. Oliveira, J. Sci. Technol. 3 (2011) 176. [39] Z.Y. Jiang, Z.X. Xie, X.H. Zhang, R.B. Huang, L.S. Zheng, Chem. Phys. Lett. 378 (2003) 313. [40] H. Yamada, Y. Kang, T. Aso, H. Uesugi, T. Fujimura, K. Yonebayashi, Soil Sci. Plant Nat. 44 (1998) 385. [41] J.C. de Lima, T.A. Grandi, R.S. De Biasi, J. Non-Cryst. Solids 286 (2001) 93. [42] M.F. Kotkata, A.E. Masoud, M.B. Mohamed, E.A. Mahmoud, Phys. E: LowDimens. Syst. Nanostruct. 41 (2009) 640. [43] P.J. Carroll, J.S. Lannin, J. Phys. Colloid 42 (1981) C6–643. [44] Z.V. Popovic´, G. Stanišic´, D. Stojanovic´, R. Kostic´, Phys. Status Solidi B 165 (1991) K109–k112. [45] G.J. Fan, F.Q. Guo, Z.Q. Hu, M.X. Quan, K. Lu, Phys. Rev. B 55 (1997) 11010.