Shape and stoichiometry control of bismuth selenide nanocrystals in colloidal synthesis

Yan Jianga, Mengmeng Haoa, Liangxing Jianga, Fangyang Liu*ab and Yexiang Liua
aSchool of Metallurgy and Environment, Central South University, Changsha 410083, China. E-mail: liufangyang@csu.edu.cn
bSchool of Photovoltaic and Renewable Energy Engineering, University of New South Wales, Sydney 2052, Australia

Received 23rd March 2016 , Accepted 25th April 2016

First published on 26th April 2016


Abstract

Three kinds of Bi–Se nanocrystals (Bi2Se3, Bi8Se9 and Bi1.007Se0.993) have been synthesized via a hot-injection approach. The colloid morphology, from hexagonal to octahedron, strongly depends on the injection temperature. The growth mechanism of Bi–Se nanocrystals is revealed here. A higher temperature could cause the reduction reaction from Bi3+ to Bi, and accelerate the reaction speed of Bi–Se, thus causing a difference of monomer concentration in bulk solution, which impacts the shapes of bismuth selenide nanocrystals as well as the composition. The three Bi–Se nanocrystals also show different optical and photoelectrocatalyst performance.


Nanocrystal engineering is of great interest over the past few years in catalysis,1 electrocatalysis,2,3 and photocatalysis,4 due to the size- and shape-controlled properties of nanomaterials. Among various methods to obtain nanocrystals, colloidal synthesis is highlighted due to its flexible processing chemistry5 and potential for large-scale production. Through controlling the reaction environment, like temperature, time, capping molecules, precursor ratio and monomer concentration, engineering shapes as well as constituents of nanocrystals is feasible. Guo et al. reported wurtzite ZnSe nanorod couples connected by twinning structures by purification and re-adding monomer approach.6 Han et al. found that the colloidal morphology of Bi2S3 nanocrystals, such as nanodots and nanorods, strongly depends on the preparation temperature and the ratio of Bi/S precursors, which lead to different optical and electrical properties.7 Sang-Min Lee et al. pointed out that the shape of CdS nanocrystals can be adjusted by the temperature-mediated phase control of the initial seeds.8 It is widely accepted that colloidal synthesis is an effective way to explore unique nanostructures of relevance for applications.

Bismuth selenide, a V–VI semiconductor, is mostly known as topological insulator material with a simple band structure and a single Dirac cone on the surface. However, with a narrow band gap (∼0.3 eV in bulk) tunable by shape, bismuth selenide is able to utilize near infrared solar energy, which is admirable for applying in photoelectrocatalyst.9 Moreover, Bi2Se3 is also popular in thermoelectric devices.10 Consequently, the investigation on shape-controlled synthesis of Bi2Se3 is of great interests. Liu et al. reported growth of Bi2Se3 nanobelts synthesized through a co-reduction method under ultrasonic irradiation,11 Yan et al. had synthesized various Bi2Se3 nanocrystals with various morphologies using CVD method,12 Kadel et al. expressed nano-flake like Bi2Se3 synthesized by hydrothermal method.13 Though there are some reports on colloidal synthesis of Bi2Se3,14,15 to date, to the best of author's knowledge, there has been no reports on the synthesis of Bi–Se nanocrystals with tunable shape and composition by a simple temperature-controlled colloidal way.

Herein, Bi2Se3, Bi8Se9 and Bi1.007Se0.993 with hexagonal, disk and octahedron-like shapes respectively were synthesized by simply controlling the reaction temperature. The growth mechanism was discussed in detail. It is found that in higher temperature, formation of shapes with lower chemical potential were promoted. Moreover, optical and photoelectrochemical properties of three Bi–Se nanocrystals had been investigated.

It is known that Se powder is not soluble in oleyl amine until heated to 205 °C,16 but our previous work shows that in assistance of dimethylamine borane (DMAB), Se powder could firstly be reduced to alkylammonium selenide and then dissolved in oleyl amine at 110 °C.17 In this work, we prepared Se–DMAB–OLA solution as Se precursor and Bi(NO3)3–OLA as Bi precursor. By injecting Se–DMAB–OLA into Bi precursor at 220 °C, 250 °C and 290 °C, three kinds of Bi–Se nanocrystals were obtained (more experiment details in ESI). In a typical method, 1.5 mmol (0.089 g) DMAB, 1.5 mmol (0.118 g) Se powder and 3 mL oleyl amine were loaded into a two-neck flask (10 mL) connected to a Schlenk line. After purging for several times, the temperature was raised up to 60 °C under vacuum for degassing about 30 min. Then the flask was heated to 100 °C under Ar for 10 min, the turbid solution gradually turned clear and yellow, as the Se powder reduced by DMAB. The Se–DMAB solution was kept at 60 °C for injection process. Meanwhile, 1 mmol (0.485 g) Bi(NO3)3·5H2O with 10 mL oleyl amine were loaded into a three-neck flask (100 mL) connected to a Schlenk line. After purging and degassing process, the flask was heated up to determined temperature (220 °C, 250 °C and 290 °C) under Ar for 30 min to form a homogenous solution. The Se–DMAB solution made before was then injected to Bi-precursors, kept for 1 h for crystal growth. The final precipitates were washed with ethanol for several times, and dispersed in toluene.

TEM images show that at lower reaction temperature, the synthesized nanocrystal show a hexagonal shape (Fig. 1a and S1). Corresponding XRD pattern (Fig. 2a) evidences that all the diffraction peaks can be related to rhombohedral Bi2Se3 phase (paraguanajuatite, JCPDS no. 33-0214). The characteristic peak represents the prominent growth orientation of Bi2Se3 nanoplates along the [015] direction.18 The selected area electron diffraction (SAED) ring patterns (Fig. 1b) could be related to the (107), (0015) and (1210) planes of Bi2Se3, and the interplanar spacing measured from high-resolution TEM (Fig. 1c) of 2.074 Å can correspond to the (110) plane. The sharp characteristic peak, distinct rings in the SAED pattern as well as clear lattice fringes reveal the high crystalline quality of Bi2Se3 nanoplates. The elemental composition of synthesized Bi2Se3 was determined by X-ray fluorescence (XRF) spectroscopy (Table 1 in ESI), the atom ratio of Bi to Se is 4.1[thin space (1/6-em)]:[thin space (1/6-em)]6, almost the same as stoichiometric ratio of Bi2Se3 phase.


image file: c6ra07570c-f1.tif
Fig. 1 (a)TEM image, (b) SEAD ring patterns and (c) HRTEM of Bi2Se3 nanocrystals; (d) TEM image, (e) SEAD ring patterns and (f) HRTEM of Bi8Se9 nanocrystals; (g) TEM image, (h) SEAD ring patterns and (i) HRTEM of Bi1.007Se0.993 nanocrystals.

image file: c6ra07570c-f2.tif
Fig. 2 (a) XRD patterns of Bi2Se3, Bi8Se9 and Bi1.007Se0.993; (b) XPS spectra of Bi 4f for Bi2Se3, Bi8Se9 and Bi1.007Se0.993.

With increasing temperature, nanodisks were found at 250 °C (Fig. 1d and S2). The XRD patterns (Fig. 2a) could correspond to guanajuatite Bi8Se9 phase (JCPDS no. 70-2610). SAED ring patterns (Fig. 1d) reveal (1022), (0048) and (0225) planes of Bi8Se9, and the HR-TEM image (Fig. 1f) shows the d-spacing of lattice fringes of 2.738 Å, which is related to the (0123) plane of Bi8Se9. In even higher temperature, octahedron nanocrystals were formed (Fig. 1g and S3). For octahedron Bi–Se, the XRD (Fig. 2a) patterns can be accordingly indexed to hexagonal Bi1.007Se0.993 phase (nevskite, JCPDS no. 82-2442). SAED patterns (Fig. 1h) show three rings corresponding to the (105), (018) and (0110) planes of Bi1.007Se0.993, and the d-spacing of 3.23 Å (Fig. 1i) can be related to (013) plane. XRF results (Table 1 in ESI) show the atom ratio of Bi to Se of 5.1[thin space (1/6-em)]:[thin space (1/6-em)]6 and 1.01[thin space (1/6-em)]:[thin space (1/6-em)]1 for Bi8Se9 and, Bi1.007Se0.993 respectively, which also correspond to the stoichiometric ratios. It is well known that the Gibb's free energy of the surface of nanoparticles increases with the surface–volume ratio, as a result, Bi2Se3 with hexagonal shape obtains the highest surface energy, following with Bi8Se9 with disk-like shape and ended with Bi1.007Se0.993 with octahedron shape. As the reaction temperature increases, the shape of Bi–Se nanocrystal turns more stable, from hexagonal to octahedron, and the atom ratio of Bi to Se gradually increases too along with the temperature. It can be concluded that higher temperature could promote the formation of stabler and Bi-richer nanocrystals. Moreover, XRD patterns ascribed to other impurity crystal phases were not detected. Average crystal size are shown in Fig. S4.

We also deduced X-ray photoelectron spectroscopy (XPS) to probe the surface valence state of three kinds of Bi–Se nanocrystals. Fig. 2b shows the high-resolution spectra of Bi 4f of Bi2Se3, Bi8Se9 and Bi1.007Se0.993. In the XPS spectrum of Bi 4f in Bi2Se3, the two symmetrical peaks appear at the binding energies of 158.2 eV (Bi 4f7/2) and 163.6 eV (Bi 4f5/2) with a splitting of 5.3 eV is in consistence with the literature.19 However, in case of Bi8Se9, Bi 4f spin–orbit (SO) doublet become broader compared to that of Bi2Se3 with an asymmetrical characteristic. Since the metallic Bi–Bi bond are located at 157.0 eV (Bi 4f5/2) and 162.3 eV (Bi 4f7/2),19 we consider that each asymmetrical peak of Bi 4f in Bi8Se9 is a convolution of two peaks, as further shown in Fig. S5, due to different bonding environment of Bi atoms (Bi–Se bond and metallic Bi–Bi bond). The phenomenon is also observed in Bi-rich surface of Bi2Se3 by Hewitt et al.20 As for Bi1.007Se0.993, the peaks of Bi 4f SO doublet become even broader. Interestingly, there are less Bi–Bi bonds and more Bi–Se bonds in Bi1.007Se0.993 compared with those of Bi8Se9 (Fig. S6). This is because theoretically, searching from Findit software we can find that Bi8Se9 is composed of Bi and Se atoms without any oxidation and reduction of the two elements (ICSD-42665[thin space (1/6-em)]21) and Bi1.007Se0.993 of Bi, Bi2.979+ and Se2− ions (ICSD-79019[thin space (1/6-em)]22). So theoretically, there should be more Bi–Bi bonds in Bi8Se9 than those in Bi1.007Se0.993, which is in consistence with our result, though there are also a little Bi–Se bonds in Bi8Se9 in our experiment. It can be concluded that in three Bi–Se nanocrystals, Bi atoms have different bonding environments, as illustrated in Fig. S7 using CrystalMaker software.

XRD, TEM and XPS results show that in oleyl amine system, by controlling the reaction temperature, shape and composition control of Bi–Se nanocrystal is available. There have been several theories investigating how temperature infects the shapes of nanocrystals. Lee reported that the shapes might be adjusted by temperature-mediated phase control of initial seeds during the nucleation processes. Under different temperature condition, the crystalline phase of nanocrystal seeds might be different.23 However in Peng's theory, they considered that temperature-dependence of shapes might be another form of concentration dependence.24 Herein, before the nucleation process and formation stage, we concern more about the Bi-precursors heated to different temperature, since changes of color in high-temperature Bi-precursors were observed. As shown in Fig. S8, when kept at 220 °C for half an hour, the color of Bi-precursor stays white, however when increasing the temperature to 250 °C or even higher (290 °C), the color of Bi-precursor suddenly turns black. The three Bi-precursors, kept under 220 °C, 250 °C and 290 °C, were centrifuged and washed several times before deposited on soda-lime glasses for XRD tests. XRD patterns of three Bi-precursors (Fig. 3) show that under 220 °C, no sharp peak is observed, implicating the absence of any well-formed crystals. As temperature increases, sharp peaks at around 26° appear, which are in consistence with bismuth (JCPDS no. 44-1246). This is resulted to the reducibility of oleyl amine under higher temperature which could reduce Bi3+ to Bi. It can be concluded that in preparation of Bi2Se3, only Bi3+ and Se2− took part in the reaction, resulted in paraguanajuatite Bi2Se3. For the formation of Bi8Se9 and Bi1.007Se0.993, besides Bi3+, Bi also took part in the reaction and resulted in Bi-rich products. Chemical reaction equations for the formation of three kinds of Bi–Se composite are proposed here:

2Bi3+ + 3Se2− → Bi2Se3

2Bi + 6Bi3+ + 9Se2− → Bi8Se9

0.345Bi + 0.662Bi3+ + 0.993Se2− → Bi1.007Se0.993


image file: c6ra07570c-f3.tif
Fig. 3 XRD patterns of Bi-precursors heated under 220 °C, 250 °C and 290 °C for formation of Bi2Se3, Bi8Se9 and Bi1.007Se0.993 respectively.

Moreover, in higher temperature, as Peng mentioned in their theory,22 the consumption of Se2−, Bi and Bi3+ monomers in bulk solution for the nucleation process as well as the number of nuclei formed in nuclei stage should both be higher than those under low temperature. Consequently, the monomers concentration is lower in high temperature solution, which means the relative chemical potential of the remaining monomers in the growth stage is also lower. Thermodynamically, to achieve the balance, octahedron nanocrystals are more stable in bulk solution of lower monomer concentration than disk-like ones, let alone hexagonal ones, since the Gibb's free energy of the surface of nanoparticles increases with the surface–volume ratio as mentioned above. In this way, as reaction temperature grows, the shape of Bi–Se nanocrystal turns from hexagonal to octahedron like. Reaction process is illustrated in Scheme 1.


image file: c6ra07570c-s1.tif
Scheme 1 Reaction process of Bi2Se3, Bi8Se9 and Bi1.007Se0.993.

Shapes impact the properties of Bi–Se nanocrystals. We have investigated the optical properties of three Bi–Se samples by dissolving them in toluene. Fig. S9 shows the room temperature optical absorption spectra of Bi2Se3, Bi8Se9 and Bi1.007Se0.993. Bi8Se9 shows the highest absorbance. All three Bi–Se nanocrystals show near-infrared absorbance, which is desirable for photocatalysis and photoelectrocatalysis. The optical absorption coefficient is calculated by using the equation:23

α = −1/c[thin space (1/6-em)]ln[thin space (1/6-em)]A
where c is the concentration of Bi–Se sample dispersed per 1 mL of toluene (around 10−5 g mL−1), A is the absorbance. The value of α is found to be higher than 104 cm−1 for three nanocrystals, showing that Bi–Se nanocrystals are direct bandgap semiconductors.25 Bandgap values are calculated by Tauc equation:26
(αhν)n = A(Eg)
where A is a constant, presents the light energy, Eg is the optical band gap energy, α is the measured absorption coefficient, n is 0.5 for indirect band gap and 2 for direct band materials. Here we choose n = 2 since Bi–Se nanocrystals are all direct band semiconductors. The plot of (αhν)2 with respect to is shown in Fig. S10. Extending the linear part of the curve on the axis, the bandgap values of 0.85 eV, 1.2 eV and 1.25 eV is obtained for Bi2Se3, Bi8Se9 and Bi1.007Se0.993, respectively.

We also investigated the photo-response abilities of Bi2Se3, Bi8Se9 and Bi1.007Se0.993 nanocrystals. A conventional three-electrode PEC cell was constructed. Three Bi–Se samples dissolved in toluene were deposited on ITO glasses as working electrodes, a saturated calomel electrode (SCE) was adopted as counter electrode and a graphite electrode worked as counter electrode. 0.1 M Na2SO4 solution was applied as electrolyte, in which the illumination intensity was 40 W cm−2 from xenon lamp. The linear sweep voltammogram (LSV) of Bi2Se3 (Fig. S11a) under on/off chopping illumination shows a decrease of photocurrent density with negative potential sweeping, implicating an n-type semiconductor character of Bi2Se3. In the PEC cell, holes transferred from the valence band of Bi2Se3 to electrolyte. Bi2Se3 film shows photo response from 0.2 V to 1.3 V vs. SCE, with an photocurrent density of 6 μA cm−2 at a bias of 1.2 V. This is at the same level of other reported Bi2Se3 materials.27 For Bi8Se9 films (Fig. S11b), it is hard to distinguish p/n type from LSV curves, and the photo response range is also narrower compared with Bi2Se3. However under 0.3 V, the photocurrent density achieves 3 μA cm−2, which still have the potential of applying in photocatalysis and photoelectrocatalysis. For Bi1.007Se0.993, the photo response is very weak (Fig. S11c). It can be concluded that both Bi2Se3 and Bi8Se9 can be used in photocatalyst and photoelectrocatalyst.

Conclusion

In summary, we have successfully synthesized three kinds of Bi–Se nanocrystals, Bi2Se3, Bi8Se9 and Bi1.007Se0.993 with hexagonal-, disk- and octahedron-like shapes respectively in a facile hot-injection way, by changing the reaction temperature. XRD analysis confirmed the formation of three Bi–Se nanocrystals, and the simulated crystal structures were given. Comprehensive analysis from TEM, XRF and XPS confirms the purity of Bi–Se nanocrystals. We also deduced a growth mechanism for three kinds of nanocrystals. From XRD analysis of Bi-precursors kept at different temperature, it is found that beyond 250 °C, Bi3+ was reduced to Bi, so at higher temperature, beside Bi3+, Bi also took part in the reaction. Moreover, the temperature also infects reaction speed, thus causing differences in shapes of nanocrystals. With different shapes and composition, Bi2Se3, Bi8Se9 and Bi1.007Se0.993 show different optical absorption and photoelectrocatalysis performances. Bi2Se3 and Bi8Se9 nanocrystals show potential of applying in photoelectrocatalyst.

Acknowledgements

This work was supported by Hunan Provincial Natural Science Foundation of China (13JJ1003 and 2015JJ2175).

References

  1. H. Mistry, F. Behafarid, E. Zhou, L. K. Ono, L. Zhang and B. R. Cuenya, ACS Catal., 2013, 4, 109–115 CrossRef .
  2. M. H. Lee, T. H. Kim, Y. S. Kim and H. K. Song, J. Phys. Chem. C, 2011, 115, 12255–12259 CAS .
  3. A. Cao and A. Manthiram, Phys. Chem. Chem. Phys., 2012, 14, 6724–6728 RSC .
  4. C. Huang and Q. Xu, J. Mater. Chem. A, 2015, 3, 13884–13891 Search PubMed .
  5. Z. Adam Peng and X. Peng, J. Am. Chem. Soc., 2002, 124, 3343–3353 CrossRef .
  6. G. Jia, A. Sitt, G. B. Hitin, I. Hadar, Y. Bekenstein, Y. Amit and U. Banin, Nat. Mater., 2014, 13, 301–307 CrossRef CAS PubMed .
  7. P. Han, A. Mihi, J. Ferre-borrull, J. Pallares and L. F. Marsal, J. Phys. Chem. C, 2015, 119, 10693–10699 CAS .
  8. S. M. Lee, S. N. Cho and J. Cheon, Adv. Mater., 2003, 15, 441–444 CrossRef CAS .
  9. Q. Zhang, J. Su, X. Zhang, J. Li, A. Zhang and Y. Gao, J. Alloys Compd., 2012, 545, 105–110 CrossRef CAS .
  10. K. Kadel, L. Kumari, W. Z. Li, J. Y. Huang and P. P. Provencio, Nanoscale Res. Lett., 2011, 6, 57 Search PubMed .
  11. H. Liu, H. Cui, F. Han, X. Li, J. Wang and R. I. Boughton, Cryst. Growth Des., 2005, 5, 1711–1714 CAS .
  12. Y. Yan, Z. M. Liao, Y. B. Zhou, H. C. Wu, Y. Q. Bie, J. J. Chen and D. P. Yu, Sci. Rep., 2013, 3, 1264 Search PubMed .
  13. K. Kadel, Synthesis and Characterization of Thermoelectric Nanomaterials, FIU Electronic Theses and Dissertations, Florida, 2014, p. 1170.
  14. Y. Min, G. D. Moon, B. S. Kim, B. Lim, J. S. Kim, C. Y. Kang and U. Jeong, J. Am. Chem. Soc., 2012, 134, 2872–2875 CrossRef CAS PubMed .
  15. X. Liu, J. Xu, Z. Fang, L. Lin, Y. Qian, Y. Wang and J. Zeng, Nano Res., 2015, 8, 3612–3620 CrossRef CAS .
  16. K. H. Park, K. Jang, S. Kim, H. J. Kim and S. U. Son, J. Am. Chem. Soc., 2006, 128, 14780–14781 CrossRef CAS PubMed .
  17. Y. Liu, J. Yang, E. Gu, T. Cao, Z. Su, L. Jiang and Y. Liu, J. Mater. Chem. A, 2014, 2, 6363–6367 CAS .
  18. M. Achimovičová, F. J. Gotor, C. Real and N. Daneu, J. Mater. Sci.: Mater. Electron., 2012, 23, 1844–1850 CrossRef .
  19. F. M. Ismail and Z. M. Hanafi, Z. Physiol. Chem., 1986, 267, 667 CAS .
  20. A. S. Hewitt, J. Wang, J. Boltersdorf, P. A. Maggard and D. B. Dougherty, J. Vac. Sci. Technol., B: Nanotechnol. Microelectron.: Mater., Process., Meas., Phenom., 2014, 32, 04E103 Search PubMed .
  21. R. M. Imamov and S. A. Semiletov, Soviet Physics - Crystallography, 1971, 15(5), 845–850 Search PubMed .
  22. E. Gaudin, S. Jobic, M. Evain, R. Brec and J. Rouxel, Mater. Res. Bull., 1995, 30(5), 549–561 CrossRef CAS .
  23. X. Peng, Adv. Mater., 2003, 15, 459–463 CrossRef CAS .
  24. Z. A. Peng and X. Peng, J. Am. Chem. Soc., 2002, 124, 3343–3353 CrossRef CAS PubMed .
  25. J. R. Ota, P. Roy, S. K. Srivastava, R. Popovitz-Biro and R. Tenne, Nanotechnology, 2006, 17(6), 1700 CrossRef CAS PubMed .
  26. C. Wang, W. Wan, Y. Huang, J. Chen, H. H. Zhou and X. X. Zhang, Nanoscale, 2014, 6, 5351–5358 CAS .
  27. C. Han, J. Yang, C. Yan, Y. Li, F. Liu, L. Jiang and Y. Liu, CrystEngComm, 2014, 16, 2823–2834 RSC .

Footnote

Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra07570c

This journal is © The Royal Society of Chemistry 2016
Click here to see how this site uses Cookies. View our privacy policy here.