Nanoscale View Article Online
Published on 14 October 2014. Downloaded by Brigham Young University on 18/11/2014 09:51:04.
PAPER
Cite this: DOI: 10.1039/c4nr03148b
View Journal
Structural evolution from Bi4.2K0.8Fe2O9+δ nanobelts to BiFeO3 nanochains in vacuum and their multiferroic properties† Sining Dong,a,b Dalong Zhang,a Yukuai Liu,a Shengwei Yang,a Tao Jiang,a Yuewei Yina and Xiaoguang Li*a In this paper, we report the structural evolution of Bi4.2K0.8Fe2O9+δ nanobelts to BiFeO3 nanochains and the related variations in multiferroic properties. By using in situ transmission electron microscopy with comprehensive characterization, it was found that the layered perovskite multiferroic Bi4.2K0.8Fe2O9+δ nanobelts were very unstable in a vacuum environment, with Bi being easily removed. Based on this finding, a simple vacuum annealing method was designed which successfully transformed the Bi4.2K0.8Fe2O9+δ nanobelts into one-dimensional BiFeO3 nanochains. Both the Bi4.2K0.8Fe2O9+δ nanobelts and the BiFeO3 nanochains showed multiferroic behavior, with their ferroelectric and ferromagnetic
Received 9th June 2014, Accepted 13th October 2014 DOI: 10.1039/c4nr03148b www.rsc.org/nanoscale
1.
properties clearly established by piezoresponse and magnetic measurements, respectively. Interestingly, the BiFeO3 nanochains had a larger magnetization than the Bi4.2K0.8Fe2O9+δ nanobelts. Moreover, the BiFeO3 nanochains exhibited a surprisingly large exchange bias with small training effects. This onedimensional BiFeO3 multiferroic nanostructure characterized by a relatively stable exchange bias offers important functionalities that may be attractive for device applications.
Introduction
Multiferroic materials have attracted significant attention due to their coexisting multiferroic orders and potential magnetoelectric coupling, which not only offer the opportunity to explore interesting physics, but also can be utilized for new classes of magnetoelectric devices, such as sensors, non-volatile memories or information processing applications.1–4 Since the control of different ferroic orders, such as ferromagnetism (FM) and ferroelectricity (FE), on the nanoscale offers unprecedented possibilities for microelectronics, several low-dimensional multiferroic structures have already been fabricated and investigated.5–9 Among these, one-dimensional (1D) nanostructures stand out as particularly important owing to their unique prospects for fabrication of new nanoscale devices.10–14 In this context, bismuth ferrite BiFeO3 (BFO), one of the most important single-phase multiferroic materials, has been widely investigated for the past ten years. BFO is a room-temperature multiferroic material, with a ferroelectric Curie temperature (TC) of about 1100 K and an antiferromagnetic (AFM) Néel
a Hefei National Laboratory for Physical Sciences at Microscale, Department of Physics, CAS Key Laboratory of Materials for Energy Conversion, University of Science and Technology of China, Hefei 230026, P. R. China. E-mail: [email protected] b Department of Physics, University of Notre Dame, Indiana 46556, USA † Electronic supplementary information (ESI) available. See DOI: 10.1039/c4nr03148b
This journal is © The Royal Society of Chemistry 2014
temperature (TN) of around 640 K.15,16 Additionally, the magnetic properties of BFO nanostructures are interesting because size effects in these systems can induce weak ferromagnetism and spin-glass as well as exchange bias.17,18 Recently, room-temperature multiferroic phase Bi4.2K0.8Fe2O9+δ (BKFO) nanobelts have been designed and fabricated, which resulted in large magnetocapacitance effects near 270 K.19 This layered perovskite compound is isostructural with the well-known high temperature superconductor Bi2Sr2CaCu2O8+δ (Bi2212), and its unit cell is composed of alternating Bi2O2 rock salt layers and BiFeO3-like perovskite layers along the c axis.19 The insertion of the Bi2O2 layers between potassium-doped perovskite-like BiFeO3 results in BKFO having a natural dielectric-multiferroic superstructure, in which the multiferroic behavior is spatially modulated. We have also found that the BKFO nanobelts are unstable and can be decomposed (primarily into BiFeO3 and Bi2O3(KBiO2)x) when heated to about 500 °C in air.19 However, this decomposition process is rather difficult to control and results in highly mixed phases, thus making it difficult to isolate pure BiFeO3. Based on the special structural features of BKFO, here we explore the process of large scale fabrication of highly pure phase BFO in the form of 1Dnanochains via an entirely different approach, using BKFO nanobelts as the intermediate material. As will be shown, the multiferroic BFO nanochains as obtained present a relatively stable exchange bias with a small training effect at room temperature, making this system especially attractive for device applications.
Nanoscale
View Article Online
Paper
2. Experimental
Published on 14 October 2014. Downloaded by Brigham Young University on 18/11/2014 09:51:04.
2.1 The synthesis of single crystalline Bi4.2K0.8Fe2O9+δ nanobelts The Bi4.2K0.8Fe2O9+δ nanobelts were prepared by a hydrothermal method. Firstly, 0.485 g (1 mmol) Bi(NO3)3·5H2O, 0.192 g (0.475 mmol) Fe(NO3)3·9H2O and 200 μL concentrated nitric acid were dissolved in deionized (DI) water, the volume of the solution being kept at 5 mL by adding appropriate amounts of DI water. This was then added dropwise to 15 mL of KOH solution (16 mol L−1), magnetically stirred for 120 min, and transferred to a stainless steel Teflon-lined autoclave of 25 mL capacity. The autoclave was heated and maintained at 180 °C for 150 minutes. Finally, the products were collected after separating by precipitation, washed three times with DI water and ethanol, and dried at 120 °C, resulting in Bi4.2K0.8Fe2O9+δ nanobelts. 2.2
Nanoscale
Bi4.2±0.60K0.8±0.12Fe2O9+δ nanobelts was deduced from statistical analysis of EDX results repeated over 100 times, Table S1, ESI†). The BKFO nanobelts grow along the [100] crystal orientation, with an obvious superlattice structure along the b-axis, which can be confirmed from the selected area electron diffraction (SAED) patterns shown in Fig. 1 and in the ESI.† Specifically, Fig. 1 shows in situ TEM images and the dynamic morphology evolution of a single BKFO nanobelt under 200 kV electron beam (e-beam) focused irradiation (the real-time in situ
The synthesis of BiFeO3 nanochains
The obtained Bi4.2K0.8Fe2O9+δ nanobelts were put into a quartz tube. The tube was pumped down to a vacuum of about 10−2 Pa and sealed. The quartz tube was then heated to 500 °C and maintained at that temperature for 120 minutes. Finally, the products were washed with 10 wt% dilute nitric acid (once), DI water and absolute ethanol (3 times), and dried at 120 °C, resulting in single-phase BiFeO3 nanochains. 2.3
Materials characterization
X-ray diffraction (XRD) patterns were measured for the BKFO nanobelts and the BFO nanochains using a Rigaku D/Max-rA Xray diffractometer. Field emission scanning electron microscopy (FE-SEM) images were obtained for these systems using a FEI Sirion microscope, and transmission electron microscopy (TEM) images were taken on a JEOL JEM-ARM200F microscope. Piezoresponse force microscopy (PFM) measurements were carried out on a Veeco DI multimode V SPM. The magnetocapacitance measurements were performed using an Agilent 4294A LCR meter, with the applied magnetic field and the sample temperature controlled by a Quantum Design PPMS instrument. For the magnetocapacitance measurements, the BKFO and BFO powders were pelletized and formed into a tetragonal bulk sample (2 × 2 × 0.5 mm3) in a stainless steel mold under a vertical pressure of 20 MPa, and silver electrodes were then deposited onto the opposite pressure surfaces. Finally, the magnetic properties of the BKFO and BFO samples were determined with a Quantum Design SQUID-VSM.
3. Results and discussion As described in Section 2 above, single-crystalline Bi4.2K0.8Fe2O9+δ nanobelts were synthesized by a simple hydrothermal method.20 The as-prepared BKFO nanobelts were obtained on a large scale without any detectable impurities, as indicated by Xray diffraction (XRD), field emission scanning electron microscopy (FE-SEM), energy dispersive X-ray analysis (EDX) and high resolution transmission electron microscopy (HR-TEM) for the nanobelts (see Fig. S1–S4, ESI.† The chemical formula of
Nanoscale
Fig. 1 Real-time TEM images of Bi nanoparticles grown from a BKFO nanobelt: (a) initial morphology image; (b) SAED pattern; TEM images of nanobelt after (c) 95 s, (d) 185 s, (e) 886 s and (g) 1227 s of in situ electron beam irradiation; (f ) enlarged view of the Bi nanoparticle in (e); (h) shows the SAED pattern of (g). The marked lattice fringes in (f ) indicate the Bi (104) face.
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 14 October 2014. Downloaded by Brigham Young University on 18/11/2014 09:51:04.
Nanoscale
TEM videos are provided as ESI Videos, SVideos 1–4†). Interestingly, a number of nanoparticles are gradually generated on the surface of the nanobelt by the e-beam irradiation. One can see that, with increasing irradiation time, more nanoparticles appear and grow larger until their diameters reach about 10 nm. By analyzing the HR-TEM lattice fringes and SAED patterns, these nanoparticles can be identified as consisting of bismuth. In addition, we find that the 200 kV e-beam is not sufficiently powerful to destroy the physical structure of BKFO, and the SAED patterns of BKFO are seen to remain sharp and bright, as presented in Fig. 1h. Fig. 2 shows that the formation of the Bi nanoparticles on BKFO nanobelts can be divided into two sub-processes: the formation of very fine Bi nanoparticles, and their fusion with neighboring Bi nanoparticles. Specifically, we note that Bi nanoparticles with a diameter smaller than 5 nm are very unstable, roughly spherical, and have lattice fringes that are rather weak and change rapidly. By contrast, a fully formed 10 nm Bi nanoparticle becomes much more stable, with the corresponding HR-TEM lattices becoming clear, as the nanoparticle with a hexagonal shape gradually forms. We note that electron irradiation damage during TEM observation has also been observed in other bismuth-based systems, such as NaBiO3 and NiBi alloys.21,22 In the present case, a number of the Bi3+
Fig. 2 Electron-beam-driven growth of bismuth nanoparticles on BKFO nanobelts. The relative irradiation times are shown in each figure. The marked lattice fringes in (d) indicate the Bi (012) face. Pi (i = 1, 2, 3, 4, 5, 6) represents Bi particles. P5 shows collapse of P2 and P1, and P6 shows collapse of P3 and P5.
This journal is © The Royal Society of Chemistry 2014
Paper
ions in BKFO may be converted to Bi0 by the electron beam and migrate out from the BKFO nanobelts due to the ultra-high vacuum environment of the TEM chamber, hence there should be many Bi vacancies inside the BKFO nanobelts after the irradiation. This kind of continuous dynamic process has not been reported for complex bismuth systems before. This observation presents a clue regarding the structural evolution from Bi4.2K0.8Fe2O9+δ nanobelts to another kind of nanostructure through such segregation of Bi from the parent compound. To confirm the effect of heating in a vacuum, the assynthesized Bi4.2K0.8Fe2O9+δ nanobelts were annealed in vacuo at 500 °C for 2 hours and then washed in dilute nitric acid solution. The final products were nanochains composed of pure single-phase BiFeO3 nanocrystals (see Fig. S4 and S5, ESI†). The 500 °C annealing process offers a significantly higher amount of energy than the TEM electron beam irradiation, so that all the extra bismuth as well as potassium is separated out or evaporated. Our results show that the BFO nanochains are covered by layers of Bi25FeO39 instead of pure bismuth because of the different conditions that exist during the vacuum annealing and during ultra-high-vacuum e-beam irradiation. As the dilute nitric acid can dissolve Bi25FeO39 but not BiFeO3,23 the additional wash in acid is used to remove the surface layer, and only pure BFO in the form of nanocrystal chains survives. In addition, oxygen vacancies in BFO can be introduced during the vacuum annealing process, and washing in strongly oxidizing nitric acid serves to re-oxidize BFO. These treatments may be advantageous for the magnetic and ferroelectric properties of the resulting BFO product.17
Fig. 3 (a, b) TEM images at different magnifications of BiFeO3 nanochains. (c) HR-TEM image and (d) SAED pattern of the BFO nanocrystal marked in (a) are also shown.
Nanoscale
View Article Online
Published on 14 October 2014. Downloaded by Brigham Young University on 18/11/2014 09:51:04.
Paper
Fig. 3 shows TEM images of a typical BFO nanochain, an HRTEM lattice image, and a corresponding SAED pattern of a BFO nanocrystal in the chain. Every particle in the nanochain is a single crystal with nearly perfect crystallinity, but with a different crystallographic orientation to the adjacent particle, indicating that the interface of the two adjoining BFO crystals is not fully epitaxial. One can thus conclude that BFO nanochains are converted from Bi4.2K0.8Fe2O9+δ nanobelts. The BFO nanocrystal presented in Fig. 3a and c has its (104) face (∼0.279 nm) aligned with the direction of the chain, which is actually equivalent to the (200) face of the original Bi4.2K0.8Fe2O9+δ (∼0.277 nm, see Fig. S3†). It also should be pointed out that the neighboring BFO nanocrystals in the chains are firmly fastened to each other, so that the chains are not easy to break by ultrasound cleaning. The key point of the successful transformation from the BKFO phase to BFO using vacuum annealing is the difference in the volatilities of different cations in BKFO. It is well established that at a given temperature the vapor pressure of K is significantly higher than that of Bi, and the vapor pressure of Fe is much lower than both of these elements. Specifically, it is expected that for BKFO in a vacuum environment at 500 °C, K will evaporate almost entirely, Bi will partially evaporate, and Fe will remain essentially unaffected.24,25 Therefore, it is suggested that K and Bi cations in BKFO have relatively low volatilization temperatures during the 500 °C vacuum annealing process, thus all K cations and some of the Bi cations can easily escape from the perovskite-like precursor. Finally, the non-volatile Fe cations re-react with the remaining Bi cations to form the pureperovskite BFO lattice. Our synthesis method may provide a useful route for preparing other complex nanomaterials from the decomposition of intermediate compounds, in which there are cations with significantly different volatilities. Piezoresponse force microscopy (PFM) was used to investigate the ferroelectric properties of the Bi4.2K0.8Fe2O9+δ nanobelts and the BiFeO3 nanochains at room temperature. From Fig. 4a–d one can see that the phase of the PFM signals can be tuned over nearly 180° by sweeping the DC electric field (Fig. 4a and b), and clear piezoelectric butterfly curves can thus be simultaneously obtained (Fig. 4c and d). It is notable that both the local PFM hysteresis loop and the butterfly curve for BFO nanocrystals exhibit a horizontal shift (see Fig. 4b and d), which may be induced by a built-in electric field in the ferroelectric nanocrystal by the shape of the specimen.26 Moreover, the large magnetocapacitance effect near 270 K observed in bulk specimens prepared from Bi4.2K0.8Fe2O9+δ nanobelts indicates strong coupling behavior between magnetic and electric orders in this system, as shown in Fig. 5a. Fig. 5b shows the roomtemperature magnetocapacitance coefficient as a function of applied magnetic field measured at 1 kHz for BFO and BKFO bulk specimens. One can see that BFO exhibits a positive magnetocapacitance effect (about 0.4% at 50 kOe) which is consistent with earlier reports on BFO-based nanosystems.27,28 By contrast, the BKFO bulk specimens show a negative magnetocapacitance effect (about −20% at 50 kOe), indicating the different origins of the magnetocapacitance effects in BKFO
Nanoscale
Nanoscale
Fig. 4 Local PFM phase and amplitude signals obtained (a, c) for a BKFO nanobelt and (b, d) for a BFO nanocrystal. The black curve in (a) is the signal of the background for the measured system.
and BFO. The giant magnetocapacitance coefficient in BKFO near the ferroelectric transition temperature might be related to its novel ferroelectricity,19 which is unclear at present. On the other hand, these low-dimensional nanomaterials with strong ferroelectric properties should be of applied interest, since they can be easily integrated with 2D monolayer materials (such as graphene or MoS2) as nanoscale ferroelectric gates, thus providing a path to next-generation functional devices. It is also interesting to study the magnetic properties of the nanochains composed of BiFeO3 nanocrystals, because the exchange coupling interactions at interfaces between contiguous particles are likely to result in new magnetic behaviors. The zero field cooled (ZFC) and field cooled (FC) M–T curves of the BFO nanochains are presented in Fig. S6 (ESI†). No
Fig. 5 Magnetocapacitance effects (a) as a function of temperature and (b) as a function of applied magnetic field for the bulk samples of Bi4.2K0.8Fe2O9+δ nanobelts and BiFeO3 nanochains.
This journal is © The Royal Society of Chemistry 2014
View Article Online
Nanoscale
Published on 14 October 2014. Downloaded by Brigham Young University on 18/11/2014 09:51:04.
indication of a spin-glass (SG) transition is observed, which is quite different from the SG-like BFO nanosystems reported earlier,17,18 and is more similar to the high quality BFO singlecrystals reported by Lebeugle et al.29 This may be attributed to two reasons: (1) the surface spins of the BFO nanochains are not frustrated due to their high degree of crystallinity with very few defects; (2) the structural phase of the BFO nanochains
Paper
obtained by our method is very pure, and there are no other magnetic impurities that could contribute to the magnetic signals. Fig. 6a and b show the M–H loops measured for the BFO nanochains at 300 K and 2 K, respectively, in the FC mode with a cooling field HFC = ±50 kOe (cooling from 400 K). The magnetization of the BFO nanochains at 50 kOe is somewhat smaller than that of individual BFO nanocrystals fabricated via the hydrothermal method,18 possibly due to a reduction of the specific surface area when the BFO nanocrystals are linked together. On the other hand, the magnetization observed for the BFO nanochains is larger than the values obtained for BKFO nanobelts at both 300 K and 2 K. More importantly, the hysteresis loops of the BFO nanochains shown in Fig. 6c and d exhibit a shift toward the direction opposite to the applied HFC, indicating the presence of exchange bias in these nanostructures. The observed shift can be described as HEB = (H1 + H2)/2, where HEB is the exchange bias field, and H1 and H2 are the left and right coercive fields, respectively. We can thus calculate that HEB(300 K) = 289.6 Oe and HEB(2 K) = 688.9 Oe. These values are much larger than those of separated individual BFO nanocrystals of similar size.18 The training effect for the BFO nanochains at 300 K and 2 K is shown in Fig. 6e and f, respectively. By making consecutive measurements of the hysteresis loop, in this case 10 times, we observe only very small changes after the second and all subsequent repetitions, both at 2 K and at 300 K. The dependence of the exchange bias fields on the repeat number (n) is presented in Fig. 6g. HEB can then be well-fitted using the recursive expression as:30 H EB ðn þ 1Þ H EB ðnÞ ¼ γðH EB ðnÞ H EB ð1ÞÞ3 ;
ð1Þ
where HEB(∞) is the exchange bias field in the limit of infinite loops, and γ is a system-dependent constant. The fitting parameters for 300 K are HEB(∞) = −97 ± 7 Oe and γ = (1.7 ± 0.2) × 10−5 Oe−2 (HFC = +50 kOe), and for 2 K HEB(∞) = 499 ± 6 Oe and γ = (1.9 ± 0.2) × 10−5 Oe−2 (HFC = −50 kOe). Such a large and relatively stable exchange bias effect as seen in the present BFO nanochains is quite rare, being different from previously reported results for single-phase magnetic nanosystems, and may be attractive for device applications. The mechanism of such unusual behavior needs further investigation.
4. Fig. 6 Exchange bias in BFO nanochains: M–H curves of the BFO nanochains measured at (a) 300 K and (b) 2 K in FC mode with HFC = ±50 kOe, and (c, d) the corresponding enlarged views in the low field range. The M–H curves of the BKFO nanobelts (black open squares) are also presented for comparison. Measurements of continuous 10 circle hysteresis loops of the training effect are also shown for (e) 300 K and (f ) 2 K. Panel (g) shows HEB as a function of the repeat number of the hysteresis sweep n obtained at 300 K (HFC = +50 kOe) and at 2 K (HFC = −50 kOe).
This journal is © The Royal Society of Chemistry 2014
Conclusion
In summary, the structural evolution from layered perovskitelike Bi4.2K0.8Fe2O9+δ nanobelts to the 1D BiFeO3 nanochains was successfully performed and studied by a specially designed vacuum annealing method. The ferroelectric properties of the BiFeO3 nanochains and the Bi4.2K0.8Fe2O9+δ nanobelts were established by PFM measurements. Most importantly, a very large and stable exchange bias effect in the as-synthesized BFO nanochains was observed, which may be attractive for magnetoelectric devices.
Nanoscale
View Article Online
Paper
Nanoscale
Published on 14 October 2014. Downloaded by Brigham Young University on 18/11/2014 09:51:04.
Acknowledgements This work is supported by the National Natural Science Foundation of China, the National Basic Research Program of China (contract nos. 2012CB922003 2015CB921201 and 2011CBA00102), the Fundamental Research Funds for the Central Universities (contract no. WK2060140009 and WK2030020026), and the China Postdoctoral Science Foundation (contract no. 2014T70590 and 2013M540513). The authors would like to address special thanks to Prof. J. K. Furdyna (University of Notre Dame) for his kind interest and helpful advice in this work.
14 15
16 17
18
References
19
1 W. Eerenstein, N. D. Mathur and J. F. Scott, Nature, 2006, 442, 759–765. 2 T. Choi, S. Lee, Y. J. Choi, V. Kiryukhin and S. W. Cheong, Science, 2009, 324, 63–66. 3 Y. Tokura and S. Seki, Adv. Mater., 2010, 22, 1554–1565. 4 N. A. Spaldin, S. W. Cheong and R. Ramesh, Phys. Today, 2010, 63, 38–43. 5 X. L. Lu, Y. Kim, S. Goetze, X. G. Li, S. N. Dong, P. Werner, M. Alexe and D. Hesse, Nano Lett., 2011, 11, 3202–3206. 6 M. Bibes, Nat. Mater., 2012, 11, 354–357. 7 J. X. Zhang, R. J. Zeches, Q. He, Y. H. Chu and R. Ramesh, Nanoscale, 2012, 4, 6196–6204. 8 N. Nuraje and K. Su, Nanoscale, 2013, 5, 8752–8780. 9 S. Hong, T. Choi, J. H. Jeon, Y. Kim, H. Lee, H. Y. Joo, I. Hwang, J. S. Kim, S. O. Kang, S. V. Kalinin and B. H. Park, Adv. Mater., 2013, 25, 2339–2343. 10 J. Cao, E. Ertekin, V. Srinivasan, W. Fan, S. Huang, H. Zheng, J. W. L. Yim, D. R. Khanal, D. F. Ogletree, J. C. Grossmanan and J. Wu, Nat. Nanotechnol., 2009, 4, 732–737. 11 D. P. Dutta, O. D. Jayakumar, A. K. Tyagi, K. G. Girija, C. G. S. Pillai and G. Sharma, Nanoscale, 2010, 2, 1149–1154. 12 P. M. Rorvik, T. Grande and M. A. Einarsrud, Adv. Mater., 2011, 23, 4007–4034. 13 N. Lei, T. Devolder, G. Agnus, P. Aubert, L. Daniel, J. V. Kim, W. S. Zhao, T. Trypiniotis, R. P. Cowburn,
20
Nanoscale
21
22 23 24 25 26 27
28
29
30
C. Chappert, D. Ravelosona and P. Lecoeur, Nat. Commun., 2013, 4, 1378. C. Liu, N. P. Dasgupta and P. D. Yang, Chem. Mater., 2014, 26, 415–422. J. Wang, J. B. Neaton, H. Zheng, V. Nagarajan, S. B. Ogale, B. Liu, D. Viehland, V. Vaithyanathan, D. G. Schlom, U. V. Waghmare, N. A. Spaldin, K. M. Rabe, M. Wuttig and R. Ramesh, Science, 2003, 299, 1719–1722. G. Catalan and J. F. Scott, Adv. Mater., 2009, 21, 2463–2485. T. J. Park, G. C. Papaefthymiou, A. J. Viescas, A. R. Moodenbaugh and S. S. Wong, Nano Lett., 2007, 7, 766–772. S. N. Dong, Y. P. Yao, Y. Hou, Y. K. Liu, Y. Tang and X. G. Li, Nanotechnology, 2011, 22, 385701. S. N. Dong, Y. P. Yao, J. Q. Li, Y. J. Song, Y. K. Liu and X. G. Li, Sci. Rep., 2013, 3, 1245. S. N. Dong, Y. K. Liu, S. W. Yang, T. Jiang, Y. W. Yin and X. G. Li, CrystEngComm, 2013, 15, 9057–9063. S. Sepulveda-Guzman, N. Elizondo-Villarreal, D. F. A. Torres-Castro, X. Gao, J. P. Zhou and M. JoseYacaman, Nanotechnology, 2007, 18, 335604. W. D. Pyrz, S. Park, D. A. Blom, D. J. Buttrey and T. Vogt, J. Phys. Chem. C, 2010, 114, 2538–2543. G. D. Achenbac, W. J. James and R. Gerson, J. Am. Ceram. Soc., 1967, 50, 437. A. N. Nesmeianov, Vapor pressure of the chemical elements, Elsevier Pub. Co., 1963. D. R. Lide, CRC Handbook of Chemistry and Physics, Taylor & Francis, 90th edn, 2009. W. H. Ma and D. Hesse, Appl. Phys. Lett., 2004, 84, 2871– 2873. D. P. Dutta, B. P. Mandal, R. Naik, G. Lawes and A. K. Tyagi, J. Phys. Chem. C, 2013, 117, 2382– 2389. D. P. Dutta, B. P. Mandal, M. D. Mukadam, S. M. Yusuf and A. K. Tyagi, Dalton Trans., 2014, 43, 7838– 7846. D. Lebeugle, D. Colson, A. Forget, M. Viret, P. Bonville, J. F. Marucco and S. Fusil, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 76, 024116. C. Binek, Phys. Rev. B: Condens. Matter Mater. Phys., 2004, 70, 014421.
This journal is © The Royal Society of Chemistry 2014
Published on 14 October 2014. Downloaded by Brigham Young University on 18/11/2014 09:51:04.
PAPER
Cite this: DOI: 10.1039/c4nr03148b
View Journal
Structural evolution from Bi4.2K0.8Fe2O9+δ nanobelts to BiFeO3 nanochains in vacuum and their multiferroic properties† Sining Dong,a,b Dalong Zhang,a Yukuai Liu,a Shengwei Yang,a Tao Jiang,a Yuewei Yina and Xiaoguang Li*a In this paper, we report the structural evolution of Bi4.2K0.8Fe2O9+δ nanobelts to BiFeO3 nanochains and the related variations in multiferroic properties. By using in situ transmission electron microscopy with comprehensive characterization, it was found that the layered perovskite multiferroic Bi4.2K0.8Fe2O9+δ nanobelts were very unstable in a vacuum environment, with Bi being easily removed. Based on this finding, a simple vacuum annealing method was designed which successfully transformed the Bi4.2K0.8Fe2O9+δ nanobelts into one-dimensional BiFeO3 nanochains. Both the Bi4.2K0.8Fe2O9+δ nanobelts and the BiFeO3 nanochains showed multiferroic behavior, with their ferroelectric and ferromagnetic
Received 9th June 2014, Accepted 13th October 2014 DOI: 10.1039/c4nr03148b www.rsc.org/nanoscale
1.
properties clearly established by piezoresponse and magnetic measurements, respectively. Interestingly, the BiFeO3 nanochains had a larger magnetization than the Bi4.2K0.8Fe2O9+δ nanobelts. Moreover, the BiFeO3 nanochains exhibited a surprisingly large exchange bias with small training effects. This onedimensional BiFeO3 multiferroic nanostructure characterized by a relatively stable exchange bias offers important functionalities that may be attractive for device applications.
Introduction
Multiferroic materials have attracted significant attention due to their coexisting multiferroic orders and potential magnetoelectric coupling, which not only offer the opportunity to explore interesting physics, but also can be utilized for new classes of magnetoelectric devices, such as sensors, non-volatile memories or information processing applications.1–4 Since the control of different ferroic orders, such as ferromagnetism (FM) and ferroelectricity (FE), on the nanoscale offers unprecedented possibilities for microelectronics, several low-dimensional multiferroic structures have already been fabricated and investigated.5–9 Among these, one-dimensional (1D) nanostructures stand out as particularly important owing to their unique prospects for fabrication of new nanoscale devices.10–14 In this context, bismuth ferrite BiFeO3 (BFO), one of the most important single-phase multiferroic materials, has been widely investigated for the past ten years. BFO is a room-temperature multiferroic material, with a ferroelectric Curie temperature (TC) of about 1100 K and an antiferromagnetic (AFM) Néel
a Hefei National Laboratory for Physical Sciences at Microscale, Department of Physics, CAS Key Laboratory of Materials for Energy Conversion, University of Science and Technology of China, Hefei 230026, P. R. China. E-mail: [email protected] b Department of Physics, University of Notre Dame, Indiana 46556, USA † Electronic supplementary information (ESI) available. See DOI: 10.1039/c4nr03148b
This journal is © The Royal Society of Chemistry 2014
temperature (TN) of around 640 K.15,16 Additionally, the magnetic properties of BFO nanostructures are interesting because size effects in these systems can induce weak ferromagnetism and spin-glass as well as exchange bias.17,18 Recently, room-temperature multiferroic phase Bi4.2K0.8Fe2O9+δ (BKFO) nanobelts have been designed and fabricated, which resulted in large magnetocapacitance effects near 270 K.19 This layered perovskite compound is isostructural with the well-known high temperature superconductor Bi2Sr2CaCu2O8+δ (Bi2212), and its unit cell is composed of alternating Bi2O2 rock salt layers and BiFeO3-like perovskite layers along the c axis.19 The insertion of the Bi2O2 layers between potassium-doped perovskite-like BiFeO3 results in BKFO having a natural dielectric-multiferroic superstructure, in which the multiferroic behavior is spatially modulated. We have also found that the BKFO nanobelts are unstable and can be decomposed (primarily into BiFeO3 and Bi2O3(KBiO2)x) when heated to about 500 °C in air.19 However, this decomposition process is rather difficult to control and results in highly mixed phases, thus making it difficult to isolate pure BiFeO3. Based on the special structural features of BKFO, here we explore the process of large scale fabrication of highly pure phase BFO in the form of 1Dnanochains via an entirely different approach, using BKFO nanobelts as the intermediate material. As will be shown, the multiferroic BFO nanochains as obtained present a relatively stable exchange bias with a small training effect at room temperature, making this system especially attractive for device applications.
Nanoscale
View Article Online
Paper
2. Experimental
Published on 14 October 2014. Downloaded by Brigham Young University on 18/11/2014 09:51:04.
2.1 The synthesis of single crystalline Bi4.2K0.8Fe2O9+δ nanobelts The Bi4.2K0.8Fe2O9+δ nanobelts were prepared by a hydrothermal method. Firstly, 0.485 g (1 mmol) Bi(NO3)3·5H2O, 0.192 g (0.475 mmol) Fe(NO3)3·9H2O and 200 μL concentrated nitric acid were dissolved in deionized (DI) water, the volume of the solution being kept at 5 mL by adding appropriate amounts of DI water. This was then added dropwise to 15 mL of KOH solution (16 mol L−1), magnetically stirred for 120 min, and transferred to a stainless steel Teflon-lined autoclave of 25 mL capacity. The autoclave was heated and maintained at 180 °C for 150 minutes. Finally, the products were collected after separating by precipitation, washed three times with DI water and ethanol, and dried at 120 °C, resulting in Bi4.2K0.8Fe2O9+δ nanobelts. 2.2
Nanoscale
Bi4.2±0.60K0.8±0.12Fe2O9+δ nanobelts was deduced from statistical analysis of EDX results repeated over 100 times, Table S1, ESI†). The BKFO nanobelts grow along the [100] crystal orientation, with an obvious superlattice structure along the b-axis, which can be confirmed from the selected area electron diffraction (SAED) patterns shown in Fig. 1 and in the ESI.† Specifically, Fig. 1 shows in situ TEM images and the dynamic morphology evolution of a single BKFO nanobelt under 200 kV electron beam (e-beam) focused irradiation (the real-time in situ
The synthesis of BiFeO3 nanochains
The obtained Bi4.2K0.8Fe2O9+δ nanobelts were put into a quartz tube. The tube was pumped down to a vacuum of about 10−2 Pa and sealed. The quartz tube was then heated to 500 °C and maintained at that temperature for 120 minutes. Finally, the products were washed with 10 wt% dilute nitric acid (once), DI water and absolute ethanol (3 times), and dried at 120 °C, resulting in single-phase BiFeO3 nanochains. 2.3
Materials characterization
X-ray diffraction (XRD) patterns were measured for the BKFO nanobelts and the BFO nanochains using a Rigaku D/Max-rA Xray diffractometer. Field emission scanning electron microscopy (FE-SEM) images were obtained for these systems using a FEI Sirion microscope, and transmission electron microscopy (TEM) images were taken on a JEOL JEM-ARM200F microscope. Piezoresponse force microscopy (PFM) measurements were carried out on a Veeco DI multimode V SPM. The magnetocapacitance measurements were performed using an Agilent 4294A LCR meter, with the applied magnetic field and the sample temperature controlled by a Quantum Design PPMS instrument. For the magnetocapacitance measurements, the BKFO and BFO powders were pelletized and formed into a tetragonal bulk sample (2 × 2 × 0.5 mm3) in a stainless steel mold under a vertical pressure of 20 MPa, and silver electrodes were then deposited onto the opposite pressure surfaces. Finally, the magnetic properties of the BKFO and BFO samples were determined with a Quantum Design SQUID-VSM.
3. Results and discussion As described in Section 2 above, single-crystalline Bi4.2K0.8Fe2O9+δ nanobelts were synthesized by a simple hydrothermal method.20 The as-prepared BKFO nanobelts were obtained on a large scale without any detectable impurities, as indicated by Xray diffraction (XRD), field emission scanning electron microscopy (FE-SEM), energy dispersive X-ray analysis (EDX) and high resolution transmission electron microscopy (HR-TEM) for the nanobelts (see Fig. S1–S4, ESI.† The chemical formula of
Nanoscale
Fig. 1 Real-time TEM images of Bi nanoparticles grown from a BKFO nanobelt: (a) initial morphology image; (b) SAED pattern; TEM images of nanobelt after (c) 95 s, (d) 185 s, (e) 886 s and (g) 1227 s of in situ electron beam irradiation; (f ) enlarged view of the Bi nanoparticle in (e); (h) shows the SAED pattern of (g). The marked lattice fringes in (f ) indicate the Bi (104) face.
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 14 October 2014. Downloaded by Brigham Young University on 18/11/2014 09:51:04.
Nanoscale
TEM videos are provided as ESI Videos, SVideos 1–4†). Interestingly, a number of nanoparticles are gradually generated on the surface of the nanobelt by the e-beam irradiation. One can see that, with increasing irradiation time, more nanoparticles appear and grow larger until their diameters reach about 10 nm. By analyzing the HR-TEM lattice fringes and SAED patterns, these nanoparticles can be identified as consisting of bismuth. In addition, we find that the 200 kV e-beam is not sufficiently powerful to destroy the physical structure of BKFO, and the SAED patterns of BKFO are seen to remain sharp and bright, as presented in Fig. 1h. Fig. 2 shows that the formation of the Bi nanoparticles on BKFO nanobelts can be divided into two sub-processes: the formation of very fine Bi nanoparticles, and their fusion with neighboring Bi nanoparticles. Specifically, we note that Bi nanoparticles with a diameter smaller than 5 nm are very unstable, roughly spherical, and have lattice fringes that are rather weak and change rapidly. By contrast, a fully formed 10 nm Bi nanoparticle becomes much more stable, with the corresponding HR-TEM lattices becoming clear, as the nanoparticle with a hexagonal shape gradually forms. We note that electron irradiation damage during TEM observation has also been observed in other bismuth-based systems, such as NaBiO3 and NiBi alloys.21,22 In the present case, a number of the Bi3+
Fig. 2 Electron-beam-driven growth of bismuth nanoparticles on BKFO nanobelts. The relative irradiation times are shown in each figure. The marked lattice fringes in (d) indicate the Bi (012) face. Pi (i = 1, 2, 3, 4, 5, 6) represents Bi particles. P5 shows collapse of P2 and P1, and P6 shows collapse of P3 and P5.
This journal is © The Royal Society of Chemistry 2014
Paper
ions in BKFO may be converted to Bi0 by the electron beam and migrate out from the BKFO nanobelts due to the ultra-high vacuum environment of the TEM chamber, hence there should be many Bi vacancies inside the BKFO nanobelts after the irradiation. This kind of continuous dynamic process has not been reported for complex bismuth systems before. This observation presents a clue regarding the structural evolution from Bi4.2K0.8Fe2O9+δ nanobelts to another kind of nanostructure through such segregation of Bi from the parent compound. To confirm the effect of heating in a vacuum, the assynthesized Bi4.2K0.8Fe2O9+δ nanobelts were annealed in vacuo at 500 °C for 2 hours and then washed in dilute nitric acid solution. The final products were nanochains composed of pure single-phase BiFeO3 nanocrystals (see Fig. S4 and S5, ESI†). The 500 °C annealing process offers a significantly higher amount of energy than the TEM electron beam irradiation, so that all the extra bismuth as well as potassium is separated out or evaporated. Our results show that the BFO nanochains are covered by layers of Bi25FeO39 instead of pure bismuth because of the different conditions that exist during the vacuum annealing and during ultra-high-vacuum e-beam irradiation. As the dilute nitric acid can dissolve Bi25FeO39 but not BiFeO3,23 the additional wash in acid is used to remove the surface layer, and only pure BFO in the form of nanocrystal chains survives. In addition, oxygen vacancies in BFO can be introduced during the vacuum annealing process, and washing in strongly oxidizing nitric acid serves to re-oxidize BFO. These treatments may be advantageous for the magnetic and ferroelectric properties of the resulting BFO product.17
Fig. 3 (a, b) TEM images at different magnifications of BiFeO3 nanochains. (c) HR-TEM image and (d) SAED pattern of the BFO nanocrystal marked in (a) are also shown.
Nanoscale
View Article Online
Published on 14 October 2014. Downloaded by Brigham Young University on 18/11/2014 09:51:04.
Paper
Fig. 3 shows TEM images of a typical BFO nanochain, an HRTEM lattice image, and a corresponding SAED pattern of a BFO nanocrystal in the chain. Every particle in the nanochain is a single crystal with nearly perfect crystallinity, but with a different crystallographic orientation to the adjacent particle, indicating that the interface of the two adjoining BFO crystals is not fully epitaxial. One can thus conclude that BFO nanochains are converted from Bi4.2K0.8Fe2O9+δ nanobelts. The BFO nanocrystal presented in Fig. 3a and c has its (104) face (∼0.279 nm) aligned with the direction of the chain, which is actually equivalent to the (200) face of the original Bi4.2K0.8Fe2O9+δ (∼0.277 nm, see Fig. S3†). It also should be pointed out that the neighboring BFO nanocrystals in the chains are firmly fastened to each other, so that the chains are not easy to break by ultrasound cleaning. The key point of the successful transformation from the BKFO phase to BFO using vacuum annealing is the difference in the volatilities of different cations in BKFO. It is well established that at a given temperature the vapor pressure of K is significantly higher than that of Bi, and the vapor pressure of Fe is much lower than both of these elements. Specifically, it is expected that for BKFO in a vacuum environment at 500 °C, K will evaporate almost entirely, Bi will partially evaporate, and Fe will remain essentially unaffected.24,25 Therefore, it is suggested that K and Bi cations in BKFO have relatively low volatilization temperatures during the 500 °C vacuum annealing process, thus all K cations and some of the Bi cations can easily escape from the perovskite-like precursor. Finally, the non-volatile Fe cations re-react with the remaining Bi cations to form the pureperovskite BFO lattice. Our synthesis method may provide a useful route for preparing other complex nanomaterials from the decomposition of intermediate compounds, in which there are cations with significantly different volatilities. Piezoresponse force microscopy (PFM) was used to investigate the ferroelectric properties of the Bi4.2K0.8Fe2O9+δ nanobelts and the BiFeO3 nanochains at room temperature. From Fig. 4a–d one can see that the phase of the PFM signals can be tuned over nearly 180° by sweeping the DC electric field (Fig. 4a and b), and clear piezoelectric butterfly curves can thus be simultaneously obtained (Fig. 4c and d). It is notable that both the local PFM hysteresis loop and the butterfly curve for BFO nanocrystals exhibit a horizontal shift (see Fig. 4b and d), which may be induced by a built-in electric field in the ferroelectric nanocrystal by the shape of the specimen.26 Moreover, the large magnetocapacitance effect near 270 K observed in bulk specimens prepared from Bi4.2K0.8Fe2O9+δ nanobelts indicates strong coupling behavior between magnetic and electric orders in this system, as shown in Fig. 5a. Fig. 5b shows the roomtemperature magnetocapacitance coefficient as a function of applied magnetic field measured at 1 kHz for BFO and BKFO bulk specimens. One can see that BFO exhibits a positive magnetocapacitance effect (about 0.4% at 50 kOe) which is consistent with earlier reports on BFO-based nanosystems.27,28 By contrast, the BKFO bulk specimens show a negative magnetocapacitance effect (about −20% at 50 kOe), indicating the different origins of the magnetocapacitance effects in BKFO
Nanoscale
Nanoscale
Fig. 4 Local PFM phase and amplitude signals obtained (a, c) for a BKFO nanobelt and (b, d) for a BFO nanocrystal. The black curve in (a) is the signal of the background for the measured system.
and BFO. The giant magnetocapacitance coefficient in BKFO near the ferroelectric transition temperature might be related to its novel ferroelectricity,19 which is unclear at present. On the other hand, these low-dimensional nanomaterials with strong ferroelectric properties should be of applied interest, since they can be easily integrated with 2D monolayer materials (such as graphene or MoS2) as nanoscale ferroelectric gates, thus providing a path to next-generation functional devices. It is also interesting to study the magnetic properties of the nanochains composed of BiFeO3 nanocrystals, because the exchange coupling interactions at interfaces between contiguous particles are likely to result in new magnetic behaviors. The zero field cooled (ZFC) and field cooled (FC) M–T curves of the BFO nanochains are presented in Fig. S6 (ESI†). No
Fig. 5 Magnetocapacitance effects (a) as a function of temperature and (b) as a function of applied magnetic field for the bulk samples of Bi4.2K0.8Fe2O9+δ nanobelts and BiFeO3 nanochains.
This journal is © The Royal Society of Chemistry 2014
View Article Online
Nanoscale
Published on 14 October 2014. Downloaded by Brigham Young University on 18/11/2014 09:51:04.
indication of a spin-glass (SG) transition is observed, which is quite different from the SG-like BFO nanosystems reported earlier,17,18 and is more similar to the high quality BFO singlecrystals reported by Lebeugle et al.29 This may be attributed to two reasons: (1) the surface spins of the BFO nanochains are not frustrated due to their high degree of crystallinity with very few defects; (2) the structural phase of the BFO nanochains
Paper
obtained by our method is very pure, and there are no other magnetic impurities that could contribute to the magnetic signals. Fig. 6a and b show the M–H loops measured for the BFO nanochains at 300 K and 2 K, respectively, in the FC mode with a cooling field HFC = ±50 kOe (cooling from 400 K). The magnetization of the BFO nanochains at 50 kOe is somewhat smaller than that of individual BFO nanocrystals fabricated via the hydrothermal method,18 possibly due to a reduction of the specific surface area when the BFO nanocrystals are linked together. On the other hand, the magnetization observed for the BFO nanochains is larger than the values obtained for BKFO nanobelts at both 300 K and 2 K. More importantly, the hysteresis loops of the BFO nanochains shown in Fig. 6c and d exhibit a shift toward the direction opposite to the applied HFC, indicating the presence of exchange bias in these nanostructures. The observed shift can be described as HEB = (H1 + H2)/2, where HEB is the exchange bias field, and H1 and H2 are the left and right coercive fields, respectively. We can thus calculate that HEB(300 K) = 289.6 Oe and HEB(2 K) = 688.9 Oe. These values are much larger than those of separated individual BFO nanocrystals of similar size.18 The training effect for the BFO nanochains at 300 K and 2 K is shown in Fig. 6e and f, respectively. By making consecutive measurements of the hysteresis loop, in this case 10 times, we observe only very small changes after the second and all subsequent repetitions, both at 2 K and at 300 K. The dependence of the exchange bias fields on the repeat number (n) is presented in Fig. 6g. HEB can then be well-fitted using the recursive expression as:30 H EB ðn þ 1Þ H EB ðnÞ ¼ γðH EB ðnÞ H EB ð1ÞÞ3 ;
ð1Þ
where HEB(∞) is the exchange bias field in the limit of infinite loops, and γ is a system-dependent constant. The fitting parameters for 300 K are HEB(∞) = −97 ± 7 Oe and γ = (1.7 ± 0.2) × 10−5 Oe−2 (HFC = +50 kOe), and for 2 K HEB(∞) = 499 ± 6 Oe and γ = (1.9 ± 0.2) × 10−5 Oe−2 (HFC = −50 kOe). Such a large and relatively stable exchange bias effect as seen in the present BFO nanochains is quite rare, being different from previously reported results for single-phase magnetic nanosystems, and may be attractive for device applications. The mechanism of such unusual behavior needs further investigation.
4. Fig. 6 Exchange bias in BFO nanochains: M–H curves of the BFO nanochains measured at (a) 300 K and (b) 2 K in FC mode with HFC = ±50 kOe, and (c, d) the corresponding enlarged views in the low field range. The M–H curves of the BKFO nanobelts (black open squares) are also presented for comparison. Measurements of continuous 10 circle hysteresis loops of the training effect are also shown for (e) 300 K and (f ) 2 K. Panel (g) shows HEB as a function of the repeat number of the hysteresis sweep n obtained at 300 K (HFC = +50 kOe) and at 2 K (HFC = −50 kOe).
This journal is © The Royal Society of Chemistry 2014
Conclusion
In summary, the structural evolution from layered perovskitelike Bi4.2K0.8Fe2O9+δ nanobelts to the 1D BiFeO3 nanochains was successfully performed and studied by a specially designed vacuum annealing method. The ferroelectric properties of the BiFeO3 nanochains and the Bi4.2K0.8Fe2O9+δ nanobelts were established by PFM measurements. Most importantly, a very large and stable exchange bias effect in the as-synthesized BFO nanochains was observed, which may be attractive for magnetoelectric devices.
Nanoscale
View Article Online
Paper
Nanoscale
Published on 14 October 2014. Downloaded by Brigham Young University on 18/11/2014 09:51:04.
Acknowledgements This work is supported by the National Natural Science Foundation of China, the National Basic Research Program of China (contract nos. 2012CB922003 2015CB921201 and 2011CBA00102), the Fundamental Research Funds for the Central Universities (contract no. WK2060140009 and WK2030020026), and the China Postdoctoral Science Foundation (contract no. 2014T70590 and 2013M540513). The authors would like to address special thanks to Prof. J. K. Furdyna (University of Notre Dame) for his kind interest and helpful advice in this work.
14 15
16 17
18
References
19
1 W. Eerenstein, N. D. Mathur and J. F. Scott, Nature, 2006, 442, 759–765. 2 T. Choi, S. Lee, Y. J. Choi, V. Kiryukhin and S. W. Cheong, Science, 2009, 324, 63–66. 3 Y. Tokura and S. Seki, Adv. Mater., 2010, 22, 1554–1565. 4 N. A. Spaldin, S. W. Cheong and R. Ramesh, Phys. Today, 2010, 63, 38–43. 5 X. L. Lu, Y. Kim, S. Goetze, X. G. Li, S. N. Dong, P. Werner, M. Alexe and D. Hesse, Nano Lett., 2011, 11, 3202–3206. 6 M. Bibes, Nat. Mater., 2012, 11, 354–357. 7 J. X. Zhang, R. J. Zeches, Q. He, Y. H. Chu and R. Ramesh, Nanoscale, 2012, 4, 6196–6204. 8 N. Nuraje and K. Su, Nanoscale, 2013, 5, 8752–8780. 9 S. Hong, T. Choi, J. H. Jeon, Y. Kim, H. Lee, H. Y. Joo, I. Hwang, J. S. Kim, S. O. Kang, S. V. Kalinin and B. H. Park, Adv. Mater., 2013, 25, 2339–2343. 10 J. Cao, E. Ertekin, V. Srinivasan, W. Fan, S. Huang, H. Zheng, J. W. L. Yim, D. R. Khanal, D. F. Ogletree, J. C. Grossmanan and J. Wu, Nat. Nanotechnol., 2009, 4, 732–737. 11 D. P. Dutta, O. D. Jayakumar, A. K. Tyagi, K. G. Girija, C. G. S. Pillai and G. Sharma, Nanoscale, 2010, 2, 1149–1154. 12 P. M. Rorvik, T. Grande and M. A. Einarsrud, Adv. Mater., 2011, 23, 4007–4034. 13 N. Lei, T. Devolder, G. Agnus, P. Aubert, L. Daniel, J. V. Kim, W. S. Zhao, T. Trypiniotis, R. P. Cowburn,
20
Nanoscale
21
22 23 24 25 26 27
28
29
30
C. Chappert, D. Ravelosona and P. Lecoeur, Nat. Commun., 2013, 4, 1378. C. Liu, N. P. Dasgupta and P. D. Yang, Chem. Mater., 2014, 26, 415–422. J. Wang, J. B. Neaton, H. Zheng, V. Nagarajan, S. B. Ogale, B. Liu, D. Viehland, V. Vaithyanathan, D. G. Schlom, U. V. Waghmare, N. A. Spaldin, K. M. Rabe, M. Wuttig and R. Ramesh, Science, 2003, 299, 1719–1722. G. Catalan and J. F. Scott, Adv. Mater., 2009, 21, 2463–2485. T. J. Park, G. C. Papaefthymiou, A. J. Viescas, A. R. Moodenbaugh and S. S. Wong, Nano Lett., 2007, 7, 766–772. S. N. Dong, Y. P. Yao, Y. Hou, Y. K. Liu, Y. Tang and X. G. Li, Nanotechnology, 2011, 22, 385701. S. N. Dong, Y. P. Yao, J. Q. Li, Y. J. Song, Y. K. Liu and X. G. Li, Sci. Rep., 2013, 3, 1245. S. N. Dong, Y. K. Liu, S. W. Yang, T. Jiang, Y. W. Yin and X. G. Li, CrystEngComm, 2013, 15, 9057–9063. S. Sepulveda-Guzman, N. Elizondo-Villarreal, D. F. A. Torres-Castro, X. Gao, J. P. Zhou and M. JoseYacaman, Nanotechnology, 2007, 18, 335604. W. D. Pyrz, S. Park, D. A. Blom, D. J. Buttrey and T. Vogt, J. Phys. Chem. C, 2010, 114, 2538–2543. G. D. Achenbac, W. J. James and R. Gerson, J. Am. Ceram. Soc., 1967, 50, 437. A. N. Nesmeianov, Vapor pressure of the chemical elements, Elsevier Pub. Co., 1963. D. R. Lide, CRC Handbook of Chemistry and Physics, Taylor & Francis, 90th edn, 2009. W. H. Ma and D. Hesse, Appl. Phys. Lett., 2004, 84, 2871– 2873. D. P. Dutta, B. P. Mandal, R. Naik, G. Lawes and A. K. Tyagi, J. Phys. Chem. C, 2013, 117, 2382– 2389. D. P. Dutta, B. P. Mandal, M. D. Mukadam, S. M. Yusuf and A. K. Tyagi, Dalton Trans., 2014, 43, 7838– 7846. D. Lebeugle, D. Colson, A. Forget, M. Viret, P. Bonville, J. F. Marucco and S. Fusil, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 76, 024116. C. Binek, Phys. Rev. B: Condens. Matter Mater. Phys., 2004, 70, 014421.
This journal is © The Royal Society of Chemistry 2014
