micro-cubes

Accepted Manuscript The influence of structural size on thermal stability in single crystalline hematite uniform nano/micro-cubes Chao Ruan, Jun Wang, Miao Gao, Guo-meng Zhao PII:

S0254-0584(16)30616-2

DOI:

10.1016/j.matchemphys.2016.08.014

Reference:

MAC 19108

To appear in:

Materials Chemistry and Physics

Received Date: 12 November 2015 Revised Date:

19 June 2016

Accepted Date: 12 August 2016

Please cite this article as: C. Ruan, J. Wang, M. Gao, G.-m. Zhao, The influence of structural size on thermal stability in single crystalline hematite uniform nano/micro-cubes, Materials Chemistry and Physics (2016), doi: 10.1016/j.matchemphys.2016.08.014. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

The influence of Structural Size on Thermal Stability in Single Crystalline Hematite Uniform Nano/Micro-cubes Chao Ruan1, Jun Wang1, Miao Gao1 and Guo-meng Zhao1,2 1 Faculty of Science, Ningbo University, Ningbo, People’s Republic of China

California 90032, USA

RI PT

2 Department of Physics and Astronomy, California State University, Los Angeles,

TE D

M AN U

SC

Abstract We have successfully synthesized monodisperse α-Fe2O3 nano/micro-cubes with sizes ranging from 40 to 840 nm. High-temperature magnetic measurements under a high vacuum were used to systematically study the size dependence of phase transition from α-Fe2O3 to Fe3O4. Through a series of characterizations, we concluded that the different sizes of particles have a significant impact on their thermal stability. The data shows a clear correlation between the onset temperature of the phase transition and the size of particles. This phenomenon can be well explained in terms of the dependence of enthalpy on surface area. Moreover, Morin transition of the α-Fe2O3 phase also have been measured for our products at the nanoscale. The result reflects that the Morin transition temperature decreases with decreasing the size of the particle, which supports finite-size scaling law according to previous findings. Our conclusions provide an important insight into many other size-dependent structural phase transitions. Key words: Hematite nano/micro-cubes; thermal stability; phase transition; Morin transition.

AC C

EP

1. Introduction Recently, with nano-devices decreasing in size, the study of limited quantum size effect and surface effect becomes increasingly important. It is known that, when the size of the magnetic material is close to or smaller than the size of characteristic parameters, the small size effect can affect important physical properties of nanostructures, such as structural phase transition temperature or magnetic ordering temperature. For example, Jiang’s group found that the change of dimension and size of metal and alloy grain can alter the temperature of thermal phase transition [1]. Li et al. investigated the relation between magnetic properties at low temperatures and nanostructures of La6Pb0.4MnO3 with the size of 5 to 100 nm [2]. In our previous work, many of our experiments were focused on the size effect upon magnetic transition temperature [3-5]. Besides, our group studied the thermal stability of α-Fe2O3 nanoring and nanotube, and found it to be closely related to the surface fraction of the (001) plane in the nanostructures [6]. We also found that the different thicknesses of α-Fe2O3 nanoring can affect their thermal stability [7]. Recently, many experimental studies have been conducted to research the size effects of phase transitions [8-14]. 1

ACCEPTED MANUSCRIPT

TE D

M AN U

SC

RI PT

Hematite (α-Fe2O3) nanostructure is one of the most attractive nano-materials and has been widely used in many scientific and technical applications [15-18]. From a fundamental point of view, it is important to research the relationship between the size of the nanostructure and related physical parameters. There have been some studies on the stability of α-Fe2O3 with different morphologies [19, 20]. Most works have focused on the oxidation from Fe3O4 to α-Fe2O3 or fromγ-Fe2O3 to α-Fe2O3, which is found to be associated with size, nanostructure and other factors [21, 22]. However, few studies have been carried out on the phase transition of uniform α-Fe2O3 nanostructures with different sizes to Fe3O4 under a high vacuum (e.g., <9.5×10–6 Torr). These investigations should be given greater attention, since they can provide important conclusions, for example, as to whether the Neel temperature can be accurately measured for the α-Fe2O3 nanoparticles [6, 23]. In our present work, α-Fe2O3 nano/micro-cubes with different sizes have been synthesized by a simple method. We varied the reaction time to gradually change the amorphous rod-like structure to quasi-cubic structure and then to control the size of the grown particles. By this hydrothermal technique, quasi-cubic α-Fe2O3 with high yields and good uniformity can be achieved. The phase stability was studied using high-temperature magnetic measurements in a high vacuum. By comparing with the curve of surface area vs. enthalpy, we show that surface area strongly influences their redox equilibrium and phase stability under high vacuum at high temperature. Moreover, we also measured the Morin temperature for two target products in nanoscale (C4 and C5) which supports the previous conclusion. Our finding provides an important insight into many other size-dependent structural transformations.

AC C

EP

2. Experiment In a typical procedure, 0.2 M FeCl3 with a volume of 40 (C1), 24 (C2), 16 (C3) and 8 ml (C4) were placed into different beakers, respectively. Then distilled water was added until the total volume of the solution in each beaker reached 80 ml. After stirring for 15 minutes, the solutions were transferred into 100 ml autoclaves with Teflon-lined stainless-steel and heated at 170 in a sealed condition for 3 hours without shaking. Then the autoclave was cooled down to room temperature. The red precipitates were collected by centrifugation at 8000 rpm, washed with deionized water and ethanol three times respectively. Finally, the target products were dried in air at 70 [24]. Hematite nanocubes (C5) were synthesized by a hydrothermal route in the presence of polyvinylpyrrolidone (PVP). In the process, 8 mmol FeCl3, 80mmol NaAc and 2.0 g of PVP were dissolved in 60 ml distilled water. After vigorous stirring for half an hour, the solution was transferred into a 100 ml Teflon-lined stainless steel autoclave for hydrothermal treatment and maintained at 200 C for 18 h. After the autoclave was cooled down naturally, the precipitate was collected by centrifugation at 8000 rpm, then washed with distilled water and ethanol, and finally dried at 70 in air [25]. These different sizes of α-Fe2O3 cubes were characterized by X-ray diffraction (XRD, Brucker-D8) spectra, and scanning electron microscopic (SEM, SU-70) 2

ACCEPTED MANUSCRIPT

RI PT

images. Magnetization was investigated using a Quantum Design vibrating sample magnetometer (VSM). The moment measurement was carried out after the sample chamber reached a high vacuum (<9.5×10–6Torr). The heating and cooling rates for the magnetic measurements from 300 K to 920 K are 20 K/min, which are slow enough to have a small negligible thermal lag (less than 3 K). The ZFC and FC measurements were under an applied magnetic field of 500 Oe with heating and cooling rates of 3 K/min.

M AN U

SC

3. Result and Discussion

AC C

EP

TE D

(a) (b) Fig. 1. (a) XRD spectra of the α-Fe2O3 samples labeled through C1 ~ C5, (b) magnifying XRD spectra of the (110) peak. Figure 1 shows x-ray diffraction (XRD) spectra of our α-Fe2O3 particles prepared with increasing amount of FeCl3 (C1~C5). The spectra do not show any sign of other phases, indicating high purity of the samples. All XRD peaks can be indexed to the hexagonal crystal structure. Interestingly, the diffraction (110) peak height increases gradually with the increase of the particle size, relative to the diffraction (104) peak height. This indicates that more fraction of the (110) plane contributes to the X-ray diffraction. Through magnified X-ray diffraction spectra of the (110) peak (Fig. 1(b)), we can clearly see that the reflection peak becomes sharper and splits into two peaks with the increase of particle size. This phenomenon demonstrates the changes of our target products’ crystallinity. It is also found that the variation trend of the full width at half maximum (FWHM) for the (110) peak is gradually increasing with the decrease of particle size. To be specific, the observed FWHMs for C1, C3, and C5 samples are 0.131, 0.167, and 0.236 (rad), respectively. This variation trend is consistent with the Debye-Scherrer equation D =

   θ

, where D is the mean

diameter of particles; k is a content equal to 0.89; λ is the wavelength of X-rays Cu-Kα radiation (λ=0.154nm) ; θ is the diffraction peak angle and B is FWHM (in radian).

3

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

Fig. 2. The SEM images of different samples: (a) C1, (b) C2, (c) C3, (d) C4, (e) C5. The morphology of the samples was analyzed by field emission scanning electron microscopy (FE-SEM, SU70, operated at 5 kV). Figure 2 shows the scanning electron microscopic images of α-Fe2O3 with different sizes. It can be clearly seen that the differently sized α-Fe2O3 structures have a relatively uniform quasi-cubic morphology and homogeneous distribution. It is also confirmed that the particle size increases significantly with the concentration of the precursor material. Figure 3 displays the histograms of sample size determined by FE-SEM. The average diameters of the samples can be estimated through the histograms, which are in accord with those obtained from the XRD results. The detailed data are listed in Table 1.

4

RI PT

ACCEPTED MANUSCRIPT

Fig.3. Histograms of the size distribution of α-Fe2O3 cubes. The solid lines are the best fitted curves by log-normal distribution functions.

SC

Table 1. Average diameters of the α-Fe2O3 samples determined by FE-SEM Sample C1 C2 C3 C4 C5 Size(nm) 840 470 255 100 40

AC C

EP

TE D

M AN U

The synthesized α-Fe2O3 cubes were further characterized by VSM. Figure 4 shows magnetization versus temperature for the α-Fe2O3 particles from 300 to 920 K. The measurement was carried out in a magnetic field of 10 kOe. It can be seen clearly that the magnetization displays a rapid rise above 729 K for C1, and 667 K, 650 K, 621 K, 609 K for C2~C5, respectively. The turning points represent the onset temperature of the phase transition from the weak-ferromagnetic α-Fe2O3 to the ferrimagnetic Fe3O4 phase. Obviously, the transition temperature shows a decrease trend with the decrease of the particle size. Due to the uniformity of the morphology and structure for different α-Fe2O3 particles, this result definitely shows more accurately the impact of particle size on its thermal stability in high vacuum conditions.

Fig. 4. Magnetization versus temperature for the α-Fe2O3 particles from 300 K to 920 K. Figure 5(a) shows transition temperature versus particle size for different 5

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

samples. It’s intriguing to find that there is a linear relationship between particle size and transition temperature. Figure 5(b) shows the enthalpy of anhydrous γ-Al2O3 versus surface area [26]. Enthalpy as a state function, represents the system of thermodynamics (internal energy). Different enthalpy value reflects the characteristics of thermodynamic stability. So the higher enthalpy value means that the greater the internal energy of system, and the more prone to occur phase transition. From figure 5(b), with increasing specific surface area, the enthalpy increases linearly that reflects the material thermal stability with decreasing particle size. So this previous result is consistent with our experimental data, which prove that enthalpy strongly influences phase stability. All of these evidences provides deeper insight into understanding of the change of phase stability by thermodynamics.

AC C

EP

TE D

a b Fig.5. (a) Phase transformational temperature versus particle size for different samples and (b) enthalpy versus mean surface area of alumina particles (the data points were taken from Ref.26) Moreover, different structural sizes of particles not only influence their phase stability, but also have impact on their redox equilibrium between α-Fe2O3 and Fe3O4. From figure 4, we can also get the maximal magnetization of C1~C5 which is 1.74 emu/g at 825 K, 5.31 emu/g at 817 K, 6.87 emu/g at 792 K, 15.32 emu/g at 764 K and 41.10 emu/g at 718 K, respectively. These values of the maximum magnetization are related to the ratios of the phase transition from α-Fe2O3 to Fe3O4 in our different target products, and further verify the relationship between particle size and redox equilibrium on the other hand. To analysis the specific ratios of phase conversion transition from α-Fe2O3 to Fe3O4, we take the XRD measurement for samples which were cooled down from 920 K. Figure 6 shows the XRD spectra. By measurement of reference intensity ratios (RIR):

% =

 

    

× %

we can approximately quantify the ratios of phase transition. And in this formula, Cx represents the mass fraction of phase x; Ix represents the intensity of phase x, which can be calculated by the area of the strongest peak. And Kx represents the RIR value of the phase x, which can get from ICSD subfile in Jade.

6

SC

RI PT

ACCEPTED MANUSCRIPT

M AN U

Fig. 6. XRD spectra of the α-Fe2O3 samples (cooled down from 920 K) Table 2 Ratios of phase transition from α-Fe2O3 to Fe3O4 in given samples which were cooled down from 920 K Fe2O3 % Fe3O4 % Component Sample

AC C

EP

TE D

C1 ~81 ~19 C2 ~78 ~22 C3 ~60 ~40 C4 ~53 ~47 For samples C1~C4, the peaks are impure which match with mixed α-Fe2O3 and Fe3O4. Through calculating the area of the strongest peak, specific ratios of phase transition were listed in Table 2. The results illustrate that since the decrease of average size, the percentage of residual α-Fe2O3 is gradually reduced from 81.45% to 52.76%. And for sample C5, the peaks are well matched with Fe3O4, indicating that almost all α-Fe2O3 has been converted into Fe3O4 completely.

(a) (b) Fig. 7 (a) Magnetization versus magnetic field for the α-Fe2O3 nanoparticles before 7

ACCEPTED MANUSCRIPT

RI PT

heating. (b) Magnetization versus magnetic field for the α-Fe2O3 nanoparticles after heating. Figure 7 (a) shows the magnetic hysteresis loops of α-Fe2O3 cubes for different samples (C1, C3, C5) from -3T to 3T at 300K before they were heated to 920 K. It exhibits a weak ferromagnetic behavior between the Morin temperature and Neel temperature [19, 27]. The magnetization doesn’t saturate even in the high magnetic fields. So in this region their magnetization curve is described by M=MS+λH. The coercivity (HC), remanence magnetization (MR), saturation magnetization (MS), high-field susceptibility (λ) for samples C1, C3 and C5 were listed in Table 3.

AC C

EP

TE D

M AN U

SC

Table 3. The coercivity (HC), remanence magnetization (MR), saturation magnetization (MS), high-field susceptibility (λ) in given sample (C1, C3 and C5). Parameter HC(Oe) MR(emu/g) MS(emu/g) λ(emu/g Oe) Sample -1 -1 C1 2990 3.24×10 1.32×10-5 2.74×10 C3 2330 1.83×10-1 2.67×10-1 1.90×10-5 C5 520 0.78×10-1 2.83×10-1 1.76×10-5 From Table 3, we can find that sample C1 displays the largest coercivity and remanence magnetization. It can be attributed to its enhanced shape and reduced crystal defects [28-30]. Samples C3 and C5 show less coercivity and remanence magnetization due to their smaller size and lower crystallinity. Moreover, the area of hysteresis loop decreases with the decrease of particle size. It indicates that smaller α-Fe2O3 structure size can reduce the energy loss effectively in the process of positive and reverse magnetization repeatedly. Figure 7 (b) shows the magnetic hysteresis loops of α-Fe2O3 cubes after they were cooled down from 920 K. The saturation magnetizations of the samples can be used to determine the fractions of the transformed Fe3O4 phase. From Figure 7 (b), we found the saturation magnetization enhance with decreasing particle size. And when the particle size was reduced to 40 nm, the saturation magnetization is about 69 emu/g, which is close to that expected for pure Fe3O4 nanoparticles. These data also verify the measurement of RIR by XRD and confirm the specific ratios of phase conversion.

(a) (b) Fig.8. The zero-field-cooled and field-cooled measurements at magnetic fields of 500 Oe from 5 K to 300 K for sample C4 (a) and C5 (b). Inset shows the dM/dT versus T plot, giving the positions of the Morin transition temperature. In addition to phase transition, Morin transition also have been researched in our 8

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

paper. Herein, size dependence of the Morin transition was studied for sample C4 and C5. Because of their nanometer dimensions, more discussion about this issue is valuable and necessary. The obtained field cooled and zero-field-cooled magnetization versus temperature curves under an external magnetic field of 500 Oe from 5 to 300 K are shown in Fig. 8. As can be seen from Fig. 8, the Morin transition temperature (TM) which is derived from the maximum of the dM/dT curve, shifted from 252 K to 208 K as the sizes of nanocubes reducing from 100 nm to 40 nm. Both of these two values are lower than the Morin temperature of bulk hematite (263 K) [31]. As we all know, there are some factors can affect the Morin transition temperature, such as crystalline anisotropy, lattice strains and crystal defects generated by different morphologies and nanostructure size [20, 32]. In our experiment, since the morphology and anisotropy were similar among different samples, the change in the Morin transition temperature can qualitatively be attributed to the nanostructure size and crystal defects. So we can conclude that increasing nanostructure size and reducing the crystal defects can enhance its Morin temperature.

EP

TE D

(a) (b) Fig.9. Magnetization versus magnetic field for C4 (a) and C5 (b) at 5 K Hysteresis loops at 5 K after zero-field-cooling for samples C4 and C5 are shown in Fig.9. From the character of low magnetic susceptibility, typical antiferromagnetic behavior was observed below TM.

AC C

4. Conclusion In summary, we showed the stability of α-Fe2O3 cubes with different sizes under high vacuum. The results revealed a strong size-dependent phase transition from weak-ferromagnetic α-Fe2O3 to ferrimagnetic Fe3O4. The effect of the surface area and related enthalpy can explain this phenomenon well, and the linear relationship between surface area and enthalpy displays a similar linear fitting compared with our experimental result. Then we further characterized the magnetic properties of α-Fe2O3 cubes with different diameters. The observed changes of the coercivity, remanence magnetization and the area of the hysteresis loop can be attributed to its enhanced shape and reduced crystal defects. All these results indicate that the phase transformation from α-Fe2O3 to Fe3O4 depends strongly on the size of the structures. Finally, the Morin temperature was measured at 500Oe for our samples in nanoscale, which monotonically increases with the size of nanoparticles.

9

ACCEPTED MANUSCRIPT

Acknowledgments This work was supported by the National Natural Science Foundation of China (11474174), the Natural Science Foundation of Zhejiang (LY14E020002) and Ningbo

RI PT

(2015A610001, xkzwl1611), the K. C. Wong Magna Foundation. Reference:

[1] X.B. Jiang, M. Jiang, M. Zhao, Physica B. 406 (2011) 4544.

[2] T. Zhang, G. Li, T. Qian, J.F. Qu, X.G. Li et al., J. Appl. Phys. 100 (2006)

SC

094324.

[3] J. Wang, W. Wu, F. Zhao, and G.M. Zhao, Applied Physics Letters. 98 (2011)

M AN U

083107.

[4] J. Wang, F. Zhao, W. Wu, and G.M. Zhao, Journal of Applied Physics. 110 (2011) 123909.

[5] J. Wang, W. Wu, F. Zhao, and G.M. Zhao, Physical review B. 84 (2011) 174440.

TE D

[6] L. Li, F.G. Li, J. Wang, G.M. Zhao, J. Appl. Phys. 116 (2014) 174301. [7] L. Li, F.G. Li, X.P. Zhang and J. Wang, NANO: Brief Reports and Reviews. 10 (2015) 1550084.

[8] J. Lee, S.G. Kwon, J.G. Park and T. Hyeon, Nano Lett. 15 (2015) 4337.

EP

[9] M. Li, E.I. Altman, Surface Science. 619 (2014) L6. [10] R.A. Mondal, B.S. Murty, V.R.K. Murthy, Physica B. 461 (2015) 10.

AC C

[11] C.B. Zhu, L. Gu et al., Adv. Funct. Mater. 24 (2014) 312. [12] X.Z. Yan, X.T. Ren et al., J. Appl. Phys. 116 (2014) 033507. [13] S.X. Zhang, I.S. Kim, and L.J. Lauhon, Nano Lett. 11 (2011) 1443. [14] E.C. Spencer, N.L. Ross et al, J. Phys. Chem. C. 119 (2015) 9609. [15] D.K. Zhong, J.W. Sun, H. Inumaru, D.R. Gamelin, J. Am. Chem. Soc. 131

(2009) 6086. [16] Z.Y. Sun, H.Q. Yuan, Z.M. Liu et al., Adv. Mater. 17 (2005) 2993. [17] L. Zhang, H.B. Wu et al., Adv. Energy Mater. 4 (2014) 1300958. [18] H.C. Yu, Y.Y. Li et al., Dalton Trans. 41 (2012) 723. 10

ACCEPTED MANUSCRIPT [19] N.M.A. Hadia, S. García-Granda, J.R. García, Materials Chemistry and Physics. 147 (2014) 1037. [20] S. Mitra, S. Das et al., Journal of Magnetism and Magnetic Materials. 321 (2009) 2925.

RI PT

[21] A. Navrotsky, C.C. Ma, K. Lilova, N. Birkner, SCIENCE. 330 (2010) 8. [22] I.V. Chernyshova, M.F.Jr. Hochella, A.S. Madden, Phys. Chem. Chem. Phys. 9 (2007) 1736.

[23] J. Wang, W. Wu, F. Zhao, G.M. Zhao, J. Appl. Phys. 109 (2011) 056101.

SC

[24] S. He, G.S. Wang, J.W. Wang, ChemPlusChem. 78 (2013) 875.

[25] M.Y. Zhu, Y Wang, D.H. Meng, J. Phys. Chem. C. 116 (2012) 16276.

M AN U

[26] A. Navrotsky, Geochem. Trans. 4(6) (2003) 34.

[27] H.M. Lu, X.K. Meng, J. Phys. Chem. C. 114 (2010) 21291. [28] X.G. Wen, S.H. Wang, Y. Ding, Z.L. Wang, S.H. Yang, J. Phys. Chem. B. 109 (2005) 215.

[29] L.M. Bronstein, X.L. Huang, J. Retrum, A. Schmucker, M. Pink, B.D. Stein,

TE D

B. Dragnea, Chem. Mater. 19 (2007) 3624.

[30] Y. Yang, H.X. Ma, J. Zhuang, X. Wang, Inorg. Chem. 50 (2011) 10143. [31] F.J. Morin, Phys. Rev. 83 (1951) 1005.

AC C

EP

[32] J. Wang, A. Victor, L. Li, et al., Nano Research. 6 (2015) 1906.

11

ACCEPTED MANUSCRIPT

Highlights 1. Hematite nano/micro-cubes were fabricated and well controlled in uniform

AC C

EP

TE D

M AN U

SC

RI PT

morphology. 2. Size dependence of phase transition has been systematically studied. 3. Enthalpy strongly influences phase stability. 4. Morin transition temperature decreases with decreasing the size of the particle.