II. General Backgrounds
3.1 Solid State Reaction
In this research, solid state reaction was used to make BFO and CFO bulk target and these bulk target ware used to make BFO-CFO thin film. To make BFO and CFO target from BFO and CFO powder by solid state reaction, precise calculation of chemical reaction of each basic material should be done. sintered at 600oC for 4h thus continued by 880oC for 480 second. Do grinding process again, thus sample ware pressed uniaxially (500 Mpa) into pellets and annealed in furnace at 600oC for 12 hours. For CFO, calculation was done similarly as that used to get BiFeO3 powder. However, in this case CoO and Fe2O3 powder ware used to prepare CoFe2O4 powder, the powder was dried and
24
ground, then calcined at 1000oC for 12 hours. Finally, it was pressed into pellets and annealed at 1000oC for 12 hours.
The XRD result shown in Fig. 17 (a) and 17 (b) evidently confirmed that the target obtained by the processes describe above were indeed of single phase materials with correct stoichiometry.
Figure17. BiFeO3 and CoFe2O4 XRD pattern.
It is noted that heat treatment process is very important in this experiment.
Although the temperatures were chosen according to the phase diagram and
25
parameters reported in the literature, we note that in order to get good quality samples, environmental conditions should be considered as well.
3.2 Pulse Laser Depositition
The set up of the PLD system is schematically illustrated in Fig. 18.
Briefly, it contains a laser generation system and a vacuum chamber with substrate and target holders.
Figure18. Schematic of PLD experiment set up.
The vacuum chamber should be able to reach a base preseure of 10-7 mT which is acheive by sequential pumping using scroll pump and turbo pump. A gauges and oksigen pipe were equiped to control pressure inside the chamber. In this study, the BFO-CFO film on STO substrate were deposited at an oxigen pressure 50 mT to 200 mT depending on what surface desired and how thick the film will be both substrate temperature and deposition time were further optimazed. The chamber also has heater and thermocouple to control the temperature used to make samples.
26
In further controls on the deposition process were accomplished by tuning the laser pulse energy, repetition rate direction and focus of the laser beam with lenses direct to chamber. Beside the main conditions in PLD system, we can also control the rotation of target and the distance between substrate and target to optimize the conditions for making good quality samples. Distance between target-substrate will affect the incident adatom from target coming to substrate.
If closer distance was predicted the incident adatoms are more abundant than that obtained if larger target-substrate distance was used within the same deposition time is rich than far distance in the same deposition time.
For BFO-CFO vertically alligned nanocomposite fabrication, two targets were used simultaneously. There are several ways to make BFO-CFO vertically alligned nancomposite. The popular way is to use single target made BCFO mixtures. BCFO target is a combination of BFO powder and CFO powder, where there is no chemical reaction between mixed BFO and CFO. Such that, BFO and CFO particles will be separated during deposition process and self assembled into nanocomposite. Howerver, in order to gain more control on the microstructure, in this reseach we used two targets to make BFO-CFO nanocomposites.
The deposition time for each target is important in fabrication of BFO-CFO vertically alligned nanocomposite by using two targets. In this case,we need to know how much deposition time is needed to build single three-dimensional islands first. Because, further growth of island will result in nanopillars embedded in matrix until nanocomposite pattern is accomplished. For BFO-CFO on STO (001), BFO-CFO would be pillar, so in that case we need to consider the duration of deposition time to make the first single CFO islands. After CFO was deposited, the BFO deposited on the STO (001) substrate will form a continous matrix because of the smaller lattice mismatch between them.
27
Fig. 19. Deposition step and time duration selected for forming BFO-CFO vertically aligned nanocomposite.
In this case, a 5-15 second deposition time combination is appropriate with total a deposition time of 30 minute. It was found that if the deposition time of CFO step is longer than 15 second, pillars cannot be constructed and film tend to collapse into layer. The 30 minutes total deposition time correspond to 180 time cycle for each target if 10 second for each target is used.
3.3 Characterization
The microstructure and physical properties of the obtained samples were characterized by XRD, scanning electron microscope (SEM), superconducting quantum interference device (SQUID), and impedance analyzer or (LCR meter), respectively.
3.3.1 X-ray Diffraction (XRD)
XRD characterization is due essential to identify the phases formed in the obtained films. The Bragg’s law gives the angles for coherent and incoherent scattering from a crystal lattice, which describes the relations between the wavelength (λ),diffraction angles (θ), and the lattice spacing between certain planes of the crystalline phases inside the film by the following relation :
2𝑑 𝑠𝑖𝑛𝜃 = 𝑛 𝜆 (4)
28
Figure 20. Schematic of the Bragg’s law.
Thus, from the diffraction peak, one can confirm the crystal structure of the material in study. In this research, XRD also used to investigate strain effect in the sample. BFO, CFO and STO have different lattice constants, resulting in strain effect among them when they are in proximity. Especially in the case of BFO-CFO on STO VAN, the BFO has a compressive in-plane strain due to the lattice mismatch with STO substrate. In addition, BFO also has a compressive strain along out-of plane direction due to the lattice mismatch with the CFO.
Strain effect in BFO due to lattice mismatch with STO and CFO makes the XRD peaks of BFO in BFO-CFO/STO VAN thin film shift as compared to that of the BFO powder.
Lattice constant of BFO was change because those strain effect which make X-ray diffraction also change. According to Bragg’s law, angle of reflection beam (θ) will change if spacing between diffracting planes (d) was changed. where d is also indicated as a lattice constant in the sample. So, when lattice constant of BFO was changed to higher or lower value caused lattice mismatch among BFO, CFO and STO, so angle of reflection beam (θ) also will be changed which will be evidence by shift of the BFO peak in diffraction pattern of XRD result.
29
3.3.2 Scannng Electron Microscope (SEM)
SEM was used to identify sample microstructure. In this research SEM was also used to reveal both the surface and cross sectional structures, especially for BFO-CFO on STO VAN to indicate nanocomposite structure. SEM is a type of electron microscope to take an images from the sample surface by scanning it with a high-energy beam of electrons. In SEM, electrons are emitted from a tungsten filament cathode and accelerated towards an anode. There are lenses to focus the beam to have spot radius from 0.4 µm to 5 µm. The electron beam will interact with the sample and its interaction volume can extend from less than 100 nm to around 5 μm into the surface. Pairs of scanning coils deflect the beam horizontally and vertically make a rectangular area scan in the sample surface.
Figure 21. Schematic illustration of a SEM (rotated by 90o counter-clockwise) [28].
Imaging can be achieved through the detection of low energy (< 50 eV) secondary electrons, or backscattered electrons. The latter consist of high-energy electrons originating in the electron beam, that are reflected or back-scattered out of the specimen interaction volume. Backback-scattered electron imaging is useful for distinguishing one material from another, since the yield of the collected backscattered electrons increases monotonically with the specimens atomic number. Backscatter imaging can distinguish elements with atomic number differences of at least three, i.e., materials with atomic number
30
differences of at least three would appear with good contrast on the image.
Because these electrons are emitted from a depth in the sample, the resolution in the image is not as good as that obtained from the secondary electrons, with a beam resolution ranging from 10 to 20 nm [28].
3.3.3 Atomic Force Microscopy (AFM)
AFM is a scanning probe microscope tool to image and measure a material in nanoscale. In AFM consist of a microscale cantilever with an atomically sharp tip, used to scan material surface. The tip could produce a Coulomb repulsive force from deflection of the cantilever according to Hooke’s law.
Then, it will effect to laser spot reflecting in photodiodes. To maintain a constant tip-to-sample distance, a feedback loop is implemented between the photodetector and the cantilever. The cantilever is maintained fixed in a tip holder, and the sample placed on top of a piezoelectric column that can move in the z direction for height adjusting and in the x, y directions for surface scanning [28].
Figure 22. Schematic of an AFM.
Two scanning modes, namely contact mode and resonace mode are commonly used in AFM operation. In the contact mode, the force between the tip and the sample is kept constant by the feedback loop by maintaining a
31
constant cantilever deflection. While in resonance mode, the cantilever is driven to oscillate near its resonance frequency by a small piezoelectric element mounted in the tip holder; the oscillation amplitude is modified as a result of tip-sample forces, so that the reflected laser beam is deflected in a regular pattern over the photodiode array, generating a sinusoidal electronic signal which is modified by the oscillation amplitude variation.
Depending on the height of the tip above the sample, different forces such as electrostatic interaction, magnetic forces, Van Der Waals attraction, water adhesion and Coulomb repulsion can play a role. If the cantilever enter intermittently into contact with the surface, the technique is known as tapping mode; otherwise, it is referred to as non-contact mode.
3.3.4 Superconducting Quantum Interference Device (SQUID)
SQUID has been widely used to measure magnetic properties of material.
The SQUID devices is due to the Josephson effect exhibiting in two parallel Josephson junctions consisting of two superconductors separated by a thin insulating layer. The device allows a change of magnetic field associated by one flux quantum to be measured. The flux which associated by Josephson junctions is quantized in units of flux quantum ϕo=2.07x10-15 T.m2.
Figure 23. Schematic illustration of a SQUID [28].
32
SQUID is operated with a constant bias current, thus the measured voltage oscillates with the phase changes between the two junctions, which depends upon the change in the magnetic flux, as illustrated in Figure 23. Detecting this circulating current enables the use of the SQUID as a magnetometer.
3.3.5 Impedance analyzer (LCR meter)
LCR meter is an equipment to measure the inductance (L), capacitance (C) and resistance (R) in material. In this research, LCR meter was used to measure and analyze electric properties of BFO-CFO on STO substrate especially capacitance and resistance dependence on temperature. For capacitance measurement, two probe is enough to get capacitance value of the sample. DC volatge drop will build across the sample, thus temperature dependent capacitance of the sample could be obtained. For resistance measurements, both two probe and four probe methode were used.
33
IV. Results and Discussion
4.1 Structure Analysis
In this research, three kinds of BFO-CFO vertically align nanocomposite were fabricated according to substrate direction. As shown in the Fig. 24, the temperature suitable fo growing the BFO-CFO nanocomposite film are 600oC, 650oC, and 700oC. Below 600oC, BFO growth was not good, i.e. only amorphous film were obtained at T=550oC (Fig. 24). On the other hand, for temperature higher than 700oC, bismuth was very easy to evaporate, making the stochiometry of the sample unstable and may even leading the formation of impurity phases, such as Fe2O3 and Bi2O3.
Figure 24. XRD pattern of BFO thin film on STO (001) substrate with different temperature growth conditions.
Deposition condition, such as pressure, repetition rate, and pulse energy are also dependent on the temperature used. In general, with the same growth
34
pressure. The thickness of BFO also depend on repititon rate. In present experiment, pressure at around 100 mT-250 mT and repitition rate at 5 Hz-10 Hz were found appropriate to grow BFO on STO. The pulse energy, not only will affect to the final film thickness, but also the formation of the impurity phase. Higher energy densities often caused more BFO to decomposed. In this reseach pulse energy for growing BFO on STO was optimized to within 250 mJ-350 mJ.
In this research we used separated BFO and CFO target to fabricate BFO-CFO thin films, consequently, the condition for growth of individual phase have to be established first. Generally, CFO needs higher growth temperature than BFO, because CFO and STO have different crystal structures. In contrast, BFO is having the structure crystal same with STO. Thus, BFO is easier to growth on STO. moreover, lattice mismatch between BFO-STO is smaller than that of CFO-STO, which also plays an important role in searching for optimal growth condition.
Figure 25. XRD pattern of CFO thin film on STO (001) substrate with different growth temperatures.
20 30 40 50 60 70 80
650 C 700 C 750 C
2 theta ( )
Intensity (Arb. Units)
CFO(400)
STO(001) STO(002)
STO(003)
35
As is evident from Fig. 25, single phase CFO can be grown on STO at temperature of 650 oC, 700 oC, and 750 oC, with pressure condition around 100 mT-200 mT, repetition rate of 5 Hz-10 Hz and pulse energy at 250 mJ-350 mJ, respectively, these conditions are almost the same as those needed for depositing BFO on STO with our PLD system. Impurity in CFO was relatively difficult to appear, because CFO is more stable and needs higher temperature to decompose into CoO and Fe2O3. Therefore, almost the same conditions growth were used to make the BFO-CFO composite film on STO, even with two separate target.
Since the wetting condition for BFO and CFO are different, when BFO and CFO are growing on STO substrate at the same time, they will self essemble to form composite epitaxial thin films. Fig. 26 displays a collections of XRD result for BFO-CFO/STO (001), (011), (111) VAN, BFO/CFO/STO (001), (011), (111) bilayer, BFO/STO (001), (011), (111) single layer, BFO powder and CFO powder, respectively.
36
Figure 26. XRD from BFO-CFO/STO VAN, BFO/CFO/STO bilayer, BFO/STO single layer, BFO and CFO powder grown on STO with different orientations. a. (001), b. (011), c. (111).
From the XRD results, it is evident that the diffraction peaks of BFO and CFO films are slightly shifted as compared to that of data base as well as BFO and CFO powders. The shifts indicated that there are substantial strain effects
37
from the substrate. Even though the lattice mismatch between BFO and STO is small. By using the Bragg equation, lattice paramaters and strain effects were calculated and summarized in Table 2 and 3.
(5) (6)
(7) (8)
Table 2. Strain effect from sample on STO (001).
Sample Film Lattice Parameter Strain 010
Table 3. Lattice mismatch from sample on STO (001).
Material Å Lattice mismatch
STO 3.91 STO-BFO=~1.3%
BFO 3.96 STO-CFO=~7.3%
CFO 8.39 BFO-CFO=~5.9%
Because STO has smaller lattice parameter than that of BFO, leading to a slight in-plane compressive strain on BFO film and a slightly larger lattice
38
parameter along the out-of plane orientation. As a result, the XRD peak shift to smaller 2θ degree. Shifts of XRD peak in BFO/STO singel layer and BFO/CFO/STO bilayer are larger than that in BFO-CFO/STO VAN. This is indicative that in BFO-CFO/STO VAN there was relaxation effect on BFO film from CFO film in the vertical side. This can be further visualized from the schematic illustration shown in Fig. 27.
Figure 27. Schematic illustration of strain and relaxation effect in BFO-CFO on STO substrate.(a.) BFO/STO single layer, (b.) BFO/BFO-CFO/STO bilayer, (c.) BFO-CFO/STO VAN.
Fig. 28, illustrates the nucleation and growth procedure for individual phases on STO with different orientations, which indicate the expected grain and surface morphologies for films and nanocomposites. Thus from Fig.29 swohs the surface structure analysis obtained by AFM measurements for single layer BFO/STO, bilayer BFO/CFO/STO and BFO-CFO/STO VAN grown on different orientations of STO substrate. From the AFM result, it can be seen that surface of CFO films is rougher that that of BFO film. This might be because CFO has different crystal structure with BFO and STO. The surface morphology of the single layer and bilayer samples appear to vary with the orientation of the STO substrates, which can be explained by the illustrations shown in Fig. 28.
39
Figure 28. CoFe2O4 (left) and BiFeO3 (right) nucleating on (a.) (001), (b.) (111), and (c.) (110) SrTiO3 surfaces [16].
For BFO-CFO/STO nanocomposite, it is obvious that there are nanopillars embedded in matrix. Depending on the orientation of the STO substrate, the shape of pillars is different. On STO (001), the embedded CFO pillar appears to be rectangular. It becomes like maze pattern on STO (011) and turns into triangular when STO (011) substrate was used. That was happen because as has been discussed in previous chapter, this is due premarily to the different wetting condition on different STO substrates direction.
Figure 29. AFM result : (from left to right) a, b, c. Single layer on STO (001), (011), (111), d, e, f. Bilayer on STO (001), (011), (111) g, h, i.
Vertically aligned nanocomposite on STO (001), (011), (111).
g. h. i.
a. b. c.
d. e. f.
40
To confirm the film surface morphologies, SEM was also used to observe nanopillar of BFO-CFO/STO VAN grown on STO substrates with different direction. As can be seen from the result shown in Fig. 30, essentially same result are observed as those revealed by AFM measurement.
Figure 30. SEM result from : BFO-CFO/STO vertical allign nanocomposite. a. (001), b. (011), c. (111).
One of primary goals of this reseach is also try to manipulate pillar density in the sample and hence the coupling between the two phases. Fig. 31 demonstrate that this can be done by varying the respective deposition time of BFO and CFO in each cycles.
Figure 31. SEM result from BFO-CFO/STO (001) vertical allign nanocomposite with variation of CFO pillar density.
The left photograph in Fig. 31 has the lowest pillar density because the deposition time are 15 second for BFO and 5 second for CFO respectively, with a total deposition time of 30 minutes, each target has experienced 180 cycles during the process. In the middle photograph, both BFO and CFO targets was deposited by using a 10/10 second cycle. Finally the right photograph is obtained by using 5 second deposition time for BFO and 15 second for CFO.
a b c
15:5 10:10 5:15
41
Since on STO (001) substrate, CFO is the pillar, so if we want to reach higher pillar density, deposition time of CFO should be higher than BFO and vice versa.
4.2 Properties Analisys
As mentioned previously, the Raman measurement showed clear phase transitions at ~140 K and ~200 K, although the exact origin of which is as yet unclear, it has been tentatively attributed to spin reorientations [20]. Thus, this study has focused more on the magnetic and electric properties at lower temperatures in particular, combination BFO and CFO is expected to result in extra coupling between them and the anomaly occured in BFO at low room temperature may be further modified.
Figure 32. M-T measurement for BFO/STO single layer on STO a. (001), b. (011), c. (111), d. From refrences [8] BFO single crystal powder.
0 50 100 150 200 250 300
42
Fig. 32 shows the temperature-dependent magnetization (M-T) obtained for single layer BFO film grown on various STO substrates. Similar to that observed in BFO single crystal powder (Fig. 32 d) the M-T beharviours of
Fig. 32 shows the temperature-dependent magnetization (M-T) obtained for single layer BFO film grown on various STO substrates. Similar to that observed in BFO single crystal powder (Fig. 32 d) the M-T beharviours of