Fig. 4.11 and Fig. 4.12 show the magnetization of Ni/O-3x3/W(111) and Co/O-3x3/W(111). From Fig. 4.11, Ni/O-3x3/W(111) reveals in plane magnetization and weak perpendicular magnetization. It means that Ni/O-3x3/W(111) is canted magnetization. As shown in Fig. 4.12, Co/O-Ni/O-3x3/W(111) performs in plane magnetization.
Figure 4.11: Hysteresis loop of Ni/O-3x3/W(111)
Figure 4.12: Hysteresis loop of Co/O-3x3/W(111)
Figure 4.13: (a) Hysteresis loop of 9.61 PML Co/O-3x3/W(111) in differ-ent angle. (b) Coercivity of 9.61 PML Co/O-3x3/W(111) as a function of rotating angle.
Fig. 4.13 and Fig. 4.14 show different angle hysteresis loop of 9.61 PML and 6 PML 3x3/W(111) respectively. Coercivity of 9.61 PML Co/O-3x3/W(111) as a function of rotating angle, as shown in Fig. 4.13(b). The largest coercivity point appears repeatedly after 60 degree rotating. It means that Co/O-3x3/W(111) reveals six-fold symmetry of magnetization. The 6 PML Co/O-3x3/W(111) shows similar behavior in magnetism, as shown in Fig. 4.14. The 6 PML Co/O-3x3/W(111) also reveals six-fold symmetry of magnetization.
Figure 4.14: (a) Hysteresis loop of 6.0 PML Co/O-3x3/W(111) in different angle. (b) Coercivity of 6.0 PML Co/O-3x3/W(111) as a function of rotating angle.
Chapter 5 Discussion
Q1: Why Ni performs the island growth and Co performs the layerwise growth on O-3x3/W(111)?
In chapter 4.2, the result reveals that Ni performs the island growth on O-3x3/W(111) and Co performs the layerwise growth on O-O-3x3/W(111), why?
It may be due to the lattice mismatch. As shown in Fig. 4.4, the lattice mis-match of Ni/W(111) and Co/W(111) are 3.95% and 2.76% respectively. The larger lattice mismatch of Ni/W(111) results in the island growth, and the suitable lattice constant of Co causes the layerwise growth for Co/W(111).
There may exist other reasons to affect growth mode, but we need more experimental data to demonstrate.
Q2: How does the growth mode affect the magnetic behavior?
The different growth mode of Ni and Co film causes different magnetic behavior. From Fig. 4.11 and Fig. 4.12, the Ni/O-3x3/W(111) reveals the canted magnetization, and Co/O-3x3/W(111) reveals the in plane magne-tization. Owing to the island growth of Ni/O-3x3/W(111), the magnetic moment of Ni can be separate into two parts, in plane magnetic moment and perpendicular magnetic moment, and it results in the canted magnetization.
However, for Co/O-3x3/W(111), it reveals layerwise growth, and the direc-tion of the magnetic moment is in plane direcdirec-tion which causes the in plane magnetization. The six-fold symmetry in magnetism of Co is also due to the layer growth because the Co films grows follow the W(111) substrate. Hence, the Co films has the same symmetry as W(111) six-fold symmetry.
Q3: Why Co and Ni are different in thermal stability of wetting layer?
About the thermal stability, a wetting layer is an initial layer of atoms that is epitaxially grown on a surface. The thickness of wetting layer is due to the lattice mismatch in theoretically. Co/O-3x3/W(111) exists 13 PML
wetting layer result from a smaller lattice mismatch 2.76 %, and the wetting layer does not exist for Ni/O-3x3/W(111) because of a larger lattice mismatch 3.95 %.
Q4: How do you demonstrate the exist of surfactant effect of oxygen?
As mentioned in chapter 4.3, O(510 eV) Auger signal always keep invari-ant after deposition, even the thickness increases over 10 PML. The similar behavior is observed after annealing, as shown in the result of thermal sta-bility. These results implies that the oxygen atom is always on the top of surface after deposition and annealing, as a surfactant. The surfactant effect of oxygen in other system had been demonstrated in previous literature[16-21]. These literatures indicates that the oxygen gas adsorbs on surface and the oxygen atom should float on the surface in order to decrease the surface energy.
Q5: Why the post-annealing is necessary?
The oxygen molecule adsorbs on the W(111) surface, and then it breaks up into atoms. With the oxygen exposing time increasing, the AES signal of O(510 eV)/W(170 eV) increases quickly and finally saturates about a maximum of 0.5. It implies that the quantity of oxygen is limited on the W(111) surface. Moreover, we also do the experiment about annealing times.
The results show that the AES signal of O(510 eV)/W(170 eV) decreases with annealing times, and it reaches a minimum of about 13 as the O-3x3 surface reconstructed. It means that the oxygen is too much to form the 3x3 surface after oxygen exposing. Hence, we must reduce the quantity of oxygen by post-annealing. Therefore, a well-ordered O-3x3 surface is formed after post-annealing with proper temperature.
Q6: Does the change of AES signal ratio of Co/W and Ni/W result from alloy effect?
From the phase diagram of Co-W and Ni-W, as shown in Fig. 5.1, it shows that no alloy forms when the percentage of Ni and Co is lower than 10
%. Hence, we can conclude that no alloy effect occurs in Co/O-3x3/W(111) and Ni/O-3x3/W(111).
Figure 5.1: The phase diagram of Co-W[28].
Figure 5.2: The phase diagram of Ni-W[29].
Chapter 6 Summary
In previous literature, the oxygen induced facetting on W(111), and we are interesting in the magnetism of magnetic material deposited on facetting surface which induced by oxygen. From our results, a 3x3 reconstruction was formed after oxygen exposing and annealing with proper temperature.
Without the two-step of annealing, no 3x3 reconstruction formed.
In the other reports, Ni deposited on a clean W(111) crystal reveals Vollmer-Weber growth mode, and Co deposited on a clean W(111) was layer by layer growth. In contrast, for Ni/O-3x3/W(111), AES signal of tung-sten was always observable, implying that Ni undergoes island-growth on O-3x3/W(111). For Co/O-3x3/W(111), Co undergoes layerwise growth on O-3x3/W(111).
About the thermal stability, the wetting layer formed after about 700 K annealing in the case of Co/W(111) and Ni/W(111). However, our experi-mental result reveals that the wetting layer is not formed after annealing, even after large amount of deposition in the case of Ni/O-3x3/W(111). Oppositely, the smaller wetting layer was observed after annealing in Co/O-3x3/W(111).
Our experimental result also indicates that the Co and Ni films which grew on O-3x3/W(111) formed 3D islands after annealing process. The AES sig-nal as a function with annealing temperature. As we see, no wetting was formed in Ni/O-3x3/W(111). Differently, the wetting layer whose thickness is 13 PML(1 ML) formed after about 600 K annealing in Co/O-3x3/W(111), and the wetting layer of Co/O-3x3/W(111) is thinner than Co/W(111) whose thickness is 1 PML(3 ML).
Finally, about the magnetism, Co/O-3x3/W(111) reveals the in plane magnetization, and Ni/O-3x3/W(111) reveals a canted magnetization.
Bibliography
Madey, Surf. Sci., 601, 3431-3449 (2007).
[5] Y. R. Niu, and M. S. Altman, S urf. Sci., 604, 1055-1059 (2010).
(1990).
[16] C. Sorg , N. Ponpandian, A. Scherz, H. Wende, R. Nunthel ,
T. Gleitsmann, and K. Baberschke, Surf. Sci., 565, 197-205 (2004).
[17] D. J. Larson, A. K. Petford-Long, A. Cerezo, S. P. Bozeman, A. Morrone,
Ciccacci, and M. Finazzi, Phys. Rev. B, 86, 075465 (2012).
[20] C. Tölkes, R. Struck, R. David, P. Zeppenfeld, and G. Comsa, Phys. Rev.
Lett., Vo.86, Num.13 (1998).
[21] H. L. Meyerheim, D. Sander, R. Popescu, W. Pan, I. Popa, and J.
Kirschner, Phys. Rev. Lett., 99, 116101 (2007).
[22] A. Grüneis, K. Kummer, and D. V. Vyalikh, New Journal of Physics, 11,
Magn. Magn. Mater., 198-199, 716-718 (1999) [27] Chiao-Sung Chi, Master thesis, (2012)
[28] TDATA https://www.tdata.com/
[29] https://www.cnnmol.com/