鈷及鐵薄膜於鎢(111)表面上的結構與磁性

全文

(1)國立臺灣師範大學理學院物理學系研究所碩士論文. Structure and Magnetism of Co and Fe Films on W(111) Surface 鈷及鐵薄膜於鎢(111) 表面上的結構與磁性林奕成 Yi-Cheng Lin. 指導教授林文欽教授宋克嘉博士. 中華民國一百年一月.

(2)

(3) 1.

(4) 中文摘要在本篇報告中我們於 100K 的溫度將鈷或鐵蒸鍍在鎢(111)的表面並升溫至室溫後，藉著低能量電子繞射以及磁光柯爾效應探討其結構與磁性，此外我們將呈現樣品熱處理過程中所觀察到的特殊磁現象。. 2.

(5) Abstract Co and Fe films grown on W(111) were studied. After low temperature (100K) deposition with post-annealing at 300K, the crystalline structure, morphology, and magnetic behavior were measured by low energy electron diffraction (LEED), and magneto optical Kerr effect (MOKE). The special annealing effect on the magnetic behavior is also discussed.. 3.

(6) INDEX 中文摘要 ........................................................................................................................................... 2 Abstract ............................................................................................................................................. 3 Chapter 1. Introduction ...................................................................................................................... 8 1.1. Motivation ........................................................................................................................... 8. 1.2. About W(111) substrate .................................................................................................. 10. 1.3. About CIMS (Current-Induced-Magnetism-Switching) ............................................... 11. Chapter 2. Experiment ................................................................................................................. 12 2.1 UHV chamber and Equipment ................................................................................................ 12 2.2 Experiment method .................................................................................................................. 14 Chapter 3. Results and discussion............................................................................................... 16 3.1. Structure of LT-growth Co/W(111) ................................................................................ 16. 3.2. Magnetism of LT-growth Co on W(111) ......................................................................... 20. 3.2.1 Case 1 .............................................................................................................................................. 20 3.2.2 Case 2 .............................................................................................................................................. 23 3.3. Effect of sample current of Co and Fe on W(111)......................................................... 25. 3.3.1. Sample A ................................................................................................................................... 27. 3.3.1.1. Effect of sample in Co/W(111) ........................................................................ 27. 3.3.1.2. Relation of temperature and bias................................................................... 30. 3.3.1.3. Relation of thickness and bias ........................................................................ 31. 3.3.1.4. About interface and surface ............................................................................ 34. 3.3.1.5. Bias of different position on sample .............................................................. 36. 3.3.2. Sample B ................................................................................................................................... 38. 3.3.2.1 Case I ............................................................................................................................ 38 3.3.2.1.1 Magnetism and effect of current on Fe/W(111) ........................................ 38 3.3.2.1.2 Relation of current density and bias ........................................................... 42 3.3.2.2 Case II ........................................................................................................................... 47 Chapter 4. Summary ..................................................................................................................... 49 Reference ......................................................................................................................................... 52. 4.

(7) List of Figures [1]. Fig 1.1-1 LEED of Co/W(111) at RT and HT growth and SPLEEM of magnetism of Co/W(111) . ____ 9 Fig 1.2-1 STM images of Pd/W{112}. [2][3]. . ________________________________________________ 10. Fig 2.1-1 Our chamber’s equipments ___________________________________________________ 13 Fig 2.1-2 The schematic illustration of AC MOKE.. [7]. _______________________________________ 13. Fig 2.2-1 The Auger spectrum of Co on W with C and O. ___________________________________ 14 Fig 2.2-2 The relation of our deposition of Co. ___________________________________________ 15 Fig 2.2-3 TPD of different thickness Co on W (111). _______________________________________ 15 Fig 3.1-1 LEED of clean W(111) and nTML Co/W(111). _____________________________________ 16 Fig 3.1-2 Top view of bcc W(111) and hcp Co. ____________________________________________ 17 Fig 3.1-3 IV-LEED of clean W(111) and nTML Co/W(111). __________________________________ 18 Fig 3.1-4 Profile chart of bcc W(111). ___________________________________________________ 18 Fig 3.1-5 Profile chart of hcp Co(0001).. [8]. _______________________________________________ 19. Fig 3.2.1-1 MOKE of nTML Co/W(111) in in-plane direction and perpendicular direction. ________ 20 Fig 3.2.1-2 MOKE of 4.7 TML Co/W(111) in in-plane direction and perpendicular direction with different temperature. ______________________________________________________________ 21 Fig 3.2.1-3 Temperature V.S. Hc and Temperature V.S. MR of 4.7 TML Co/W(111) in in-plane direction and perpendicular direction. _________________________________________________________ 22 Fig 3.2.2-1 MOKE of nTML Co/W(111) in in-plane direction and perpendicular direction. ________ 24 Fig 3.3-1 Shape and size of sample A and B. _____________________________________________ 25 Fig 3.3-2 LEED patterns of sample A and B in 150eV. ______________________________________ 26 Fig 3.3.1.1-1 MOKE of about 10 TML Co/W(111) in in-plane direction and perpendicular direction with different current passed and temperature. _________________________________________ 27 5.

(8) Fig 3.3.1.1-2 Relation of current and bias and fitting in Fig 3.3.1.1-1. _________________________ 28 Fig 3.3.1.1-3 MOKE of about 10 TML Co/W(111) in perpendicular direction with reversal current passed. ___________________________________________________________________________ 28 Fig 3.3.1.2-1 Device of radiation heating. _______________________________________________ 30 Fig 3.3.1.2-2 MOKE of about 10 TML Co/W(111) in perpendicular direction by radiation heating. _ 30 Fig 3.3.1.3-1 Current V.S. bias of nTML Co/W(111) in perpendicular direction.r ________________ 31 Fig 3.3.1.3-2 Relation between thickness and bias per current. _____________________________ 32 Fig 3.3.1.3-3 Current V.S. bias of Fe/W(111) in perpendicular direction. ______________________ 32 Fig 3.3.1.3-3 Fitting of Current V.S. bias of nTML Co/W(111) and Fe/W(111) in perpendicular direction. _________________________________________________________________________ 33 Fig 3.3.1.4-1 MOKE of about 10 TML Co/O2/W(111) in perpendicular direction.________________ 34 Fig 3.3.1.4-2 MOKE of O2/ about 10 TML Co/W(111) in perpendicular direction. _______________ 35 Fig 3.3.1.5-1 MOKE of different positions of about 10 TML Co/W(111) in perpendicular direction. 36 Fig 3.3.1.5-2 The bias 2A passed V.S. positions of horizontal (along Y axis) and vertical (along Z axis) of about 10 TML Co/W(111) in perpendicular direction.r __________________________________ 37 Fig 3.3.1.5-3 The perpendicular direction magnetic field from sample holder at the front surface along the vertical and horizontal center line. Data by Dr. Ker-Jar Song. _______________________ 37 Fig 3.3.2.1.1-1 MOKE of about 20 TML Fe/W(111) in in-plane direction and perpendicular direction with different current passed. ________________________________________________________ 39 Fig 3.3.2.1.1-2 MOKE of about 20 TML Fe/W(111) in perpendicular direction with -3A to +3A. passed _________________________________________________________________________________ 39 Fig 3.3.2.1.1-3 Current V.S. bias of about 20 TML Fe/W(111) in perpendicular direction. _________ 40 Fig 3.3.2.1.1-4 Loop of Kerr signal V.S. linear reversal current passed of about 20 TML Fe/W(111) in perpendicular direction. _____________________________________________________________ 41 Fig 3.3.2.1.2-1 MOKE which scanned along the longer side (from left to right side) of about 20 TML Fe/W(111) in perpendicular direction . _________________________________________________ 42. 6.

(9) Fig 3.3.2.1.2-2 Position from left to right side V.S. bias with 2A of about 20 TML Fe/W(111) in perpendicular direction. _____________________________________________________________ 43 Fig 3.3.2.1.2-3 MOKE with positions of the center’s left side with 2A and ±1000 Oe magnetic field of about 20 TML Fe/W(111) in perpendicular direction. _____________________________________ 44 Fig 3.3.2.1.2-4 The contour map of bias with positions of the center’s left side with 2A of about 20 TML Fe/W(111) in perpendicular direction. _____________________________________________ 44 Fig 3.3.2.1.2-5 MOKE with positions of the center’s right side with 2A and ±1000 Oe magnetic field of about 20 TML Fe/W(111) in perpendicular direction. ___________________________________ 45 Fig 3.3.2.1.-6 The contour map of bias with positions of the center’s right side with 2A of about 20 TML Fe/W(111) in perpendicular direction. _____________________________________________ 45 Fig 3.3.2.1.2-7 The contour map of current density of around center of Sample B. _____________ 46 Data by Jai-Wei Shin ________________________________________________________________ 46 Fig 3.3.2.2-1 MOKE of Fe/W(111) in in-plane direction direction with sample current passed. Data by Jai-Wei Shin. ______________________________________________________________________ 47 O. Fig 3.3.2.2-2 MOKE of Fe/W(111) in perpendicular direction with 9A passed and rotate 1.4 from direction of magnetic field. Data by Jai-Wei Shin. ________________________________________ 48 Fig 3.3.2.2-3 Loop of Kerr signal V.S. linear reversal current passed of Fe/W(111) in in-plane direction. Data by Jai-Wei Shin. ________________________________________________________________ 48. 7.

(10) Chapter 1. Introduction 1.1 Motivation Previously, we had performed a lot of experiment about the faceting surface of Pd/W{112} which is prepared by depositing over 1 TML Pd on W(111) and then annealing to about 900K. Much background knowledge of W(111) has been accumulated in our lab. Not only the above reasons, but also we had the devices and techniques to understand the structure, growth, and magnetism of the sample at low temperature. So we had try to prepare Fe/Pd/W{112} in order to explore the phenomenon and application on this system. In Ref. [1], they have shown not only the growth of Co on W(111) with 1 x 1 structure at low temperature and 6 x 6 structure at high temperature (like Fig. 1.1-1) but also magnetic behavior. Therefore, we are interested in how are the structure and magnetism at low temperature growth. In addition, we found the special effect with current passing through Co/W(111). So we use devices which are available in our chamber to understand thoroughly. And we hope this effect will be applied in future.. 8.

(11) Fig 1.1-1 LEED of Co/W(111) at RT and HT growth and SPLEEM of magnetism of Co/W(111)[1].. 9.

(12) 1.2 About W(111) substrate It is known that the thin films of transition and noble metals such as Pt or Pd are capable to induce thermal faceting phenomenon on W(111) surface. And the surface of W(111) with Pd-covering becomes faceting of 3-sided {112} pyramids after about 700K ~ 900K annealing if coverage of Pd is greater than a certain amount. Like Fig. 1.2-1. The pyramids of faceting are a very good method in preparing reproducible single atom tips for STM or as the field emission sources. Surface of W(111) is rough so that structure of the deposition is complicated. And the magnetism which is deposited ferromagnetic mater is less to study.. Fig 1.2-1 STM images of Pd/W{112}[2][3].. 10.

(13) 1.3 About CIMS (Current-Induced-Magnetism-Switching) How to let magnetism be transferring or switching by electric current is usually studied. Everyone who investigates it hopes to have application in our world. The name about CIMS is usually seen in the systems of GMR (giant magneto resistance). The CIMS of GMR is that the magnetization of a thinner (soft) FM (ferromagnetic) layer of a FM/NM/FM (NM is nonmagnetic metal) nanopillar is flipped relative to that of a thicker (hard) FM layer by applying a current.[4] One kind of possible reason which makes magnetism reversal is spin hall effect. The spin hall effect consists of an appearance of spin accumulation on the lateral surfaces of a current-carrying sample. The spin Hall effect was first observed in GaAs at 20K by Awschalom and Yuichiro Kato in 2004[5]. It consists of a spin current flowing in a transverse direction to the charge current in a non-magnetic material and in the absence of an applied magnetic field. The result is a measurable accumulation of “spin up” and “spin down” electrons at opposite edges of the conducting channel. The electrons with fixed direction have possibility to make magnetic domain to switch. It charges of opposite sign appear on the opposing lateral surfaces to compensate, acting on electrons in an applied magnetic field. Another famous system of magnetism reversal current-induced is Rashba effect who originates from the macroscopic electric field in a semiconductor quantum well. [6]. Due to a typical conduction band offsets at the interface of two different materials. the electrons are confined in a quantum well. If the potential well is asymmetric, the electrons which are 2-dimensional gas are moving in an effective electric field. In the reference system of the electron this electrical field transforms into a magnetic field in normal direction. Then it is hopeful to make the spin reversal in perpendicular direction. 11.

(14) Chapter 2. Experiment 2.1 UHV chamber and Equipment In order to sensitiveness for surface science in our experiment, we perform the UHV (ultra high vacuum) chamber and let the base pressure be about 10-10 torr with baking chamber and TSP (titanium sublimation pump). Our devices which are equipped on chamber like Fig 2.1-1. The AES (Auger electron spectrum) is due to analysis what element on surface. So we can use the AES to check the sample surface is clean (without oxygen and carbon) or not. The LEED (low energy electron diffraction) shown the pattern which helps us to analyze the structure in horizontal on surface. And in our chamber, the sample is rotated 4∘, then we take the (0, 0) beams intensity in order to determine the vertical interlayer distance. MOKE (magnetic-optical Kerr effect) is a kind of magnetic-optical effect which describes that linear polarized laser beam becomes elliptical polarized on ferromagnetic sample reflected from magnetized media. The AC MOKE we set helps us analyze the magnetism by hysteresis curves in in-plane direction and perpendicular direction. The schematic illustration of AC MOKE is Fig. 2.1-2. RGA (residual gas analyzer) can let us know what residual gas in our vacuum and track one kind gas in our chamber. So we use it to take the TPD (temperature program adsorption) in order to measure and check what elements and how much capacity deposition on our surface of substrate. Our sample W(111) is spot-wielded on two Ta (tantalum) rods and then mounted on the 4-axis manipulator with a liquid nitrogen dewar which cools sample to about 100K. Not only but also we can pass current to let the sample heating in order to anneal or flash. And we also spot-wield the C-type thermocouple on the back of the substrate to check the temperature of sample. We deposit Co or Fe on 12.

(15) W(111) by evaporators, EFM3, which let rods of Co or Fe heating by filament and then take high voltage to be splattered on sample.. Fig 2.1-1 Our chamber’s equipments. Fig 2.1-2 The schematic illustration of AC MOKE.[7]. 13.

(16) 2.2 Experiment method At first, we need our W(111) to be clean by passing about 40 ~ 70 A on sample to let it flash to over 2000K for many times. Then take AES to check the oxygen and carbon existed or not. Like Fig. 2.2-1. If both oxygen and carbon are excited or just only oxygen, it should need more flash. If only carbon on the spectrum, we will dose O2 to let be carbon oxide and then flash for many times to let it get out off surface. When we deposit the Co, we can let substrate be about 100K by LN2. We check AES again to know thickness by the ratio of minimum to maximum values of W’s second pick and Co’s third pick. We had done the relation curve in Fig. 2.2-2 by TPD before. And let sample anneal to 300K for 5 minutes in order to make sure that the structure could be well. Then take LEED, IV-LEED, and MOKE. If we need to know thickness accurately, we will take TPD which is like Fig. 2.2-3.. Fig 2.2-1 The Auger spectrum of Co on W with C and O.. 14.

(17) Fig 2.2-2 The relation of our deposition of Co.. Fig 2.2-3 TPD of different thickness Co on W (111).. 15.

(18) Chapter 3. Results and discussion 3.1 Structure of LT-growth Co/W(111) In the part, we will show the growth of Co on W(111) and discuss if they match or not. Fig 3.1-1 shows the LEED patterns of clean W(111) and different thickness films Co on W(111). We can see the 1x1 points of W(111). And the points of the nearest center are dimmer while the Co files be thicker. We could know the lattice constant of W (111) and Co are 4.47 Å and 2.51 Å. In our ideal model, the Co will rotate 30o to match with substrate of W(111). Like Fig. 3.1-2, the Co rotates 30o to match the W, and the lattice mismatch is just 3%. So we can say that Co grows well on W(111) at low temperature.. Fig 3.1-1 LEED of clean W(111) and nTML Co/W(111).. 16.

(19) Fig 3.1-2 Top view of bcc W(111) and hcp Co.. The Fig. 3.1-3 is the IV-LEED with clean W(111) and different coverage thickness Co on W(111). We can use the Bragg condition and the de-Brogile relation, , and. ,. to calculate the changes of (0, 0) beams’ intensity which lets sample is rotated. =. 5o . d is the vertical interlayer distance, Ek is the kinetic energy of the incident electron beam, and V is the potential cost for electrons to escape from the atoms. We take the picks in Fig. 3.1-3 to last two functions respectively and average them. We solve d = 0.94 Å in black lines which is clean W(111) and low coverage and d = 1.98 Å in red lines which is high coverage.. 17.

(20) Fig 3.1-3 IV-LEED of clean W(111) and nTML Co/W(111).. And about the structure of normal direction, we can know the ideal case of W(111) and hcp Co(0001) in Fig. 3.1-4 and 5. We can see that the distance of two layers of the nearest is 2.735 Å / 3 = 0.912 Å from Fig. 3.1-4 which is similar with our lattice constant 0.94 Å of clean W(111) and low coverage Co. And the one of high coverage is similar with the hcp Co’s lattice constant 2.049 Å.. Fig 3.1-4 Profile chart of bcc W(111).. 18.

(21) Fig 3.1-5 Profile chart of hcp Co(0001).[8]. By the LEED and IV-LEED data, we regard the Co growth on W(111) is a kind of pseudomorphic. At lower thickness Co seem to be bcc structure on W bcc(111). And the structure of high coverage Co maybe is hcp.. 19.

(22) 3.2 Magnetism of LT-growth Co on W(111) In this part, we will study two cases of magnetism of Co/W(111) which are two substrates of W(111) at low temperature growth. Two cases have large different in magnetism. We can’t differentiate between two substrates by LEED. But the substrate in case 2, the carbon pick in AES has never been seen when the sample was in the chamber at beginning.. 3.2.1 Case 1. Fig 3.2.1-1 MOKE of nTML Co/W(111) in in-plane direction and perpendicular direction. Fig. 3.2.1-1 shows the hysteresis curves in different thickness Co on W(111) and taken at low temperature. By this, we can see the sample without magnetism in thicker thickness Co on W(111) (2.8 and 3.6 TML). And the loop shown in in-plane direction and not yet see loop of perpendicular direction in 5.2 TML. But we will know the both of hysteresis curves existed later. When thickness is bigger than 20.

(23) 5.2TML, the coercivity of loop in in-plane direction have not changed much and in perpendicular direction have been smaller. In general, the coercivity and magnetic resonance are increased with the magnetic films thickness increasing, but in our case is not.. Fig 3.2.1-2 MOKE of 4.7 TML Co/W(111) in in-plane direction and perpendicular direction with different temperature. In Fig. 3.2.1-2, it shows the hysteresis curves of 4.7TML at different temperature in in-plane direction and perpendicular direction by passing current. And Fig. 3.2.1-3 show relation of the coercivity and magnetic resonance with temperature in 4.7 TML Co on W(111). We will show about the bias of loops of perpendicular direction in next part. Both of coercivity of in-plane direction and perpendicular direction are decreasing as well as temperature increasing clearly. The magnetic resonance of in-plane direction is decreasing with temperature increasing, too. But its of perpendicular direction has not clearly tending with temperature. And we can see 21.

(24) the temperature of the loops disappeared in in-plane direction and perpendicular direction are different, which are about 450K and 580K. It lets us know the domains of in-plane direction and perpendicular direction in two directions are different. The total magnetism are form different magnetic axis. Another reason is that the easy axis transfers to perpendicular direction only with temperature increasing.. Fig 3.2.1-3 Temperature V.S. Hc and Temperature V.S. MR of 4.7 TML Co/W(111) in in-plane direction and perpendicular direction.. 22.

(25) 3.2.2 Case 2 This case of magnetism seems to be same with case 1 in in-plane direction. In Fig. 3.2.2-1, the loop of in-plane direction has shown in 4.6 TML Co on W(111) which resembles with case 1’s 5.2 TML. But the hysteresis curves of perpendicular direction are never observed in all of thickness which we take. At first we regard that their coercivity are much larger than our system’s magnetic field. So we pass the sample current to let temperature be higher in order to make the coercivity to be smaller in 11.8 TML. In other hand, we hope to check the effect of current-induced bias existed or not. But in Fig. 3.2.2-2, we can’t observe any loop in perpendicular direction. By case 1 and 2, this system appears that some kinds of factor we don’t know and observe. It causes the magnetism is sensitive very much in perpendicular direction. The loops in perpendicular direction seem to be existed independently, and they will be bias by current-induced. In next part, we will discuss many kinds of possible reasons in this effect.. 23.

(26) Fig 3.2.2-1 MOKE of nTML Co/W(111) in in-plane direction and perpendicular direction.. Fig 3.2.2-2 MOKE of about 11.8 TML Co/W(111) in in-plane direction and perpendicular direction with different temperature. 24.

(27) 3.3 Effect of sample current of Co and Fe on W(111) In this part I will show two kind of form of sample W(111). One kind is Fig. 3.3-1’s Sample A and another one (Sample B) which induce to test the relation of current density. About the cross sectional, the sample A is about 3.00mm X 0.30mm = 0.90mm2, and the sample B is about 3.00mm X 0.40mm = 1.20mm2 in the side and about 1.50mm X 0.40mm = 0.60mm2 in the center. The sample B’s mass is bigger than A, so sample B is more hard to heat by current. Not only the form but also the crystal axis of them is different. The LEED patterns of Fig. 3.3-2 are shown where are different. The crystal axis of longer side of Sample A is [1-10], and which of Sample B is [11-2]. So we will discuss the effect of current in different case in order to understand this special effect in this system.. Fig 3.3-1 Shape and size of sample A and B.. 25.

(28) Fig 3.3-2 LEED patterns of sample A and B in 150eV.. 26.

(29) 3.3.1Sample A 3.3.1.1. Effect of sample in Co/W(111). At first, we show the loops of MOKE in in-plane direction and perpendicular direction of about 10 TML Co on W(111) at low temperature growth which passed sample current (Fig. 3.3.1.1-1). In Fig. 3.3.1.1-1, the loops in in-plane direction have not changed when we pass the sample current except which temperature is higher. But which in perpendicular direction, the loops have been bias obviously. And the bias in perpendicular direction is higher with current and temperature ascended. We suppose which the bias is related with current only, and we will discuss really causes latter. So we take a relational graph between current and bias in Fig. 3.3.1.1-2. By the fit curve in Fig. 3.3.1.1-2, we find that the sample current and bias are a perfect linear relation. In in-plane direction, we can there are no bias or say the bias in in-plane direction is much smaller than 2.4 Oe (by measure point).. Fig 3.3.1.1-1 MOKE of about 10 TML Co/W(111) in in-plane direction and perpendicular direction with different current passed and temperature. 27.

(30) 250. Bias (Oe.). 200 150 fit curve Y = 28.18X + 4.85. 100 50 0 0. 1. 2. 3. 4. 5 Current (A). 6. 7. 8. 9. 10. Fig 3.3.1.1-2 Relation of current and bias and fitting in Fig 3.3.1.1-1.. By the way, when we pass the reverse direction current on another sample of 10 TML Co/W(111), the direction of bias of loop is also opposite. Like Fig. 3.3.1.1-3.. Fig 3.3.1.1-3 MOKE of about 10 TML Co/W(111) in perpendicular direction with reversal current passed. In this case, something induced loop bias. The reasons which let bias happened could be temperature, current, or couple of temperature and current caused. Of 28.

(31) course, it’s could be the electromagnetic induction of the sample holder made. But which an easy square loop wants to make about 28 Oe/A electromagnetic field is impossible really. Later, we will show this effect is related with current induced bias only.. 29.

(32) 3.3.1.2. Relation of temperature and bias. In last part, we can see the bias increased with the current and temperature. In this part, we let change be the temperature only, and discuss the relation of temperature and bias. In this case, we set a filament which is the form of whirlpool back of the sample in order to radiation heating. Like Fig. 3.3.1.2-1. Fig. 3.3.1.2-2 shows three kind of temperature of about 10TML Co on W(111). In this picture, the loops don’t have bias with temperature increase. So we exclude the reason of temperature or couple of temperature and current induced bias.. Fig 3.3.1.2-1 Device of radiation heating.. Fig 3.3.1.2-2 MOKE of about 10 TML Co/W(111) in perpendicular direction by radiation heating.. 30.

(33) 3.3.1.3. Relation of thickness and bias. At 3.3.1.1, we regard the linear relation between with current and bias. Then, we prepare many sample with different thickness Co on W(111) in order to understand the relation between the bias and the thickness in Fig. 3.3.1.3-1. The distribution of points is with a kind of regulation. The bias seems to be in direct proportion to current. And the bias and the thickness seem to be not clearly related much in Fig.3.3.1.3-2 which is studied the relation of slope by we fit in Fig. 3.3.1.3-1 (just the bias per current) and thickness. We also use the Fe/W(111) to test the bias in Fig. 3.3.1.3-3. In the picture, the sample of Fe has no much change with other samples of Co. The bias is 27.88 Oe/A for Fe. It seems to tell us that this effect is existed which the hysteresis curves are observed in perpendicular direction no matter what kind’s ferromagnetism metal. By the way, the effect is about 24.8 Oe/A at all of thickness Co/W(111) and Fe/W(111) in Fig. 3.3.1.3-3 by our fitting.. Fig 3.3.1.3-1 Current V.S. bias of nTML Co/W(111) in perpendicular direction.r 31.

(34) Fig 3.3.1.3-2 Relation between thickness and bias per current.. Fig 3.3.1.3-3 Current V.S. bias of Fe/W(111) in perpendicular direction.. 32.

(35) Fig 3.3.1.3-3 Fitting of Current V.S. bias of nTML Co/W(111) and Fe/W(111) in perpendicular direction.. 33.

(36) 3.3.1.4. About interface and surface. In directly perceived through the senses, we regard that the effect be happened from the special structure of Co on W(111). So we hope to discredit the interface between Co and W in order to want to let the bias be changed or not. In this case, we dose the O2 with 10-7 torr for 5 minutes on W at low temperature. Then deposit about 10 TML Co and anneal to room temperature. The data of MOKE in perpendicular direction is Fig. 3.3.1.4-1. We can see that the coercivity become increased at lower temperature. It still have bias with 10A passed. And the bias is about 20.0 Oe/A.. Fig 3.3.1.4-1 MOKE of about 10 TML Co/O2/W(111) in perpendicular direction. And we prepare another 10TML Co/W(111) sample with dosing O2 (10-7 torr for 5 minutes) in order to discuss the effect of surface. In Fig. 3.3.1.4-2, the coercivity seems to become more increased. The bias exist with passed the current but the bias become 35.5 Oe/A.. 34.

(37) Fig 3.3.1.4-2 MOKE of O2/ about 10 TML Co/W(111) in perpendicular direction. In last two cases, we cannot be sure what the interface and surface contribute in this system of bias. But the interface and surface seem to be not sensitive with this effect.. 35.

(38) 3.3.1.5 Bias of different position on sample This part, we try to see the bias in different position. If the bias is contributed from the electromagnetic induction of the sample holder, it will be sensitive with position. So we take MOKE of about 10 TML Co/W(111) passed 6A on area about 4 X 2 mm2 around the center. The picture of position and loop is Fig. 3.3.1.5-1.. Fig 3.3.1.5-1 MOKE of different positions of about 10 TML Co/W(111) in perpendicular direction.. And we analyze the bias of horizontal (along Y axis) and vertical (along Z axis) in Fig. 3.3.1.5-2. the picture show that the bias is no change very much. In other words, it’s not sensitive with position. By the calculation of Fig. 3.3.1.5-3, we can see that the magnetic field from sample holder along Y-axis and Z-axis, and specially the field seems to be reversal along Z-axis. It is not matched with the Fig. 3.3.1.5-2.. 36.

(39) Fig 3.3.1.5-2 The bias 2A passed V.S. positions of horizontal (along Y axis) and vertical (along Z axis) of about 10 TML Co/W(111) in perpendicular direction.r. Fig 3.3.1.5-3 The perpendicular direction magnetic field from sample holder at the front surface along the vertical and horizontal center line. Data by Dr. Ker-Jar Song. By sample A, we cannot understand the relation of current density and bias clearly. So we make the sample B in order to show the different current density on one sample.. 37.

(40) 3.3.2 Sample B In sample B, we change to use the deposition of Fe on W(111) which about 20 TML Fe of all most. We will discuss the magnetism and use the shape of this sample to take more ideal of this effect. In this substrate, we have seen two cases of magnetism, Case I and Case II in different moments.. 3.3.2.1 Case I 3.3.2.1.1 Magnetism and effect of current on Fe/W(111) This Case I is similar to Case 1 of 3.3.1. The MOKE data of in-plane direction and perpendicular direction in around center is Fig. 3.3.2.1-1. Roughly, the magnetism of Fe/W(111) is same as Co/W(111) in this sample. The bias is unexcited in in-plane direction. And it is not only excited in perpendicular direction but also seems to be larger than sample A. Then we pass the two directions current on the sample and the Fig. 3.3.2.1-2 shown. The bias of center part of sample is shown in Fig. 3.3.2.1-3. In Fig. 3.3.2.1-3, it have a perfect linear relation of current and bias which is increased to about 180 Oe/A which is much bigger than sample A, and it has heating not much than 3K by 4 A passed. So we can say this effect by sample current is only dependent with the current.. 38.

(41) Fig 3.3.2.1.1-1 MOKE of about 20 TML Fe/W(111) in in-plane direction and perpendicular direction with different current passed.. Fig 3.3.2.1.1-2 MOKE of about 20 TML Fe/W(111) in perpendicular direction with -3A to +3A. passed. 39.

(42) Fig 3.3.2.1.1-3 Current V.S. bias of about 20 TML Fe/W(111) in perpendicular direction.. In the part of 3.3.1.3, we have shown the bias on Fe/W(111) is not changed with Co/W(111). Maybe the reasons of the bias increased are the current density and the axis of crystal with current passed of W. And the front one is shown in next part that the bias is still larger than last sample on the sample of right and left sides which the cross-sectional area is bigger than sample. So the possible reason is the axis of crystal with current passed of W. Because the effect of the bias by current induced is bigger and linear relation with current, we can use no much current passed which lets loops change not much in order to take the loop to be bias to another side. So we just try to use current only to let the magnetism reversal. The Fig. 3.3.2.1-4 is successful to be shown that.. 40.

(43) Fig 3.3.2.1.1-4 Loop of Kerr signal V.S. linear reversal current passed of about 20 TML Fe/W(111) in perpendicular direction.. 41.

(44) 3.3.2.1.2 Relation of current density and bias The Fig. 3.3.2.2-1 shows the loops which we scan from left side to right side with 0 A and 2 A passed in order to click the different current density’s bias. And we can see the clearly change in Fig. 3.3.2.2-2 with the bias passed 2 A. We can see the bias is the biggest in the center of the sample this picture. The distribution of the cross-sectional area is that the center is smaller than left and right side. So the current density is also bigger in the center. Just like Fig. 3.3.2.2-2 which is made sense.. Fig 3.3.2.1.2-1 MOKE which scanned along the longer side (from left to right side) of about 20 TML Fe/W(111) in perpendicular direction .. 42.

(45) Fig 3.3.2.1.2-2 Position from left to right side V.S. bias with 2A of about 20 TML Fe/W(111) in perpendicular direction.. As this relation with current density, we try to check the loops of positions around the sample center which passed 2A more clearly in Fig. 3.3.2.2-3 and Fig. 3.3.2.2-5, and analyze the contour map of bias with three dimensions in Fig. 3.3.2.2-4 and Fig. 3.3.2.2-6. And let we see the Fig. 3.3.2.2-7 which is the current density of sample B by calculating. Compare with Fig. 3.3.2.2-4, 3.3.2.2-6, and 3.3.2.2-7. The Fig. 3.3.2.2-7’s boundaries of the contour lines are matched with Fig. 3.3.2.2-4 and 3.3.2.2-6 roughly. So we regard this effect is related with current density.. 43.

(46) Fig 3.3.2.1.2-3 MOKE with positions of the center’s left side with 2A and ±1000 Oe magnetic field of about 20 TML Fe/W(111) in perpendicular direction.. Fig 3.3.2.1.2-4 The contour map of bias with positions of the center’s left side with 2A of about 20 TML Fe/W(111) in perpendicular direction.. 44.

(47) Fig 3.3.2.1.2-5 MOKE with positions of the center’s right side with 2A and ±1000 Oe magnetic field of about 20 TML Fe/W(111) in perpendicular direction.. Fig 3.3.2.1.-6 The contour map of bias with positions of the center’s right side with 2A of about 20 TML Fe/W(111) in perpendicular direction.. 45.

(48) Fig 3.3.2.1.2-7 The contour map of current density of around center of Sample B. Data by Jai-Wei Shin. 46.

(49) 3.3.2.2 Case II This case is happened after Case I, and we can’t repeat Case I. It’s different with Case I. We take loops of in-plane direction of MOKE only and we can’t observe magnetism of perpendicular direction (Fig. 3.3.2.2-1 and 3.3.2.2-2). It’s similar to Case 2 of 3.2.2. Not only but also we can see the in-plane direction loops is bias by sample current (Fig. 3.3.2.2-1). The bias is about 1 Oe/A, and the coercivity is about 18 Oe. So we can just switch magnetism (Kerr signal) by current. Like Fig. 3.3.2.2-3 When we rotate 1.4o from magnetic field of perpendicular direction, we can observe the loop which is bias with current in perpendicular direction (3.3.2.2-2). We regard those loops of perpendicular direction is from magnetism of in-plane direction. Because the coercivity of in-plane direction divides by Sin1.4o equals about 750 Oe which is enough to let magnetic field of perpendicular direction switch magnetism of in-plane direction completely.. Fig 3.3.2.2-1 MOKE of Fe/W(111) in in-plane direction direction with sample current passed. Data by Jai-Wei Shin.. 47.

(50) Fig 3.3.2.2-2 MOKE of Fe/W(111) in perpendicular direction with 9A passed and rotate 1.4O from direction of magnetic field. Data by Jai-Wei Shin.. Fig 3.3.2.2-3 Loop of Kerr signal V.S. linear reversal current passed of Fe/W(111) in in-plane direction. Data by Jai-Wei Shin.. 48.

(51) Chapter 4. Summary Co on W(111) at low temperature (100K) after annealing to 300K reveals well ordered structure with LEED 1 X 1 pattern. The hcp (0001) Co matches on W(111) surface by ideal model and the lattice mismatch is only 3% in horizontal plane. In normal direction, the Co films follow the substrate W bcc(111) surface to be bcc structure in low coverage. And the Co films’ structure becomes hcp (0001) which is like the bulk of Co gradually with Co thickness increasing. In other words, Co is bcc on W(111) from the beginning and transfers to hcp of itself in high coverage. About magnetism of Co on W(111) at low temperature growth, we can observe the magnetism of in-plane direction with hysteresis curves in about 5 TML Co thicknesses after annealing to room temperature. But we have two different results in our experiment, the magnetism of Co/W(111) in perpendicular direction at low temperature growth is not sure to be excited. If it is observed (Case 1), it will be existed in about 5 TML like in in-plane direction. And the magnetism of perpendicular direction is independent with in-plane direction. Not only but the loops in perpendicular direction will be bias when we pass current on the sample. In another W(111) sample, we can’t see the loops in perpendicular direction, just like it’s not magnetic in perpendicular direction (Case 2). This part is uncontrolled and repeated by us presently. The effect which is found by annealing lets the hysteresis curves be biased. We know the bias is linear with sample current, and related with current density. It seems to happen on the system of ferromagnetic matter (like Co or Fe) on W(111). The value of bias per unit is related besides current density and crystalline axis which we observed. We have several cases of current-induced bias. It seems to have some kind of sensitive conditions to make this effect which we will research in the future. 49.

(52) There are 4 kinds of magnetic cases of Co or Fe on W(111) in our system. The main features are observed in Case 1 and 2 of Co/W(111), and Case I and II of Fe/W(111). In Case 1 is that we observed the magnetism of in-plane direction and perpendicular direction in Co/W(111). The loops of Co or Fe on W(111) were biased In perpendicular MOKE only. The bias is linear with sample current. The average value slope of bias/current is about 25 Oe/A. Though the loops of perpendicular direction are possible from in-plane direction, the bias is observed only in perpendicular direction. In Case 2 is that we only saw the loops in in-plane direction which are never biased with current in Co on W(111). Case I is similar to Case 1 but it is Fe on W(111) and the average value slope of bias/current is about 180 Oe/A which is much more than Case 1. Case II which is only loops in in-plane direction and bias with current is in same sample with Case I. We can’t repeat Case I again when Case II appears. It seems to exist some sensitive reasons of magnetism in this system. Maybe there are some new physics, or not. MOKE uses optic to study magnetism indirectly. If we want to understand this special effect clearly, we should use more direct methods to measure.. 50.

(53) 51.

(54) Reference 1. K. L. Man, R. Zdyb, S. F. Huang, T. C. Leung, C. T. Chan, E. Bauer, and M. S. Altman. Phys. Rev. B 67, 184402 (2003). 2. T.E. Madey, C.-H. Nien et al, Surf. Sci. 438(1999), 191 3. C.-H. Nien and T.E. Madey, Surf. Sci. 380(1997), L527 4. K. W. Cheng, C. Yu, L. K. Lin, Y. D. Yao, Y. Liou, J. H. Huang, and S. F. Lee, APPLIED PHYSICS LETTERS 96, 093110 (2010) 5. Y. K. Kato, R. C. Myers, A. C. Gossard, D. D. Awschalom, SCIENCE, (2004), VOL 306 6. I. M. Miron, G.Gaudin, S. Auffret, NATURE MATERIALS, VOL 9, (2010) 7. K. Sato, Japn. J. Appl. Phys. 20, 2403 (1981) 8. http://cst-www.nrl.navy.mil/lattice/struk/a3.html. 52.

(55)