穿隧掃描顯微鏡與場離子顯微鏡研究 納米結構的自組裝機制與控制方法
全文
(2) Contents 摘要 Abstract. Chapter One : Introduction ...................................................................1 References................................................................................................... 9. Chapter Two : Experimental Techniques and Methods ............... 11 2.1 Instrumentation ................................................................................... 11 2.2 Experimental procedure ...................................................................... 24 Reference .................................................................................................. 27. Chapter Three : Study of self- assembled of Co-TPP on Pb/Si (111) surface by STM ............................................................................. 28 3.1 Introduction......................................................................................... 28 3.2 Disorder result of Co-TPP adsorbed on Si (111) ................................ 44 3.3 Geometry transition of Co-TPP on Pb/Si (111) - √7 × √3 (or HIC) depend temperature. ........................................................................... 48 3.4 Morphology conversion of Co-TPP adsorbed on Pb/Si (111) – SIC phase .................................................................................................. 83 3.5 Summary ............................................................................................. 98 References............................................................................................... 101.
(3) Chapter Four : Fabrication of Mo pyramidal-shape single atom tips covered by a noble metal .......................................................... 105 4.1 Introduction....................................................................................... 105 4.2 The pure Mo system ......................................................................... 108 4.3 Pd-induced faceting .......................................................................... 111 4.4 Mo(111) pyramidal tip covered with Pt, Rh, and Ir .......................... 113 4.5 Summary ........................................................................................... 117 References............................................................................................... 118. Chapter Five : Field ion microscopy study of 1 × 2 reconstruction on Pt surfaces ......................................................................................... 120 5.1 Interdiction ........................................................................................ 120 5.2 Saw-tooth structure reconstruction ................................................... 123 5.3 Ascending motion ............................................................................. 127 5.4 Stability ............................................................................................. 131 5.5 Summary ........................................................................................... 132 References............................................................................................... 134. Chapter six : Conclusions ................................................................... 135.
(4) 摘要 自組裝是透過物件自身的交互作用力組合成元件的機制,並且 自組裝結構是一種最低能量也最穩定的結果。當機械或電子設備的 漸漸小型化而使得製造也將越來越費工耗時,因此物件的自組裝是 一種經濟而有效的方式。 在這篇論文中,介紹了三個關於自組裝的研究。第一部分是以 穿隧掃描顯微鏡(STM)研究 Co-TPP 分子自組裝在不同鍍量(1 ~ 1.3 ML)的矽(111)表面。我們發現透過調整鉛的鋪附量可以改變分子的 自組裝結構:第一種自組裝結構是分子會以三種不同的結構(鞍型, 平面型的和異平面型)表現形成各自的結構域在 √7 × √3 的 Pb/Si (111)基底結構上。這結構中我們還可以發現平面型和異平面型的 Co-TPP 分子形貌會隨溫度相變。第二種自組裝結構是鞍型與平面型 的 Co-TPP 分子會形成交錯排列成有週期性且更為緊密的結構在「線 條狀不相稱相(SIC)」的 Pb/Si(111)基底結構上。這樣的轉換機制來 自於 Co-TPP 的鈷原子和 Pb/Si(111)襯底的相互作用。 表面的皺化與失蹤原子列的產生,都是為了得到最低的表面自 由能而去改變表面的形貌。所以第二部分是研究鉬單原子針的自組 裝。我們利用場離子顯微鏡觀察純鉬針與鈀,鉑,銠,銥鋪附鉬針 經退火後的皺化結果。金字塔形單原子針已於形成鈀,鉑,銠鉬覆 鉬針。會有兩種類型的金字塔結構形成,分別為 1、3、10 或 1、6、.
(5) 15 的結構。然而,純鉬和銥附鉬針因為表面能異向性差異不足以及 銥容易退吸附及與鉬合金而無法形成單原子針。 最後我們同樣利用場離子顯微鏡研究鉑 1 × 2 的失蹤原子列重 構在鉑(110)和鉑(331)的表面。對於鉑(110)面, 經退火到 450K 發現 從 1 × 1 過渡到 1 × 2 結構是以跳躍或下行的原子運動產生。對於鉑 (331)而言類似的轉變發生在加熱至 600K,特別的是形成上兩層都為 1 × 2 結構的鋸齒模型。我們提出一種新的結構模型解釋鉑(331) (1×2)重構。. 關鍵字:掃描穿隧顯微鏡,場離子顯微鏡,自組裝行為,四苯基鈷 卟啉,單原子針,失蹤原子列。.
(6) Abstract Self-assembly is to coalescence many tiny elements to form a component by the self-interaction force, meaning that microstructures automatically assemble on the substrate. When the device miniaturization makes the artificial or mechanical manufacture impossible, the component self-assembly is an economical and effective way. In this thesis, three parts of investigations are included. The first part is to study how structure of cobalt(II) - tetraphenylporphyrin (Co-TPP) self-assembly on different phases of Pb plated on Si (111) surface by STM. We found the short-ranged domains and periodic islands of self-assembly structure can be adjusted by changing coverage of Pb. For one case, the molecular- layer constitutes with three kinds of domain formed by different types which were saddle, planar and iso-planar on Pb/Si (111) - √7 × √3 phase. The structure included low-temperature phase transition for planar and iso-planar conformation. For another case, periodic structure consists of staggered of saddle and planar conformation strips on SIC phase, and the structure is more closely than the first. Such the transition mechanisms are produced in the interaction of the center metal atom of Co-TPP and Pb/Si (111) substrate..
(7) The second part is the Mo single-atom-tips (SATs) are self-assembly forming the faceting phenomena. It has been studied for pure Mo and Pd, Pt, Rh and Ir covered Mo tips by FIM. The pyramidal-shape single atom tips were formed on Pd, Pt and Rh covered Mo tips. Two types of pyramidal structure which is stacked by 1, 3, 10 or 1, 6, 15 atoms for the top three layers were found. However, no single atom tips were found for pure Mo tips and Ir covered-tips due to the insufficient surface-free-energy anisotropy and easy to alloy with Mo. The finally part investigated an atomic structure of a 1 × 2 reconstruction of Pt (110) and Pt (331) crystallographic planes by FIM. For the Pt (110) a transition from a 1 × 1 phase into the 1 × 2 missing-row structure is observed upon annealing treatment at 450 K which is interpreted in terms of jumping or descending atom motions. For the Pt (331) similar transition takes place upon heating at 600 K, and the atomic arrangement in two top 1 × 2 layers is compatible with so called saw-tooth model. We propose a new structural model for the Pt (331) - (1 × 2) reconstruction.. Keywords: Scanning tunneling microscope, Field ion microscope, Self-assembly, Tetra-phenyl cobalt porphyrin, Single atom tip, Missing-row.
(8) Chapter One. Introduction In the future, the smaller and more sophisticated electronic components would not be easy manufactured. We'll be able to snap together the fundamental building blocks of nature easily by nanotechnology, inexpensively and in most of the ways permitted by the laws of nature. The self-assembly formed by atoms or molecules to materials, and next that interlink to components or equipment and specific functions system. This will let us continue the revolution in computer hardware to its ultimate limits: molecular computers made from molecular logic gates connected by molecular wires. This has been a dream away of the application of nanotechnology industry. This new pollution free manufacturing technology will also let us inexpensively fabricate a cornucopia of new products that are remarkably light, strong, smart, and durable. Self- assembly is the fundamental principle which generates structural organization on all scales from molecules to galaxies. The self-assembly is a consequence of specific of rearrangement process for disordered system of existing components forms an ordered or. 1.
(9) organized structure, which local interactions among the components themselves, without external direction.. However, when the. constitutive components are molecules, the process is termed molecular self-assembly. In essence, the basic nanotechnology is a bottom-up technology like "house made of brick". The molecular self-assembly is likely to further assemble into components. Generalized self-assembly mechanism can be defined as the natural tendency of the physical system. The self-assembly mechanism can be classified as either static or dynamic. In static self-assembly, the ordered state forms as system approaches equilibrium, reducing its free energy through the energy exchange by internal and external of system. Various configurations of the particle random mobile to aggregation by thermal diffusion which gradually form a stable state, and finally the system reached equilibrium. In a dynamic system may produce a more complex state, such as water molecules in the ice chamber selfassembled into ice crystals. Similarly, the water molecules selfassembled into complex snowflake conditions in turbulent clouds frequent changes of temperature and humidity. So the many fascinating structures in nature are self- assembly phenomena. In this thesis, we will discuss three different research topics. 2.
(10) about the self-assembly. The first is the experimental of molecular self- assembly on interface of metal- semiconductor, and the second is how to make the thermal stability single-atom-tip (SAT) used by molybdenum (Mo), and the third is to discuss the missing- row formation mechanism on platinum (Pt) surface. This thesis will discuss the emphasis on the first study. Molecular self- assembly is the assembly of molecules without guidance or management from an outside source. In the experiment of molecular self-assembly on Metal-semiconductor interface, we study self-assemble behavior the cobalt (II) - tetraphenylporphyrin (Co-TPP) molecules adsorbed on Pb/Si (111) surface. We found that the main assembled morphology of Co-TPP adsorbed on Pb/Si (111) - √7 × √3 (or HIC) phase is substantially the same it assembled to the noble metal (including Au, Ag and Cu) [1 - 3]. Assembling molecules and molecules depend on T-stacking interaction. There have three types of saddle, planer and iso-planer conformation formed on the surface, different on the noble metal. Interestingly, our can to control the morphology structure of molecular self-assembling result by changing the coverage of Pb plated on Si (111) are surprised we. We are very pleased to get this result of “directing or directed self-assembly”. As previously. 3.
(11) described, the Co-TPP will distort formed three type conformations, there can be forming separate each short-range order domains adsorbed on Pb/Si (111) - √7 × √3 (or HIC) phase. However, the Co-TPP assemble to ordered and periodic structure formed on SIC phase substrate when the 1.3 ML coverage of Pb on Si (111). This special case is only formed by the two conformation, which saddle and planar. We will discuss that difference of the Co-TPP self-assembled structure in this thesis.. a. b. Fig. 1.1.1. (a) Molecules are disordered on the surface. (b) Those rearranged to orderly structure through molecule-molecule or molecule-substrate interactions.. The part of thermal stability SAT. Our studied faceting to SATs phenomena in the pure Mo and Pd, Pt, Rh and Ir covered Mo tips by. 4.
(12) field ion microscopy (FIM). There we know the self-assembly attributed to the particles free energy tends to minimum and stable results. A crystal will always have several lower surface-free-energy index facets formed on surface, like (111) and (100) in face-center-cube (fcc) or (211) and (100) in body-center-cube (bcc) crystals. Therefore, the atoms like to prefer composition low surface-free-energy facets. There has a simple case of faceting from W (111) surface shown in Fig 1.1.2. Since bcc (211) plane surface-free-energy is lower than (111). The flat W (111) surface [Fig 1.1.2 (a)] will faceting much pyramids [Fig 1.1.2 (b)] after annealing, which three prism faces formed by (211), (112) and (121) planes of each pyramids. So, if the bcc (111) plane can be form on the distal end of the tip [Fig 1.1.2 (c)]. This will form a pyramid, even a SAT [Fig 1.1.2 (d)]. However, we can use this self-assembly property to make a SAT. Early, the SAT of tungsten (W) [5, 6] and iridium (Ir) [7] tips induced by noble metals- (Pd, Pt, Rh and Ir) covered and oxygen gas after annealing, relatively. In our system, we found Pd, Pt, Rh covered Mo tip are easy induce faceting to SATs after appropriate annealing temperature. There Mo {111} surface were replaced by {211} facets and form two types of pyramidal structure with the. 5.
(13) number of atoms constituting their three top layers of 1, 3 and 10 or 1, 6 and 15, respectively. Interestingly, we did not obtain single atom tips by pure Mo and Ir/Mo system.. Fig. 1.1.2. The formation of {211} facets on a flat crystal: (a) flat (111) surface, (b) much pyramids faceting on the surface, and on a pherical crystal: (c) (111) plane on dome of the tip, (d) {211} facet gradually replaced other plane. [4]. The final part is about the study of the missing-row of Pt surface by FIM system. The “missing-row” reconstruction of noble metal (110) surfaces has been the subject of much theoretical and. 6.
(14) experimental attention [8 - 14]. The (1 × 2) missing-row reconstruction is found on clean Au, Pt, and Ir. Of course, the Pt (110) (1 × 1) reconstruction to (1 × 2) missing-row structure is easy formed by annealing to 450 K as same find in our experiment. Through the removal of every other row, the flat (110) surface is converted into a sequence of tiny (111) facets, which are favored by the lower (111) surface free energy relative to that of (110). Because Pt is fcc structure, it is consistent described hereinbefore with the (111) plane should be a minimum energy than other. However, some high index plane would like rearrange to lower surface-free-energy facet of fcc (111) when the atoms obtained activation energy, like Pt (331) plane.. a. b. Fig. 1.1.3. (a) The model is original fcc (311) - (1 × 1) plane. (b) The model is saw-tooth of fcc (311) - (1 × 2) structure after reconstruction.. 7.
(15) In Pt (331), the saw-tooth model is the focus of our research. It is two layers (1 × 2) structure formed by Pt (331) - (1 × 1) after annealing. as shown in Fig. 1.1.3. Earlier, there are two models of the reconstruction have been proposed. One’s the one-time process of rearranged of saw-tooth model. In the other is two-times processes of double missing- row model. We will clarify which model is consistent with experimental results.. 8.
(16) References [1] W. Auwärter, K. Seufert, F. Klappenberger, J. Reichert, A. W. Bargioni, A. Verdini, D. Cvetko, M. D. Angela, L. Floreano, A. Cossaro, G. Bavdek, A. Morgante, A. P. Seitsonen, and J. V. Barth, Phys. Rev. B 81 (2010) 245403. [2] K. Comanici, F Buchner, K Flechtner, T Lukasczyk, J.M. Gottfried, H. Steinrueck and H. Marbach, Langmuir 24 (2008) 1897–901. [3] L. Scudiero, D. E. Barlow and K.W. Hipps, J. Phys. Chem. B. 104 (2000) 11899–905. [4] A. Szczepkowicz, A. Ciszewski, R. Bryl, C. Oleksy, C.H. Nien, Q. Wu, T. E. Madey, Sur. Sci. 599 (2005) 55. [5] H.S. Kuo, I.S. Hwang, T.Y. Fu, Y.C. Lin, C.C. Chang, T.T. Tsong, Jpn. J. Appl. Phys. 45 (2006) 8972-8983. [6] T.Y. Fu, L.C. Cheng, C.H. Nien, T.T. Tsong, Phys. Rev. B. 64 (2001) 113401. [7] I. Ermanoski, K. Pelhos, W. Chen, J.S. Quinton, T.E. Madey, Sur. Sci. 549 (2004) 1-23. [8] K.M. Ho, K.P. Bohnen, Phys. Rev. Lett. 59 (1987) 1833. [9] K.M. Ho, K.P. Bohnen, Europhys. Lett. 4 (1987) 345. [10] Marcel den Nijs, Phys. Rev. Lett. 66 (1991) 907. [11] M. Sturmat, R. Koch, K.H. Rieder, Phys. Rev. Lett. 77 (1996) 5071. [12] C. Hbfher, J.W. Rabalais, Phys. Rev. B. 58 (1998) 9990. [13] J.C. Campuzano, M.S. Foster, G. Jennings, R.F. Willis, W.N. Unerti, Phys. Rev. Lett. 54 (1985) 2684. [14] J.C. Campuzano, A.M. Lahee, G. Jennings, Surf. Sci. 152–153. 9.
(17) (1978) 265.. 10.
(18) Chapter Two. Experimental Techniques and Methods In this chapter we will first mention about the basic principle of STM and FIM. Subsequently, I will briefly introduce the instrument with the experimental procedure.. 2.1 Instrumentation 2.1.1 Ultra-high vacuum system The STM and FIM experiments were conducted in an ultrahighvacuum (UHV) system. Our system needs to connect different working range of the pumping system. The pumping system is consisting by a dry pump, a turbo pump, a titanium sublimation pump (TSP), and an ion pump. The base pressure of this vacuum system is 1×10-10 torr. The dry pump is used first to lower pressure in the vacuum chamber to ~10-3 torr. Then the turbo pump automatically starts to lower the pressure to the 10-7 torr range. At this lower pressure, the ion pump turns on. As the pressure drops to ~10-7, we start to bake. 11.
(19) the chamber at about 120 °C for over 12 hours (FIM) or longer (STM). After the chamber cools down to RT, we gain the ultra-high vacuum about 1 × 10-10 torr.. 2.1.2 UHV LT-STM system All experiments completed in ultra-high vacuum (UHV). USM 1400 system is a low-temperature STM in an UHV chamber. Fig. 2.1.1 (a) is the picture of our system, and (b) is STM part inner the main chamber. This STM system has three separated UHV chambers. With the combination of liquid helium and nitrogen cryostat and three layers metal shielding outside the STM part, this cryostat is constructed as two tanks so that it can save liquid helium consumption. The outer tank is for liquid nitrogen, while the inner tank is for liquid helium. However, the operating minimum temperature can be lowered to 2 K. But, our all experiments conducted at 77 K. A prepare chamber isolated from the STM chamber, because it can let us prepare the sample and not interfere the main chamber. The preparation processes include cycles of sputtering and annealing for the substrates and deposition of target materials. Thus we have a sputtering system, an annealing system, and deposition systems in our prepare chamber. Finally, a load-lock. 12.
(20) chamber isolated from prepare chamber can let us charge sample between atmosphere and UHV.. Fig. 2.1.1. (a) The photo image for USM- 1400S STM. Each arrow represents one of the components on the machine. (b) Major part of the STM.. 13.
(21) 2.1.3 Principle of scanning tunneling microscopy (STM) Scanning Tunneling Microscope (STM) was invented by G. Binnig and H. Rohrer in 1981 [1]; they shared the 1986 Nobel Prize in Physics for their invention. With the abilities of real-space surface image, atomic resolution and local density of states (LDOS) electrical analysis, STM has been widely used in many fields, such as condensed-matter physics, chemical and biology.. Fig. 2.1.2. Model is the tunneling effect in the STM. Rectangular potential barrier and particle wave function U(z). The energy of the tunneled particle is the same but the amplitude is decreased.. 14.
(22) In classical physics an electron cannot penetrate into or across a potential barrier if its energy E is smaller than the potential Ψ(z) within the barrier. But, quantum mechanics treatment predicts an exponential decaying solution for the electron wave function in the barrier, the tunneling model shown in Fig. 2.1.2. For a rectangular barrier we get. In quantum mechanics, we know the Schrödinger’s equation [2] describes an electron with energy E moving in a potential U(z). ħ d2 − 𝛹(𝑧) + 𝑈(𝑧)𝛹(𝑧) = 𝐸𝛹(𝑧) 2𝑚 d𝑧 2. (2.1.1). For a rectangular barrier we get 𝛹(𝑧) = 𝛹(0)𝑒 −𝜅𝑧. (2.1.2). √2m(𝑈 − 𝐸) ħ. (2.1.3). where 𝜅=. The probability of finding an electron behind the barrier of the width d is 𝑊(𝑧) = |𝛹(𝑧)|2 = |𝛹(0)|2 e−2ĸ𝑧. (2.1.4). In scanning tunneling microscopy a small bias voltage V is applied so that due to the electric field the tunneling of electrons results in a tunneling current I. The height of the barrier can roughly. 15.
(23) be approximated by the average workfunction of sample and tip [3]. U=. 1 (𝑈 + 𝑈𝑡𝑖𝑝 ) 2 𝑠𝑎𝑚𝑝𝑙𝑒. (2.1.5). If the voltage is much smaller than the workfunction eV << U, the inverse decay length for all tunneling electrons can be simplified to 𝜅=. √2m𝑈 ħ. (2.1.6). The current is proportional to the probability of electrons to tunnel through the barrier: 𝐸𝐹. 𝐼∝. |𝛹𝑛 (0)|2 e−2ĸ𝑧. ∑. (2.1.7). 𝐸𝑛 =𝐸𝐹 −𝑒𝑉. Therefore, the current markedly changes as the distance varies. The STM can do measurements of surfaces up to atomic scale by utilizing this property. Besides, before the exponential component, there is another component which is also can modify the tunneling current. This component is related to the LDOS [3]. The LDOS at a location z with the energy E for a sufficiently small є → 0 is defined as: 𝐸. 1 𝜌(𝑧, 𝐸) ≡ ∑ |𝛹𝑛 (𝑧)|2 є 𝐸𝑛 =𝐸−є. 16. (2.1.8)..
(24) The LDOS shows the number of electrons per unit volume per unit energy. From Eq. 2.1.2, Eq. 2.1.7, and Eq. 2.1.8, tunneling current can also be expressed as: 𝐼 ∝ 𝑉𝜌(𝑧, 𝐸). (2.1.9).. The LDOS can be obtained by calculating the derivative of the current: d𝐼 ∝ 𝜌(𝑧, 𝐸) d𝑉. (2.1.10).. Therefore we can get the LDOS of the sample by doing the scanning tunneling spectroscopy (STS). Here is the measurement of change in current as a function of applied voltage with the tip held at a constant height. Since dI/dV is proportional to the density of states of the sample, p, the derivative of this curve corresponds to the LDOS. STM has become a key tool for the investigation of superconducting materials by combining the high spectroscopic energy resolution possible in tunneling measurements with the high spatial resolution a scanning probe technique can provide.. 17.
(25) 2.1.4 UHV FIM system Field ion microscope (FIM) was invented by Erwin E. Müeller in 1951 [4] at the Pennsylvania State University. The FIM invention was first time to observe atoms by human. It was developed from its forerunner, the field emission microscope (FEM) [5]. The instrument features is a sharp tip sample mounted on an electrically insulated stage, and that will be cooled to cryogenic temperatures (20 to 80K) in a ultrahigh vacuum chamber shows in Fig. 2.1.3 (a). The field ion image of the sample is formed on a micro-channel plate and phosphor screen assembly that is positioned approximately 10 cm in front of the sample, instrumentation Diagram shows in Fig.2.1.3 (b). To produce a field ion image, controlled amounts of image gas are admitted into the vacuum system. The type of image gas used depends on the material under investigation; common images gases are neon, helium, hydrogen. The FIM imaging mechanism shows in Fig. 2.1.4. When high electric field is applied on a sharp surface of a tip, and imaging gas atoms are ionized at the protruding sites on the sample surface [6, 7]. These ions hit screen along the electric field line, forming brightly imaging spots. Hence, the FIM is a projection type microscope of an atomic resolution with an approximate magnification of a few million. 18.
(26) a. b. Fig. 2.1.3 (a) Schematic diagram of detail of the sample holder, sample loop and cooled cryostat contact. (b) The photo image for FIM. Each arrow represents one of the components on the machine.. times. By applying higher electric filed, surface atoms of the specimen are also ionized. This high electric field produced evaporation phenomenon is usually called field evaporation if the surface atoms are lattice atoms, and is called field desorption if they are adsorbed atoms. By this field evaporation process, it can be. 19.
(27) desorption layer by layer of atoms on surface. So, the field evaporation mechanism is easily to clean the shape surface of a tip. Field evaporation is a field induced process which involves the removal of atoms from the surface itself at very high field strengths and typically occurs in the range 2-5 V/Å . Atoms always evaporate from the surface, so the special resolution in the depth direction is a mono-atomic layer. A unique feature of the atom probe compared with the other analytical instrument is its extremely high spatial resolution and the equal detection efficiency for light elements.. Fig. 2.1.4 Schematic diagrams the imaging mechanism of FIM.. 20.
(28) 2.1.5 Principle of FIM system In 1928, Oppenheimer observes the field ionization phenomena formed by quantum tunneling effect [8]. However, this inference was confirmed in FIM experiments by E.W. Müller in 1951. In free atoms, the electrons are trapped in the potential wells in an atom shows in Fig. 2.1.5 (a).The atoms ionized needs us to give the energy greater than “I” of the inner electron. But the potential will be changed when the atoms placed in an electric field shows in Fig. 2.1.5 (b). Therefore, we can use the Wentzel-Kramer-Brillouin (WKB) to calculate the probability of tunneling out of barrier:. 8m 12 x2 1 DE ,V ( x) exp 2 V ( x) E 2 dx x1 . (2.1.11). E : Total energy of eletron V(x) : Barrier m : Eletronic mass : Planck' s constant divided by 2π x1 , x 2 : The initial position and end position of the tunneling electron. When this atom gas is close to the metal surface, the Eq. 2.1.11 with an electron potential approximate formula:. 21.
(29) V ( x) . e2 e2 e2 eFx xi x 4 x xi x. (2.1.12). F : Electric field intensity x : The distance between the electrons and metal. x : The distance between the ions and the metal. i First part : Coulomb electrostatic potential between electron and ion. Second part : The obtained energy of x position from metal of the electron in electric field. Third and fourth parts : Consider Coulomb electrostatic of image charge.. Figure 2.1.5 (c) show the barrier width narrows as the atom gas close to metal surface which critical value xc. This value determines whether occur the atomic electron tunneling. However, the energy of the electron from the gas atom must coincide with, or be higher than, the lowest available conduction level in the metal, which is close to the Fermi level. If this condition is not fulfilled, there are no vacant energy levels in the metal available for the tunneling electron [6, 9, 10, 11–16]. After considering the relevant conditions can be obtained the formula:. e2 1 eFxc I ' ( a i ) F 2 I 4 xc 2 αa : Atomic polarization α : Polarization of ions has been formed. i I': Critical distance equivalent Ionization energy. φ: The work function of the metal surface. I : Ionization energy.. 22. (2.1.12).
(30) a. b. c. Fig. 2.1.5 The principle of field ionization.. 23.
(31) 2.2 Experimental procedure 2.2.1 Tip preparation Both the STM scanning tip or FIM tip samples were produced using etching. Making a good tip is an important issue in the STM and FIM experiments. In our experiments, tip is prepared by electrochemical etching method. In STM, we used polycrystalline tungsten wire with 0.5 mm in diameter and 99.9% purity is chosen to make a STM tip. The tungsten wire is etched in the 5M NaOH solution by applying a direct current (DC) voltage 7V. In FIM, we used polycrystalline Mo and Pt wire with 0.125 mm in diameter and 99.98% purity is chosen to make a FIM sample. The Mo sample is like to tungsten etching used by NaOH (or KOH). But the Pt is more difficult, the electrolyte is heated to molten state of the 4:1 proportion NaNO3 (s) and NaCl (s) mixture. The tip’s shape is checked by an optical microscopy (OM) before transferred the tip to the UHV chamber.. 2.2.2 Experimental procedure of STM and FIM There we use a simple experiment flowchart illustrating procedure in our experiments, STM past shows in Fin. 2.2.1 and FIM in Fig. 2.2.2.. 24.
(32) Fig.2.2.1 STM experiments flowchart.. 25.
(33) Fig.2.2.2 FIM experiments flowchart.. 26.
(34) Reference [1] G. Binnig, H. Rohrer (1986). "Scanning tunneling microscopy". IBM Journal of Research and Development 30: 4. [2] Richard L. Liboff, Introductory Quantum Mechanics (Addison Wesley, New York, 1980). [3] C. Julian Chen, Introduction to Scanning Tunneling Microscopy (Oxford University Press, New York, 1993). [4] E. W. Müller, Z. Physik, 131 (1951) 136. [5] E. W. Müller, Z. Physik, 106 (1937) 541. [6] J.R. Oppenheimer, Phys. Rev. 32 (1928) 361. [7] E.W. Müller, J. Appl. Phys. 27(5) (1956) 474–476. [8] E.W. Müller, Phys. Rev. 102(3) (1956) 618–624. [9] E.W. Müller, T.T. Tsong, Field Ion Microscopy, Principles and Applications (Elsevier, New York, NY, 1969) [10] T.T. Tsong, Atom-Probe Field Ion Microscopy: Field Emission, Surfaces. [11] R. Gomer, Field Emission and Field Ionisation (Havard University, Cambridge, 1961) References 27. [12] D.G. Brandon, Philos. Mag. 7(78) (1962) 1003–1011. [13] D.G. Brandon, Br. J. Appl. Phys. 14(8) (1963) 474. [14] E.W. Müller, Acta Crystallogr. 10(12) (1957) 823–823. [15] E.W. Müller, Science 149(3684) (1965) 591–601 [16] T.T. Tsong, Surf. Sci. 70 (1978) 211. 27.
(35) Chapter Three. Study of self- assembled of Co-TPP on Pb/Si (111) surface by STM 3.1 Introduction The bottom up approach, where assembling basic units (atoms or molecules) into larger nanostructures on surfaces, is of great recent interest both on fundamental researches and technological applications. In particular, molecules have clearly demonstrated their potential to be building units for practical applications, such as, organic light emitting diodes (OLEDs), organic photovoltaic solar cells, and organic fieldeffect transistors (OFETs). At the nanoscale, the controlled assembly of highly organized architectures on surfaces using functional molecules is highly demanded. The central challenging is how to well perceive and manipulate the interactions between molecules and supporting substrates. Due to their wild applications both in Chemistry and life science, polyenic and aromatic compounds have been studied as the molecular layers adsorbed on various surfaces. Within this class, porphyrins and metalloporphyrins are preferentially investigated; as these compounds. 28.
(36) are very robust, polyfunctional, volatile and readily available with a range of different electronic and steric properties. For sub-monolayer studies, moreover, they are perfectly suitable because they can be prepared on surfaces in vacuum, which can be explored by a high-spatial resolution local probe technique, scanning tunneling microscopy (STM). Many STM studies of tetraphenylporphyrin (TPP) molecules adsorbed on various metallic surfaces have been carried out recently. The interaction of the phenyl legs with the surface is found to play a key role in their physical properties, from molecular adsorption conformations to growth of larger molecular complexes [1 - 3]. By replacing phenyl substituents, the adsorption property of TPP molecules varied from forming highly ordered stacking molecular structures to staying as an isolated molecule [4, 5]. On the contrary, efforts have been also made by replacing the H2 center with metallic ions, no influence on the adsorption but only on electronic properties were reported [6, 7]. One can therefore consider an alternative way: to modify the surface property in order to obtain desired molecular structures. We therefore propose to place an atomic layer of “soft” material between molecules and surfaces as a tuning layer. By properly controlling the structure of the tuning layer, the properties of. 29.
(37) molecules are expected to be modulated accordingly. In this study, cobalt (II) - tetraphenylporphyrin (Co-TPP) molecules were first time prepared on one monolayer (ML) of Pb on Si (111) surface. We chose Pb/Si (111) system because it has been well studied for decades and most importantly, various surface reconstructions can be prepared with Pb coverage’s, annealing history and temperatures [8 - 12]. This is then a perfect tuning layer for us to learn how Co-TPP molecules interacting with substrates. At a Pb coverage of about 1 ~ 1.2 ML, Co-TPP molecules formed short-ranged ordered domains where three adsorption conformations were observed. With a Pb coverage of about 1.3 ML, surprisingly, the orientation of molecular domains changed and arranged along with one of the Pb surface directions. On this surface, only two adsorption conformations were observed and a preferred adsorption site was found.. 3.1.1 Molecule H2-TPP and Co-TPP Tetraphenylporphyrin, abbreviated H2-TPP, is a synthetic heterocyclic compound that resembles naturally occurring porphyrins, which form the base structure for both heme and chlorophyll and vitamin B12, as well as a variety of other hemoproteins consisting of porphyrin combined with metals and proteins. The name of porphyrin. 30.
(38) comes from the Greek word porphyra, which means purple. Porphyrins absorb light in the visible spectrum and have been used as dyes. The study of naturally occurring porphyrins is complicated by their low symmetry and the presence of polar substituents. Tetraphenylporphyrin is hydrophobic, symmetrically substituted, and easily synthesized. The porphyrin can coordinate with many metal ions (such as Fe, Co, Cu, and Zn) to form stable metal–porphyrin coordination complexes, which are widely existed in nature with special physiological activity and excellent properties in electron transfer, energy conversion, and nonlinear optical-limiting [13], [14] and [15]. There has a H2-TPP model show in Fig. 3.1.1, it like a macrocyclic conjugated organic molecules with four phenyl rings bonding to single bond. There two hydrogen atom in the center of the macrocycle, it can be easily replaced with a metal atom becomes metal- tetraphenylporphyrin.. 31.
(39) Fig. 3.1.1 Illustrate the molecule models of the tetraphenylporphyrin.. 3.1.2 Si (111) The Si (111) is chosen as the sample in the experiments. The silicon atom has four bonding possibilities, and the crystal structure of silicon is a diamond cubic lattice, as shown in Fig. 3.1.2 (a). The sample is cut from a 4” Si (111) wafer. The miscut angle is 0.1∘and the dimension of the sample is 16 mm × 2 mm × 0.3 mm. Before transferring to the UHV chamber, the Si sample is cleaned by ultrasonic vibration in acetone, methylalcohol and deionized water for 10min, respectively.. 32.
(40) The (111) - orientation for samples results in an extremely flat surface that is suited well for the purpose of this project. Figures 3.1.2. (b) and (c) show the structure of the (111) layers [16]. If this is the configuration of the atoms at the surface, it is denoted as a 1 × 1 surface.. a. b. c. Fig. 3.1.2 The basic Si unit cell and (111) planes. (a) Si unit cell. Diamond cubic lattice. (b) Si (111) surface, side view. (c) Si (111) surface, top view.. 33.
(41) The clean Si (111) - 7 × 7 surface is prepared by high temperature annealing from 900 K. The Si (111) - 7 × 7 is the thermal stable equilibrium structure at the end of the reaction pathways. The surface structure is observed with 12 adatoms and large holes at the corners of the Si (111) - 7 × 7 unit cells from STM image. Further, the unit cell is divided into two equilateral triangles called the faulted half unit cells and unfaulted half unit cells resulting in unbalanced density of electron states at the special tunneling conditions. The Si (111) - 7 × 7 reconstruction is described by the model of the DimerAdatom-Stacking-Fault (DAS) family of reconstructions referred to as the 7 × 7 DAS model as shown in the model of Fig. 3.1.3 [17, 18]. The unfaulted domain means that the atomic stacking is following the sequence of A - B - C and so on in the direction normal to the surface as indicated by the half cell on the right side of the model, but the faulted domain is not allowing the stacking arrangement in the left half cell of the model. The unbalanced contrast between the faulted and unfaulted domains in the same unit cell is due to the DOS feature that the faulted stacking domain is observed resulting in an only chlorine-terminated rest-adatom layer without adatoms.. 34.
(42) a. c. b. Fig. 3.1.3 Atomically clean surface of Si (111) - 7 × 7. STM images of (а) filled and unfilled (b) electron states of surface; (c) schematic representation of surface (plan and side views) in accordance with Takayanagi DAS model (dimer- adatom- stacking fault) [18]. Yellow circles represent Si atoms, red circles - dimerizated Si atoms, and blue - second layer Si rest-atoms. Elementary cell 7 × 7 is highlighted with a diamond. Half of the elementary cell with package defect is marked as FH (faulted half); the half with no package defect is marked as UH (unfaulted half). You may see that the half of the cell with package defect depicted in the STM image of filled states (a) appears brighter than the other half. Maximums in the STM image correspond to adatoms.. 35.
(43) 3.1.3 Pb phase on Si (111) surface The Pb phase on Si (111) surface is a complicated system. There have various phase structures can be formed on the Si (111) substrate depending on the coverage and different thermal treatment [19 - 22]. In this thesis, the different phase of Pb interlayer face are occupies an important impact of the molecules self-assembly. We will briefly illustrate the different structural phase for 1 ~ 1.3 ML coverage Pb on Si (111) system. We first prepared Si (111) - 7 × 7 surface with deposition 1 ML Pb atoms (1ML = 7.84 × 1014 atoms/cm2) at room temperature, followed by annealing at 600 K for a few seconds (about 3 ~ 4 sec). After that, we often could see the Pb/Si (111) - 1 × 1 phase. At low temperature, the Pb/Si (111) - 1 × 1 phase is transformed to row- like √7 × √3 phase. Next, we added more Pb atoms onto the above first-stage deposition sample at room temperature. It could be called second-stage deposition. Then we can be found Pb/Si (111) - HIC or SIC phase when the Pb coverage increase to 1 ~ 1.3 ML. The detailed phase structure of Pb on Si (111) substrate will be described as follows.. 36.
(44) e. Fig. 3.1.4 (a) Room-temperature STM topograph showing regions of the 1 × 1 phase surrounded by Si (111) - 7 × 7. (Info: 2.0 V, 0.15 nA). Red scale bar: 5 nm. (b) High-resolution image of the 1×1 phase. Blue scale bar: 1 nm. (c) STM image of √7 × √3 domains at 190 K. (Info: 2 V, 0.15 nA) (d) High-resolution image of the √7 × √3 phase [10]. (e) Illustrate the atomic models of the 1 × 1 and √7 × √3 transition model.. 37.
(45) A. 1 ML coverage of the Pb/Si (111) - 1 × 1 phase, RT After the above procedures of sample preparation, the Pb atoms destroy Si (111) - 7 × 7 and rearrange to form Pb/Si (111) - 1 × 1 phase [23] surrounded by Si (111) - 7 × 7 at room temperature, as shown in Fig. 3.1.4 (a). The Pb- covered regions is mainly a 2D Pb monolayer structure on Si (111) substrate. Fig. 3.1.4 (b) shows the atomic resolution image which can be taken at low bias for both polarities. The image exhibits a hexagonal 1 × 1 structure with a small corrugation of ~ 0.2 Å . The Pb coverage of Pb/Si (111) - 1 × 1 phase has been determined to be about one monolayer by Rutherford backscattering (RBS) experiments [23 - 25]. The atomic model for 1 × 1 phase is drawing in left side of Fig. 3.1.4 (e); the Pb atom is adsorbed on the Si adatom (T1 site).. B. 1 ML coverage of the Pb/Si (111) - √7 × √3 phase, LT When the sample is cooled to low temperature, the Pb/Si (111) 1 × 1 structure is transformed reversibly into a low- symmetry rowlike √7 × √3 structure [26 - 28]. Figure 3.1.4 (c) shows the typical image of Pb/Si (111) - √7 × √3 surface structure. The image exhibits that the Pb- covered regions have the stable row-like structure with {-211} orientations, as expected from the substrate symmetry. The. 38.
(46) atomic resolution image, as shown in Fig. 3.1.4 (d), demonstrates the trimer row-like structure. Figure 3.1.4 (e) illustrated the phase transition atomic model of Pb/Si (111) - √7 × √3 phase. In this model, some atoms are displaced from their T1 sites slightly to H3 and some still on the T1 sites. Therefore, the √7 × √3 structure can be seen as a distorted 1 × 1 structure and the coverage of Pb of the √7×√3 structure is also 1 ML.. C. 1~1.2 ML coverage of the Pb/Si (111) - HIC phase, LT In particular after the second-stage deposition, two different phases have been discussed extensively: the HIC (hexagonal incommensurate phase) and the SIC (striped incommensurate phase) [29]. When the increase the Pb coverage to 1 ~ 1.2 ML, the surface will change to HIC phase in the monolayer. Based on STM images of Fig. 3.1.5 (a), it was suggested that these phases are six-fold degenerate, which implies that Pb can occupy two different binding sites (i.e., H3 and T4 sites) since the symmetry of the √3 × √3 unit cell is only three-fold. The HIC structure attributed to the added Pb atoms squeeze inner to monolayer. Therefore, this stress caused by like-trimer pattern of Pb clusters on H3 and T4 sites in a triangle domain and the domain walls formed by dimer on B2 site. The HIC. 39.
(47) model is show in Fig. 3.1.5 (b). Ex: according to the coverage are 1 ~ 1.2 ML, so the trimer pattern should form by four atoms (tetramer) [30]. Our model simply presents the STM image. We do not participate in the discussion of this result of trimer or tetramer.. 40.
(48) a. b. Fig. 3.1.5. (a) STM image is the Pb/Si (111) - HIC phase. The two different triangle domains mean the Pb atoms trimer (tetramer) close on H3 and T4 sites [31]. (b) Illustrate the atomic models of the HIC model. There marks by H3 and T4 site domains and a √3 × √3 unit cell.. 41.
(49) D. 1.3 ML coverage of the Pb/Si (111) - SIC phase, LT When the Pb coverage increase to 1.3 ML [30, 32], the monolayer surface will be rearranged to Pb/Si (111) - SIC phase. Based on STM images of Fig. 3.1.6 (a), it phase pattern look like striped along to {-211} orientations, each striped line is a domain of H3 or T4. Similarly, like the HIC phase, that Pb can occupy H3 and T4 two different binding sites since the symmetry of the √3 × √3 unit cell. High-resolution image shows in Fig. 3.1.6 (b), the different direction trimer pattern of Pb clusters are adsorbed on H3 and T4 sites and the domain walls formed by dimer on B2 site. The SIC model is show in Fig. 3.1.6 (c). In this model, we can simply distinguish Pb clusters adsorbed on which site. Due to the Pb clusters are arrangement along the direction [-211], like clusters line. However, we model show four lines on H3 site and five lines in T4 site of the Pb trimer with dimer cluster between there. Interestingly, the phase call incommensurate is mean not very orderly phase. So we often can to found random distribution of the different number of Pb cluster lines on H3 or T4 domains.. 42.
(50) a. b. c. Fig. 3.1.6. (a) Over view the STM image of Pb/Si (111) - SIC phase, the striped line parallel the [1-21] direction (Info: 2.0 V, 1.0 nA). (b) High-resolution image of SIC phase (Info: -0.05 V, 0.2 nA) . It show the different direction trimer on H3 and T4 sites marks by arrows. (c) Illustrate the atomic models of the SIC model. There marks by H3 and T4 site domains and a √3 × √3 unit cell.. 43.
(51) 3.2 Disorder result of Co-TPP adsorbed on Si (111) Originally, we want to study what kind of mechanism behavior or mechanism of the molecules adsorbed on the semiconductor. In this case, we direct deposited Co-TPP on the Si (111) - 7 × 7 surface. The molecules are evaporated on substrate at room temperature then STM images are scanned after cooling to 77 K. According to the experimental results, the molecules adsorbed on the surface are chaotic and disorderly. However, we still can to sort out some different adsorption mode of Co-TPP on Si (111) - 7 × 7 surface. Figure 3.2.1 (a) shown a few Co-TPP have a tendency to adsorb on corner hole of the 7 × 7 surface. A benzene ring of Co-TPP is found to locate in corner hole easily. Because Si (111) substrate is three symmetrical, these molecules with four-fold symmetry have six orientations which were leaning against corner hole on the surface. The heights of the Co-TPP’s four phenyl legs are different. The results indicate that the molecule is not parallel to the plane. Black arrows indicate six different direction of the Co-TPP. In such an event, the adsorption probability of each direction of Co-TPP is substantially the same, therefore no significant difference between faulted and unfaulted position site as a matter-of-course. Figure 3.2.1 (b) shows the six orientations adsorbed model of the Co-TPP on. 44.
(52) corner hole of the Si (111) - 7 × 7 surface.. F UF. a. b. Fig 3.2.1. (a) The Co-TPP tendency adsorbed on corner hole site, and there has six orientations formed on the Si (111) - 7 × 7(Info: V = -2.0 V, I = 60 pA). Black arrows indicate six different direction of the Co-TPP. (b) This is six models of the molecule adsorbed on corner hole.. In addition, there a few molecules will be lying on the surface, shown in Fig. 3.2.2 (a). STM image shows a Co-TPP lying on the triangle of faulted, site which indicated by an arrow. Whether adsorbed on faulted or unfaulted sites, the behaviors are similar. As the interaction between molecule and triple symmetry substrate, the Co-TPP has three orientations adsorbed on a site. Figure 3.2.2 (b). 45.
(53) shows the model of Co-TPP adsorbed in plane, and it have three directions at faulted and unfaulted site, respectively.. F UF. a. b. Fig 3.2.2. (a) There is a Co-TPP adsorbed on faulted site, and parallel the plane (Info: -2.00 V, 24.8 pA). (b) The model shows the molecule adsorbed on faulted and unfaulted site.. As the semiconductor have many dangling-bonds on the surface. We speculated that molecules and dangling-bonds have a strong interaction and they may even form bond. Forasmuch as the interaction, molecules are difficult to move on the surface. According to the results, we could not find molecular islands are formed by self-assembly on Si (111) - 7 × 7 surfaces. So, the surface will be. 46.
(54) disorderly when we increase the coverage of Co-TPP, shown in Fig. 3.2.3. Therefore, we choose to deposite 1 ML Pb as a buffer layer between the molecular and Si (111) surface.. Fig 3.2.3. The disorderly pattern is high coverage of Co-TPP result. We cannot to found any self-assembly islands as the dangling-bond. (Info: -2.0 V, 56 pA). 47.
(55) 3.3 Geometry transition of Co-TPP on Pb/Si (111) - √7 × √3 (or HIC) depend temperature. Since molecules cannot move on Si (111) surface, so we deposited Pb as a buffer layer. The buffer layer has lower activation energy of diffusion makes the molecules can move easily on the surface. Since the Pb/Si (111) system has many different phases, therefore the molecule-molecule and molecule-substrate interactions are important issues for the formation of highly ordered Co-TPP molecular films. Consequently, we get many special experiment results about the molecules self-assembled to order islands on the Pb/Si (111) surface. However, we will discuss molecule selfassembly on Pb/Si (111) at room and low temperature in this section.. 3.3.1 Self-assembled Co-TPP arrays: self-assembly on Pb/Si (111) at room temperature Firstly, we discuss room temperature case of the Co-TPP molecules adsorbed on Pb/Si (111) - 1 × 1 (or HIC) surface. Whereas molecules were easy mobile on metallic film at room temperature, the molecules are susceptible to move by the scanning tip. So, we need to restrict the movement of molecules by increased Co-TPP coverage approx. 1 ML. Since the system has a large thermal drift,. 48.
(56) there we temporary need not to discuss the rotation angle, relationship between molecular islands and Pb/Si (111) substrate in this section. Figure 3.3.1 (a) and (b) shows three STM images of the short range ordered structure resulting from Co-TPP deposition on Pb/Si (111) - 1 × 1 (or HIC) at 300 K. There (a) and (b) are the filled and empty states results in same area, the black arrow mark the main same defect. We identify A and B distinctively the main different areas. Three different geometries of Co-TPP are found on the surface. The difference brightness of these molecule types just from by electronic structure of LDOS. Three distinct conformations are identified type I, II and II+ mark at rad circles. The type I and II patterns were look to be a dumbbell and four-leaf-clover. However, the areas A and B are constituted by type I and II of Co-TPP, respectively. There have used four Co-TPP models mark a unit cell of molecular island for area A and B formed filled and empty states, severally. In the boundary between A and B, we can see that the exchanged of the type I and II in before Fig. 3.3.1 (a) and after (b) images. The type I mark from green arrows and blue pointing type II. Since the type I and II are arrangement in same layer and we consider the. 49.
(57) distinct conformations of Co-TPP were adsorbed on different site. However, this conformation transformed may be formed by the substrate change. Since the Pb atoms are very easy shift from the original site in the room temperature. So, we believe these types transform are due to the interface changes by Pb atoms shift. However, there is not exchange position by two molecules of type I and II in this case. Usually, the different domains orderly formed by distinct types in molecular island, this boundary have same orderly condition matching arrangement between two types. Another wall has a dislocation of boundary between two domains, the two domains formed by two different arrangement directions molecules. Figure 3.3.2 showing the two different domain walls. We have used molecular model to marks which a molecule, unit cell and boundaries position. One wall is the matching arrangement between type I and II shows in the right side and mark at black arrow. This is also reflected in the Fig. 3.3.1 (a) and (b). The dislocation boundary was shown in image left side and mark at white arrow. The green line and arrow to marked dislocation interval of boundary of two domains. There shows the molecules are turn life and right arranged formed boundary which not same period direction of two molecular domains.. 50.
(58) a. b. Fig. 3.3.1. The two images are same area scanning images of (a) filled and (b) empty state. We then assign molecules in domain A and B are formed by type I and II, respectively. Three main visible types. 51.
(59) are mark in red circle. We simply used four molecules to mark a unit cell in molecular layer. Since the two images are successive scanning images, we can found the type transform by I to II and II to I in before and after images. The green and blue arrows marks the type I and Type II. [Info: (a) -2.0 V, 0.38 nA; (b) 2.0 V, 0.38 nA]. Fig. 3.3.2. There have two different boundary shown in image. Black arrow mark matching boundary formed by homogeneous period of two domains. White arrow mark dislocation boundary formed by different direction of distinct molecular domains. Molecule model marks a single, unit cell and boundary position. The green line and arrow to marked the dislocation interval of two domains. (Info: -2.0V, 0.60 nA). 52.
(60) Type II+ is a special conformation and very small proportion in the surface, it pattern look to be a Ninja dart formed in filled states shows in Fig. 3.3.1(a). We usually can be found type II+ appear on different location in successive two images shows in the Fig. 3.3.3 (a) and (b). So, the Type II+ likes molecule diffusion on the molecular island. Whereas the type II+ apparent height obtain altitude difference is 0.04 nm in the line profile shows in the Fig. 3.3.3 (c), it’s very less than 0.3 nm of a Co-TPP when the system experiment at low temperature. Moreover, it coupled with type II+ geometry same as type II in empty states shows in the Fig 3.3.1 (b), it’s not adsorbed a molecule on the top. We consider two possibilities: one is the orbital charged by a heterogeneity atom adsorbed on a type II molecule. Another is the difference formed by a small angle of rotation of Co-TPP. So, the type II+ likes molecules diffusion on the molecular island in these successive patterns.. 53.
(61) a. b. c. Fig. 3.3.3. Two successive scanning images of (a) before and (b) after (Info: -2.0 V, 0.38 nA), we can be found some type II+ appear or disappear in the molecular island. A to E arrows mark the type II+ appear or disappear position. A, D position appears and B, C, E not type II+ in pattern (a). After the A, D disappears and B, C, E appears in pattern (b). (c) is apparent height of the rad line profile of type II+ in the pattern (b).. 54.
(62) 3.3.2 Self-assembled Co-TPP arrays: self-assembly on Pb/Si (111) at room temperature Generally, we understand the behavior of Co-TPP on Pb/Si (111) - 1 × 1 (or HIC) at room temperature. In the following, we have studied the interaction between molecules and the substrate when the system is cooling to 78 K. Firstly we prepared Co-TPP molecules on the 0.5 ML Pb plated- Si (111) surface sample as shown in Fig. 3.3.4. As previously described, the 1 × 1 or HIC phase of Pb film will be transformed into √7 × √3 phase of the substrate surface after cooling below 250 K. The Co-TPP’s formed short - ranged order domains on the surface, where three different molecular domains are identified, as marked by A, B and C. In our case, √7 × √3 and HIC phase have mixed on Pb/Si (111) surface attributed to the impact of the adsorbed molecules. However, we believe different domains of molecular islands are adsorbed on different phases of substrate. Since the Pb coverage on Si (111) is 1 ML, we can easily determine the each direction with the guide by the 7 × 7 unit cell, marked with a yellow rhombus in the upper half of Fig. 3.3.4. At the same time, the structure of √7 × √3 phase can be at guidance [1-21] with green arrows as a mark. However, we can easily to identify the molecular islands rotation 15° relative to the [-12-1] direction. In. 55.
(63) addition one diagonal of the unit cell is aligned with the Si (111) [-110] close-packed directions and another diagonal with the [11-2] directions. Accordingly, six possible domains of Co-TPP assemblies exist and are observed in the experiments. This result of orientation related molecular islands and our substrate is the same as Co-TPP self-assembly on Au (111) [3, 33, 34], Ag (111) [35] and Cu (111) [2, 36, 37]. Those example STM image shows in Fig. 3.3.4 (a) to (b).. b. a. c. Fig. 3.3.4. The Co-TPP self-assembled on (a) Au (111) [3], (b) Ag (111) [35] and (c) Cu (111) [37].. 56.
(64) With a clear inspection of the Fig. 3.3.4, the appearance of molecules exhibits differently among three types of domains. For simply, we then assign molecules in domain A, B and C as type I, II, and III molecules, respectively. Figure 3.3.5 (a) to (c) shows zoom-in STM images of the three conformations of molecules with molecular structures overlaid. The conformation is two-fold symmetric of single molecule of type I and III and quasi-four-fold symmetric of type II. We used molecule model to mark which a single Co-TPP and red dashed box to mark unit cell. In contrast to Fig. 3.3.4, the submolecular resolution allows us to determine the orientation of the Co-TPP within the layers and to determine the dimensions of the unit cell and its orientation relative to the underlying Pb/Si (111) - √7 × √3 lattice precisely. Within a single domain, the three types construction of the corresponding images was done by simply placing the scaled models in the 1.42 ± 0.02 nm square unit cell with orientation of the Co-TPP, while the subtended angle is 90° ± 1°. This molecular lattice structure has also been reported for the monolayer of Co-TPP on both the Ag (111) [35, 38 - 41, 42] and Au (111) [3, 33, 43] surfaces, which the lattice is the typical arrangement for various TTP’s on different substrates, Fe-TPP/Ag (111) [5, 42], H2-TPP/Ag (111) [3, 39, 41, 44, 45] Ni- and Cu-TPP/Au (111) [33, 46],. 57.
(65) Cu-TPP/Au (111) [3]. The few difference of the Co-TPP adsorbed directions and lattice size (unit cell ~1.35 nm) on Cu (111), because the other interaction between nitrogen and Cu [36, 37]. Hence, the order of the Co-TPP molecules reflects that observed metal-TPP’s on low reactivity surfaces typically, with corresponding high mobility and strong lateral intermolecular interaction. The principal molecular axes of the individual Co-TPP molecules, are found to be in parallel alignment, and show an azimuthal rotation of 20° ± 0.5° together. According to above results, we set up a model in Fig. 3.3.5 (d). Thus the self-assembly in well-ordered domains rely on intermolecular interactions. Considering the lattice constant of ~1.42 nm and the overall dimensions of the Co-TPP molecules as well as the flat adsorption geometry, after reports that the intermolecular forces must originate from the periphery of the molecules, i.e., the phenyl rings. A close inspection of Fig. 3.3.5 (d) reveals that the phenyl-legs of neighboring molecules are oriented in a T-stacking manner. Such a T-stacking geometry, where the H atom of one ring points toward the center of the adjacent ring is well characterized by calculations for benzene dimers [47 - 50], the T-stacking interaction diagram shows in Fig. 3.3.6. This arrangement is known to induce attractive inter-. 58.
(66) actions and thus stabilizes the molecular assembly. Thus, the T-stacking intermolecular lateral interactions are responsible for the observed arrangement; they stabilize the square packing and dominate over site specific adsorbate-substrate interactions. This holds true at least for Fe-TPP, Co-TPP, Zn-TPP and H2-TPP on Ag (111) [42]. There, the table 3.1 to integrate information about the Co-TPP’s self-assembled on various substrates. Due to the interaction between molecules and molecules most of the molecules are self-assembled to big islands. But some Co-TPP molecules trap in defect sites. In particular, the molecules will be standing vertically in default sites marked at purple arrows. We just see two phenyl legs pattern of a Co-TPP in STM images.. 59.
(67) Fig. 3.3.4. The image is the Co-TPP molecules self-assembly on Pb/Si (111) - √7 × √3 (or HIC). Since the Pb coverage is 0.5 ML, we still can to see Si (111) - 7 × 7 and yellow rhombus mark a unit cell. We can easily to identify each orientation of the substrate used by 7 × 7 and a few √7 × √3 pattern, the main direction of [1-21] and [10-1] marked at green arrows. The substrate has mixed √7 ×√3 and HIC phase which is formed by Pb plate- Si (111) after molecules adsorption at low- temperature. The Co-TPP film is identified to three different molecular domains, as marked by A, B and C. Three different have same lattice size and the unit cell marked by red boxes. The two diagonals of the unit cells are aligned with the Si (111) [-211] and [0-11] mark by blue arrows. (Info: 2.0 V, 0.95 nA ). 60.
(68) a. b. c. d. Fig. 3.3.5. Zoom-in STM images of the Fig 3.3.4 domain A shows the type I in (a), B is type II in (b) and C is type III in (c). The type I and III pattern are two-fold symmetric and type II is quasi-four-fold. We use a Co-TPP model to mark a molecule and a red box to mark a unit cell. According above results, we mimic a model in (d). Three structures are suited to the model of the relationships. (Info: 2.0 V, 0.95 nA ). 61.
(69) a. b. Fig. 3.3.6. (a) T-stacking configuration was hydrogen of benzene pointed at the center of the other ring, resulting in a perpendicular motif of neighboring phenyl rings of Co-TPP shows in (b).. Co-TPP plated-. Pb/Si(111) - 1 × 1 (or √7×√3 or HIC). Au (111). Ag (111). Cu (111). Unit cell. Lattice orientation. Interaction. Square 1.42 ± 0.02 nm • a = 1.45 ± 0.05 nm b = 1.40 ± 0.05 nm [3] • a ~ 1.4 nm [33] Square (90˚ ± 2˚) 1.405±0.02 nm [35]. • a = 1.45 ± 0.05 nm [2] b = 1.40 ± 0.05 nm • a ~ 1.35 ± 0.05 nm [37]. T-stacking interaction Diagonal aligned [-211] and [0-11] • T-stacking • N - Cu interaction (axis of two N aligned [-110]). Table 3.1. Here to compare the Co-TPP’s self-assembled on various substrates about the unit-cell, lattice orientation and interaction.. 62.
(70) After discussing the self-assembly morphology of Co-TPP on Si (111) - 1 × 1 (or HIC) at different temperature, we now proceed to investigate the adsorption-induced deformation of the molecules. There are two special events observed in our experiments: One is the three conformations formed on the Pb/Si (111) surface. This experiment results are some difference to the Co-TPP adsorbed on Au (111) [3], Ag (111) [35] and Cu (111) [2]. There is only type I conformation adsorbed on noble metals. Two is the type III conformation appeared at low- temperature. This means that the type III is generated by phase transition.. 63.
(71) 3.3.3 Molecular conformation and the geometry transition at low temperature Type I is saddle-shape conformation usually can be observed at the Co-TPP adsorption-induced distortion on noble metals, Au (111) [3], Ag (111) [35] and Cu (111) [2, 37] shown in Fig. 3.3.3 (a) to (c). The Fig. 3.3.7 (a) and (b) shows the main images and model of saddle-shape conformation. The two-fold symmetric appearance of Co-TPP points to a conformational adaptation on adsorption. The filled and empty states have different characteristics. Obviously, the individual molecular appearance drastically changes and now the phenyl substituents dominate the appearance of the Co-TPP in the STM image, there we focus in occupied state. Individual porphyrin molecules appear in their well-known shapes with four outer lobes/four sidewise extensions- depending on the actual tip status and applied bias voltage-reflecting the phenyl legs. Porphyrin cores adsorb in saddle geometry with two pyrrolic units bent upward, which appear as two prominent maxima when tunneling out of occupied states, and two pyrrolic units bent downwards, which are basically invisible in STM images. It should be noted that the dihedral angles of the phenyl groups are related to the macrocycle distortions, as these moieties are coupled by hydrogen and steric. 64.
(72) interactions. According to the research of Co-TPP [35] and Fe-TPP [5] molecules on Ag (111), the two outer maxima along the main axis are tentatively assigned to two opposite upward bent pyrrole rings expressing the saddle-shape macrocycle distortion described by an angle θpyrrole, whereas the central protrusion originates from the Co. In addition, the outline of the molecule indicates that the phenyl legs are not oriented perpendicular to the molecular core, but alternately rotated by a given angle θphenyl around the C - C bond connecting them to the macrocycle, the saddle model shown in Fig. 3.3.7 (c). As pure STM imaging is generally insufficient to determine the molecular conformation due to contribution from both electronic and geometric effects, we cannot know the distorted angle of the pyrrole and phenyl legs. Johannes V. Barth group investigate the distortion by near-edge x-ray adsorption fine-structure (NEXAFS) [35]. They quantify the adsorption-induced distortion of molecules/metal by angle-dependent NEXAFS spectroscopy, a technique sensitive to the orientation of molecular orbitals with respect to the substrate. This combined STM-NEXAFS approach was successfully applied to resolve the adsorption structure of Co-TPP on Ag (111) and Cu (111). They found the results are distorted 45˚, 35˚ of phenyl legs and subsequent macrocycle relaxation yielding pyrrole-moiety tilts of 30˚,. 65.
(73) a. b c. Fig. 3.3.7. (a) and (b) are detailed image of filled and empty state of type I (saddle conformation) showing molecular resolution. Besides the molecular main axis (three aligned maxima, dashed line) four dimmer protrusions corresponding to the phenyl groups can be identified. The molecular packing is described by a square unit cell with side length a=1.42 ± 0.02 nm. [Info: (a) -1.0 V, 0.98 nA; (b) 1.6 V, 0.98 nA] (c) Saddle conformation model. The three blue colors all is C, so the light and dark of blue colors just to mark that up and. 66.
(74) down of the distortion pyrrole. The pyrrole up or drop has a relationship with rotation of the phenyl legs. There up pyrroles with central protrusion of Co.. 20˚ relative to the plane of the porphyrin core formed at Co-TPP on Ag (111) [35] and Cu (111) [2], respectively. Therefore, saddle conformation still has slight variations take shape at different systems. There compare results of different adsorbed conformation of Co-TPP on various substrates with our experiment are shown in table 3.2. In our system, the type II and III are special structure adsorbed on the Pb/Si (111) surface. As mentioned earlier, the saddle conformation only to observed at Co-TPP adsorption on noble metals in all most case. More particular, the type II will be phase transition to type III after sample cooling to low-temperature. We first analyze the geometry of type II and III of Co-TPP. Type II is planar conformation [51] formed on Pb/Si (111) surface, the filled and empty states images and model shows in Fig. 3.3.8 (a) and (b). There same as room temperature results. The observed quasi-four-fold symmetry has to be attributed to four- fold symmetric adsorption geometry of the porphyrin macrocyle. Therefore, this structure seems not distortion to be formed on the surface. In the filled state STM image of Fig 3.3.8 (a), the individual. 67.
(75) molecules show four bright terminal protrusions, between which are found a further set of lesser intensity. Based on the molecular dimensions, the terminal protrusions are attributed to the phenyl groups. In turn, the pyrrole groups of the porphyrin macrocycle, which are located between adjacent phenyl groups, are tentatively assigned to the four core protrusions. As the same, the unoccupied state STM image shown in the Fig. 3.3.8 (b) reveals a striking same occupied state in the observed intermolecular features. We draw a simple model in Fig. 3.3.8 (c), the planar conformation mean is the macrocycle not distortion and parallel to the surface. The four-fold symmetry pattern of type II since the phenyl legs tilts of ~ 90˚ relative to the macrocycle plane. Type III conformation just appeared at low-temperature, its phase transition by type II structure. The type III filled and empty states images and model shows in Fig. 3.3.9 (a) and (b). The occupied state like the type II is the quasi-four-fold symmetry shapes with four outer lobes. But two-fold symmetry pattern shows in the unoccupied state. The results very like the Cl-Mn-TPP [52] and H2-TPP [42] adsorbed on Ag (111) surface pattern. However, we compare the H2-TPP assembly on metals results. Although the TPP maintained the porphyrin macrocycle flat, but the pyrrole legs still rotations. 68.
(76) a. b. c. Fig. 3.3.8. (a) and (b) are detailed image of filled and empty state of type II (planar conformation) showing molecular resolution. There shows quasi-four-fold symmetry image in occupied and unoccupied states. The molecular packing is described by a square unit cell with side length a=1.42 ± 0.02 nm. [Info: (a) -0.8 V, 0.19 nA; (b) 1.7 V, 0.70 nA] (c) Planar conformation model. since the phenyl legs tilts of ~ 90˚ relative to the macrocycle plane.. 69.
(77) relative to the plane. According to calculated macrocyclic conformation of the H2-TPP molecule using the ab initio molecular orbital method [53]. The ground state conformation is formed through about 63° rotations of pyrrole relative macrocycle [54]. Interestingly, above description is H2-TPP on the metal case. But we believe the same conformation occurred in our system, just cannot to measure the angle of rotation of four pyrrole legs. Furthermore, the type III two-fold symmetric pattern appeared in empty state image is like type I shown in Fig. 3.3.10. We believe that two types have the same direction of rotation of the phenyl ring, just different angles. As previously described, the type I distortion the porphyrin macrocyle attributed to the interaction between hydrogen and hydrogen of pyrrole and lesser angle of phenyl legs. Therefore, the type III phenyl legs should be rotation, just the large angle not impact pyrrole. Such there type III structure is ground state exposition consistent with Takashi Yokoyama et al. [54]. Since the type III structure transform by type II of Co-TPP, so we according the distortion result called “iso-planar”, model shows in Fig 3.3.9 (c). Compared with the Fig 3.3.8 (c), we can easily understand the subtle differences between iso-planar and planar structure is rotation of the phenyl leg. The point is the rotation angles. 70.
(78) of phenyl leg ( the θphenyl need greater than 63˚ ) just avoid distortion macrocycle by the interaction between hydrogen and hydrogen of pyrrole.. Co-TPP conformation. Saddle shape Distortion Saddle. No info.. Planar. Θpyrrole = 0˚ Θphenyl = 90˚. Pb/Si(111) √7×√3 (or HIC). Au (111). Ag (111). Cu (111). Iso-planar (ground state). Θpyrrole = 0˚ [54] Θphenyl = 63˚ ( By ab initio molecular orbital method ). Saddle. No info. (But Au-TPP on Au (111) [55] result is same Co-TPP on Ag(111)). Saddle. Θpyrrole = 30˚ [35] Θphenyl = 45˚ ( By NEXAFS ). Saddle. Θpyrrole = 20˚ [2] Θphenyl = 35˚ ( By NEXAFS ). Table 3.2. There compare the different adsorbed conformation of Co-TPP on various substrates.. 71.
(79) a. b. c. Fig. 3.3.9. (a) and (b) are detailed image of filled and empty state of type III (iso-planar conformation) showing molecular resolution. There shows quasi-four-fold symmetry image in occupied and unoccupied states. The molecular packing is described by a square unit cell with side length a=1.42 ± 0.02 nm. [Info (a) -0.9 V, 0.29 nA; (b) 1.9 V, 0.19 nA] (c) Iso-planar conformation model. since the phenyl legs tilts of 90˚ > θphenyl > 63˚ relative to the macrocycle plane. We use the four arrow mark the direction of the longer side.. 72.
(80) Fig. 3.3.10. The STM image is show two empty states conformations of type I and III. Their main difference of two patterns is the bright center in the type I, not appear in type III. We used molecular models mark the type I and III. (Info: 2.0 V, 0.48 nA). Accordingly, the type III is ground state conformation since the phenyl leg angle rotation of type II. The type II would like transition to type III after cooling to low-temperature. Therefore, the type II is unstable structures easily affect the quality of STM image. In contrast, the type II and III structure is susceptible to interconversion by scanning tip. Figure 3.3.11 (a) to (c) show the three different times image scanning in same region, the geometry transition between type II and III induced by scanning tip. In Fig. 3.3.11 (a) to (b), the type II. 73.
(81) to III and III to II marked in black and red dotted line area, respectively. The type II transition back to III formed red dotted line area will be come next in Fig. 3.3.11 (b) to (c), and new area transition from II to III barked in blue dotted line area. The transition model shows in Fig. 3.3.11 (d). The difference between the two models is rotated of the phenyl legs. Interestingly, such transition can be independent of single leg in a molecule. We can easily observe the morphology at the domains boundary between type II and III shows in Fig. 3.3.12. In the STM image, some molecules form two-face type in boundary marks by red arrow and molecular models. The two-face types include half type II half III and three- quarters of II (or III quarter of the III (or II). According this result coupled the unoccupied state of STM topography of the phenyl legs of type II and III, the nearly circular and egg-shaped pattern shows in Fig. 3.3.8 (b) and Fig. 3.3.9 (b), relatively. We can easily identify whether the rotating of single phenyl leg of a Co-TPP.. 74.
相關文件
進而能自行分析、設計與裝配各 種控制電路,並能應用本班已符 合機電整合術科技能檢定的實習 設備進行實務上的實習。本課程 可習得習得氣壓-機構連結控制
If amendment is necessary, please proceed before submission: Back to Form E Main Menu > Enter Step 1: Update by Student > Select Retrieve Record after
Microphone and 600 ohm line conduits shall be mechanically and electrically connected to receptacle boxes and electrically grounded to the audio system ground point.. Lines in
We were particularly impressed by the large garden which is looked after by the students and used to grow fruit, herbs and vegetables for the midday meal which the school serves free
Two examples of the randomly generated EoSs (dashed lines) and the machine learning outputs (solid lines) reconstructed from 15 data points.. size 100) and 1 (with the batch size 10)
3: Calculated ratio of dynamic structure factor S(k, ω) to static structure factor S(k) for "-Ge at T = 1250K for several values of k, plotted as a function of ω, calculated
List up all different types of high-sym k (points, lines, planes) 2...
The beam we selected for the bright-field image was the central (transmitted) beam of the diffraction pattern. Selecting other diffracted beam produces