In-plane anisotropy of two-dimensional materials
For the photoelectric properties of 2D materials, the linear polarized photo detection is a novel field. In terms of the in-plane structure of materials, they can be divided into symmetric materials and anisotropic materials. Materials with low in-plane symmetry have high angular sensitivity, and are potential candidate for linear polarized photodetector.
Principle of spin LED
The spin-LED is a LED with a spin injector deposited on it. The LED is generally a semiconductor heterostructure which includes a source of electrons (n-doped zone), a source of holes (p-doped zone), and intrinsic zone (non-doped). We called this heterostructure a PIN type structure. In spin-LEDs, the radiative recombination of spin-polarized electrons with unpolarized holes results in the emission of circularly polarized light. Figure 1.11 shows the schematic of a spin-LED used for electrical spin injection and optical detection. The analysis of the circular polarization of the emitted light makes it possible to characterize the efficiency of the spin injection. Researchers often use LED structures comprising quantum wells or quantum dot structure based on InGaAs/GaAs system.
The optical detection of an electrically injected spin polarization in a semiconductor was demonstrated for the first time in 1999 independently by Fiederling et al. and Ohno et al.. Fiederling et al. successful fabric the GaAs-based spin-LED with a spin injector of BeMnZnSe (Figure 1.12 (a)) and measured the circular polarization of 90% under 6K (Figure 1.12(b)). Since the easy magnetization axis of the spin injector was in the plane, a strong magnetic field (1.5T) was applied to orientate the magnetization out of the plane. This is necessary to satisfy the optical selection rules in the case of optical detection from surface emission. A maximum degree of electroluminescence circular polarization noted P= ++ − + − +− − ,where and are the intensities of the right and left circularly polarized components of the luminescence, respectively.
Ohno et al. used a GaMnAs ferromagnetic semiconductor to inject spin-polarized electrons. The device structure schematic and low temperature EL spectral are shown in Figure 1.13 (a) and (b). They obtained a spin injection into the GaAs with a circular polarization of the order of 10% at low temperature. Here, the magnetization of GaMnAs ferromagnetic semiconductor is in-plane. In this case, no need to apply an external magnetic field for edge light detection.
Chemical Vapor Deposition (CVD)
Figure 2.1 shows the machine of the Chemical Vapor Deposition (CVD). The chemical vapor deposition method refers to a reaction process in which a chemical reaction occurs between chemical substances in the gas phase at a high temperature condition. The target compounds are gradually deposited on the substrate. CVD is also the most effective method to directly grow large-scale, single-layer or double-layer two-dimensional materials. The size, number of layers, doping and defect of two-dimensional materials can be controlled through combining different experimental parameters such as temperature, growth time, air flow rate, and tube pressure.
The concentration of reactants is the key to the growth process. The main method of controlling the concentration is to control the ratio of reactants and the flow rate of the gas during growth. For example, during the silicon growth, the precursors of the reaction are SiH4 and H2, and the flow rate of the precursors is controlled. High-quality large-area silicon single crystals can be grown at the flow rate per cubic meter of 7.4×10 −7 mol/s. For transition metal chalcogenides, the reactants are solid oxides and sulfur (selenium) powder. Taking MoS2 as an example, the ratio of Mo source and S source directly affects the shape of growing MoS2. As shown in Figure 1.7, A regular triangle structure (Mo atoms at the edges) is formed when Mo:S is greater than 1:2. When Mo:S equal to 1:2, it is a regular hexagon. When Mo:S is less than 1:2, it is an inverted triangle structure (the edges are S atoms). At this time, the ratio of Mo:S concentration is controlled by the gas flow rate of the carrier gas (Ar). The fast flow rate of Ar gas leads to a slower concentration of sulfur, while the slow flow rate of Ar gas leads to a higher concentration of sulfur.
Temperature is the most important parameter in CVD growth. At the nucleation point of the gas and solid interface, high temperature controls the thermal process, and low temperature controls the mechanical process. In recent reports, when the upper triangle grows from the same nucleation point at the center of the bottom triangle, the temperature is higher than 800°C. In addition, temperature is also an important condition for controlling doping. Taking MoS2 as an example, its CVD growth temperature is about 650°C. For doped MoS2, a higher growth temperature is required. As reported by Li Bo et al, the growth temperature of Co-doped MoS2 is around 680°C。
In addition to reactant concentration and temperature, substrate is also the most important part of CVD. The main substrates are silicon wafers, silicon wafers with oxide layers, copper substrates, sapphire substrates, and mica wafers. Heavily doped silicon wafers with an oxide layer on the surface are the first choice for the growth of TMDCs two-dimensional materials. This is for the convenience and quickness of subsequent device preparation. Back-gate field effect transistors can be prepared in one step and the properties of the samples can be quickly obtained. Copper substrate and sapphire substrate are the first choice for large-scale growth. Because copper substrate and sapphire substrate have crystal orientation, the substrate with the same crystal orientation as the growth material can be selected before growth to reduce lattice mismatch and make the material easier deposit on the substrate. transport equipment. Most of the chemical vapor transport equipment are dual-temperature tube furnaces. The function of the dual-temperature zone is to provide a small temperature difference for the reaction zone, so that the sample is vaporized in the high-temperature zone and transported to the low temperature zone for growth. Similar to CVD, the chemical vapor transport (CVT) method is also one of the most important methods for growing materials. Unlike CVD, CVT mainly grows two-dimensional single crystal materials. The size of the bulk material is on the order of centimeters. And the crystal orientation is consistent. In CVT growth, there is no need for a substrate, and the reaction occurs in a high-vacuum (less than 1×10-3 Pa) sealed glass tube. The main control parameters are: growth temperature, growth time, reactant concentration, catalyst, etc.
During the CVT growth process, the range of growth temperature and growth time is wide. The selection of growth temperature is mainly determined by the evaporation temperature of the raw material and the growth temperature of the resulting material. In order to fully vaporize the raw materials in the glass tube and fully mixed, the temperature setting should firstly raise the dual temperature zone to be higher than the evaporation temperature of the raw material at the same time. Secondly, we reduce the temperature to near the growth temperature and maintain a temperature difference of 50-100 °C, which is beneficial to the gaseous reactants in the glass tube from high temperature flowing to the low temperature zone. The product will be synthesized near the low temperature zone.
The concentration of the reactants is the decisive factor for the success of the synthesis. The reactants in CVT growth are generally simple substances and a small part of chlorides. Taking SnS2 as an example, the ratio of Sn:S in the glass tube is 1:5. Excessive S can fully combine with Sn and act as a transport gas to enhance the saturated vapor pressure in the glass tube to obtain high quality SnS2 crystals. If the ratio of Sn:S in the glass tube is 1:2, a mixture of SnS and SnS2 can be grown. when the ratio of Sn:S is 1:1, Sn exists in the compound in a sub-state corresponding to SnS crystals. The carrier concentration of CVT grown single crystal is higher than that of CVD growth.
Catalysts are also an indispensable part of the reaction. Commonly used halogen elements are used as catalysts. For example, high-quality MoS2 generally uses liquid bromine (Br) as a catalyst. GeSe uses iodine as a catalyst (I2). Chlorine (Cl2) is also an catalyst for 2D crystal material growth.
Micro-nano processing technology
Figure 2.7 is a schematic diagram of a device based on two-dimensional materials. Because two-dimensional materials are generally small, micro-nano processing technology is used to prepare devices and test them.
In the first step, a substrate with a mark is prepared. A layer of photoresist (S1813) is spin-coated on the substrate. The prepared marking pattern is projectively transferred to the photoresist using an ultraviolet lithography process.Then the metal is vapor-deposited on the surface of the photoresist through an evaporation process. Finally, the metal is stripped through the Lift-off process to obtain a substrate with a marked pattern metal.
The second step is the transfer of two-dimensional materials. The thin layer of two-dimensional material is transferred to the prepared substrate with markings. Here, CVD growth can be used directly, or transfer by the mechanical lift-off method introduced.
Electron Beam Lithography (EBL)
Electron beam lithography is a high-precision exposure process. The principle is to use electron beams to directly scan, transfer, and copy graphics on the substrate with PMMA. Compared with ordinary lithography technology, electron beam exposure breaks through the diffraction limit of ordinary lithography. According to the theory of diffraction limit, the resolution of the exposure pattern is related to the wavelength of the light used in lithography. The shorter wavelength of light , the higher accuracy can be achieved by photolithography. According to the matter wave theory, the wavelength of electrons is much smaller than ultraviolet light. The wavelength of an electron with an energy of 100eV is only 0.12nm. Therefore, the resolution of electron beam lithography can be as high as 10nm.
The process EBL includes: substrate cleaning → PMMA spin-coating → heating (130°C/3 minutes) → electron beam lithography→ development → fixing (IPA) → hard film. Figure 2.8 is the physical image of the EBL equipment used in this paper. This equipment includes: sampling chamber, exposure chamber (vacuum degree of 10-7 mbar), electron emission chamber (vacuum degree of 10-9 mbar). The equipment shown in the picture is the most used electron beam equipment in this paper, and its acceleration voltage can reach up to 30KV.
Monolayer MoS2 was grown by chemical vapor deposition (CVD) using highest purity (6N) gases and precursors on SiO2 substrate (purchased from 2D semiconductor Inc.). The MoS2 substrate is firstly introduced into an MBE system to perform an annealing at 150°C during 1h to desorb the water and CO2 on the surface (MBE is shown in Figure 3.1 (b)). When the sample is cooled down to RT, the sample was transferred to another sputtering system (with a base pressure of 5×10-8torr) without breaking the vacuum (sputtering is shown in Figure 3.1(a)). Multilayer stack with the structure of MgO (1.5nm) /Co0.4Fe0.4B0.2 (1.2nm) /Ta (5nm) was deposited at RT. After taking out the sample from the vacuum, we have performed RTA treatment on the sample at different temperatures for 3min (The RTA is shown in Figure 3.2).
Figure 3.1(a) shows magnetron sputtering, which includes: transmission chamber, growth chamber, vacuum system, water cooling system, gas path system and control system. One end of the transmission chamber relates to the vacuum tube, and the other end relates to the growth chamber. There are three target positions in the growth chamber of this equipment, like MgO, CoFeB and Ta.
The Heterojunction of MoS2/MgO/CoFeB/Ta
The advantage using the full coverage MoS2 substrate allows us to characterize the magnetic property of sample directly by superconducting quantum interference device (SQUID) magnetometry. To obtain an ultrathin CoFeB layer with PMA on MoS2, we have optimized CoFeB thickness and annealing temperature (Tan).
Figure 3.6 (a) displays the out-of-plane magnetization vs. external magnetic field (M-H) curves for annealed spin-injectors (Tan=250°C) with different CoFeB thicknesses measured at 10K. From 1nm to 1.2nm, the saturation field quickly decreases and the remanence increases with the increase of CoFeB thickness. However, when the thickness is larger than 1.2nm, the saturation field increases and the remanence decreases with the increase of CoFeB thickness. This can be understood that the CoFeB layer is discontinued and it exhibits a superparamagnetic character with a thickness lower than 1.2nm. When the thickness is larger than 1.2nm, the CoFeB can form continuous film exhibiting ferromagnetic property.
Table of contents :
Chapter 1: Development and application of two-dimensional materials
1.1.1 Crystal Structure
1.1.2 Electrical properties
1.2 Transition Metal Dichalcogenides (TMDs)
1.2.1 Crystal Structure
1.2.2 Electrical and Photoelectric Properties
1.3 In-plane anisotropy of two-dimensional materials
1.3.1 Crystal structure
1.3.2 Photoelectric Properties
1.4 Application of two-dimensional materials
1.4.1 Spin Light-Emitting Diodes (LED)
1.4.2 Spin photodiode
1.4.3 Linear Polarized Photodetection
Chapter 2: Experimental techniques
2.1 Synthesis of two-dimensional materials
2.1.1 Chemical Vapor Deposition (CVD)
2.1.2 Chemical Vapor Transport (CVT)
2.1.3 Mechanical Exfoliation
2.2 Synthesis of metal film
2.2.1 Molecular beam epitaxy (MBE)
2.2.2 Magnetron sputtering
2.3 Micro-nano processing technology
2.3.1 Electron Beam Lithography (EBL)
2.3.2 UV lithography
2.4 Basic characteristic parameters
2.4.1 Magnetic Film
Chapter 3: Ta/CoFeB/MgO on full coverage monolayer MoS2
3.2 Heterojunction Processing and characterizations
3.2.1 Heterojunction Processing
3.3 Experimental results and discussion
3.3.1 Monolayer MoS2 characterization
3.3.2 The Heterojunction of MoS2/MgO/CoFeB/Ta
3.3.3 The Heterojunction of MoS2/MgO/CoFeB/Pt
3.3.4 The Heterojunction of MgO/CoFeB/Ta
3.4 First-principles calculation
3.4.1 Calculation method
3.4.2 MoS2/7ML MgO/7ML Fe Heterojunction
3.4.3 MoS2/3ML MgO/7ML Fe Heterojunction
3.4.4 Band structure of Fe/MgO/MoS2
Chapter 4: GeAs based polarized photodetector
4.2 GeAs crystal structure
4.3 Characterization of samples and devices
4.3.1 Transmission electron microscope (TEM)
4.3.2 Angle-resolved polarized Raman spectroscopy
4.3.3 Polarization absorption spectrum of GeAs
4.4 Photoelectric properties
4.4.1 Ordinary photoelectric properties
4.4.2 Polarized photoelectric properties
4.5 First-principles calculation
4.5.1 Band structure calculation
4.5.2 Calculation of absorption spectrum
Chapter 5: IV-VI compounds based Photodetector
5.1 ZnSnS alloy based Photodetector
5.1.1 Crystal structure and characterization
5.1.2 Optical properties
5.2 SnS based Photodetection
5.2.1 Crystal structure and characterization
5.2.2 Optical properties
Chapter 6: Summary and Prospects