溅射沉积过程英文

CHAPTER 5 Sputter Deposition Processes D. Depla1, S. Mahieu1 and J.E. Greene2 1Department of Solid State Sciences, Ghent University, Belgium 2Materials Science and Physics Departments and the Frederick Seitz Materials Research Laboratory, University of Illinois, Urbana, Illinois, USA Summary 253 5.1 Introduction: How Popular is Sputter Deposition? 254 5.2 What is Sputtering? 255 5.3 How are the Energetic Particles Generated? 261 5.4 Efficient Trapping of Electrons Leads to Magnetron Sputter Deposition 267 5.4.1 Post Magnetrons 267 5.4.2 Planar Magnetrons 270 5.4.3 Rotating Cylindrical Magnetrons 272 5.4.4 Some General Features of Magnetrons and Magnetron Discharges 272 5.4.5 Powering the Magnetron 277 5.5 Reactive Magnetron Sputter Deposition 280 5.5.1 Hysteresis of Reactive Gas Pressure and Discharge Voltage 280 5.5.2 Understanding the Hysteresis Behavior: Modeling the Reactive Sputter Process 280 5.5.3 Circumventing the Hysteresis Problem 286 5.6 Moving Toward the Substrate 287 5.6.1 Sputtered Particles 287 5.6.2 Other Particles Arriving at the Substrate 290 5.7 Sputter-Deposited Thin Films: Morphology and Microstructure 290 5.7.1 Zone I Films 291 5.7.2 Zone T Films 292 5.7.3 Zone II Films 294 5.8 Conclusions 294 Summary Sputter deposition is a widely used technique to deposit thin films on substrates. The technique is based on ion bombardment of a source material, the target. Ion bombardment results in a vapor due to a purely physical process, i.e. the sputtering of the target material. Hence, this technique is part of the class of physical vapor deposition techniques, which includes thermal Copyright © 2010 Peter M. Martin. Published by Elsevier Inc. All rights reserved. 253 254 Chapter 5 evaporation and pulsed laser deposition. The most common approach for growing thin films by sputter deposition is the use of a magnetron source in which positive ions present in the plasma of a magnetically enhanced glow discharge bombard the target. This popular technique forms the focus of this chapter. The target can be powered in different ways, ranging from direct current (DC) for conductive targets to radio frequency (RF) for non-conductive targets, to a variety of different ways of applying current and/or voltage pulses to the target. Since sputtering is a purely physical process, adding chemistry to, for example, deposit a compound layer must be done ad hoc through the addition of a reactive gas to the plasma, i.e. reactive sputtering. The undesirable reaction of the reactive gas with the target material results in a non-linear behavior of the deposition parameters as a function of the reactive gas flow. To model this behavior, the fluxes of the various species toward the target must be determined. However, equally important are the fluxes of species incident at the substrate because they not only influence the reactive sputter deposition process, but also control the growth of the desired film. Indeed, the microstructure of magnetron sputter-deposited films is defined by the identity of the particles arriving at the substrate, their fluxes, and the energy per particle. 5.1 Introduction: How Popular is Sputter Deposition? One way to compare sputter deposition with other deposition techniques is to count the relative number of scientific publications and patents published each year for each deposition technique. In order to provide a baseline for the rate of increase in publications in general, we first determine the number of publications per year that refer to the combination of keywords ‘gold’ OR ‘silver’ OR ‘copper’ on the Web of Sciences [1]. Similarly, a baseline for published patents is easily found by counting the number of patents published annually by entering the search term ‘the’ in the Delphion Database [2]. Figure 5.1(a) is plot of the relative number of scientific papers per year per deposition technique. Figure 5.1(b) provides the same information for patents. For the different physical vapor deposition (PVD) techniques, magnetron sputtering is clearly used extensively in the scientific community, and competes with pulsed laser deposition (PLD) as the most important deposition technique. From the search in the patent database, it can be concluded that sputter deposition is still the most popular technique. Hence, a chapter on sputter deposition in a book about thin films is quite relevant since this technique is applied in both research laboratories and industrial plants to deposit a wide variety of materials. In this chapter, we focus on key aspects of sputter deposition. Describing the physics behind the sputter process, i.e. the interaction between the ion and the target, is a first priority. However, since many review articles are available [3–9], only the essential points are discussed here. Then, a basic system design is described. It should be noted that in this chapter we exclude ion beam sources, which have been well reviewed in the literature [10], owing to Sputter Deposition Processes 255 Figure 5.1: (a) (left): Normalized number of publications per year per deposition technique based on data from the Web of Science. (b) (right) Normalized number of publications per year per deposition technique based on data from Delphion. limitations associated with scalability and power supply options. In the following sections several discharge sputter deposition configurations are discussed. Sputter deposition is also used to deposit compound films by adding a reactive gas to the discharge. This, however, greatly increases the complexity of the deposition process, and explains the ongoing interest in academia to investigate this technique of which several aspects are not completely understood. Sputtered target atoms are ejected with substantial kinetic energy, of the order of or larger than bond energies, and hence can significantly affect film growth kinetics and microstructure. Thus, energy loss mechanisms during transport in the gas phase are important. The chapter ends with a discussion of the typical microstructure of sputter-deposited coatings. 5.2 What is Sputtering? Sputtering is the ejection of atoms by the bombardment of a solid or liquid target by energetic particles, mostly ions. It results from collisions between the incident energetic particles, and/or resultant recoil atoms, with surface atoms. A measure of the removal rate of surface atoms is 256 Chapter 5 Figure 5.2: Sputtering yield Y of Cu as a function of the energy of Ar+ at normal incidence as calculated using the SRIM code. Note that Y(EAr+) is linear over the typical range of operation during magnetron sputtering (EAr+ = 250–750 eV). the sputter yield Y, defined as the ratio between the number of sputter-ejected atoms and the number of incident projectiles. Excellent review articles on sputtering are available in the literature [3–9], and only the essential features are discussed here. Based on the large amount of experimental (e.g. [11]) and calculated data as a function of ion and target material, several trends are apparent. For a given ion mass and target, Y exhibits a maximum as a function of ion energy as well as a minimum (threshold) energy. An example is shown in Figure 5.2 for Ar+ bombardment of Cu. When comparing the sputter yield of target materials bombarded by a given ion at constant energy, one notices a trend related to the position of the element in the periodic table (see Figure 5.3 and the following discussion). Several authors have derived equations describing the sputter yield as a function of energy and projectile–target combinations. P. Sigmund is the father of these theories. His work ‘Theory of sputtering I. Sputter yield of amorphous and polycrystalline targets’, published in Physical Review [12], is a benchmark in this field. According to the theory of Sigmund, the sputter yield near threshold, i.e. at low ion energy, is given by Y = 3 4π2 α 4M1M2 (M1 + M2)2 E Us (5.1) Sputter Deposition Processes 257 Figure 5.3: Dependence of the sputter yield of several elements (ordered according their position in the periodic table) calculated using SRIM (initial conditions: 300 eV Ar, other input parameters where set at the standard values given by SRIM: lattice binding energy, surface binding energy displacement energy, and normal incidence). with E the energy of the projectile, and M1 and M2 the masses of the projectile and the target atom (in amu). Us is the surface binding energy and α a dimensionless parameter depending on the mass ratio and the ion energy. At low energy, and mass ratios M2/M1 lower than 1, α is of the order of 0.2. This equation can be understood as follows. An incoming ion transfers its momentum to the target atoms which explains the term 4M1M2/(M1 + M2)2 with a maximum when M1 = M2. To sputter an atom from the target, momentum transfer from the ion-induced collision must overcome the surface barrier, given by the surface binding energy Us. Therefore, we can expect an inversely proportional relationship between the yield and the surface binding energy. Based on Eq. (5.1), we can expect that for the energy range of interest for sputter deposition, the sputter yield will vary linearly with the ion energy (see Figure 5.2). The behavior of the sputter yield over the periodic table can also be understood from Eq. (5.1), because the sputter yield is defined by momentum transfer and surface binding energy. However, differences in atomic density among different materials also affect Y through variations in the range (depth) of momentum transfer. In addition to the theory of Sigmund, heuristic approaches based on semi-empirical equations, and simulations (for an overview on sputter yield simulations, see [13]) are also available. 258 Chapter 5 Table 5.1: Overview of the terms in Yamamura formulae Symbol Definition Q(Z2) Tabulated dimensionless parameter ˛ 0.249(M2/M1)0.56 + 0.0035(M2/M1)1.5, M1 ≤ M2 0.0875(M2/M1)−0.15 + 0.165(M2/M1), M1 ≥ M2 Eth 6.7 � Us, M1 ≥ M2 1+5.7(M1/M2) � Us, M1 ≤ M2 � 4M1M2 (M1+M2)2 � W (Z2) 1+(M1/7)3 W(Z2) Tabulated dimensionless parameter s Tabulated dimensionless parameter k Linhard electronic stopping coefficient Sn(E) 84.78Z1Z2 (Z2/31 +Z 2/3 2 ) 1/2 M1 M1+M2 s TF n (ε) ε Lindhard–Scharff–Schiott reduced energy sTFn (ε) Reduced nuclear stopping cross-section For more details, see [14]. Commonly used semi-empirical formulae for the calculation of the sputter yield were developed by Yamamura et al. [14]. The equations are valid for the bombardment of monoatomic solids by projectiles at normal incident, and the sputter yield Y(E) is given by Y (E) = 0.042Q(Z2)α(M1/M2) Us Sn(E) 1 + �kε0.3 ⎡ ⎣1 − √ Eth E ⎤ ⎦ s (5.2) with E, M1, M2, and Us as defined for Eq. (5.1). Us, the surface binding energy, is intimately connected with, and explains the presence of the threshold energy (Eth) for sputtering (Table 5.1). The other symbols are defined in Table 5.1. Although these equations provide a value for the sputter yield Y as a function of ion energy and material choice, they are not instructive in explaining the sputtering process in detail. Hence, some authors have developed simpler, and more transparent, models of the sputtering process. An excellent example is the work of Mahan et al. [15], in which the sputter yield Y(E) is derived based on the following assumptions. The effective number of recoiling target atoms created per incident ion is multiplied by the probability that the recoil is close enough to the surface to escape and by the probability that the recoils are traveling toward the surface, or Y = E Eavg Rpr Rpp 1 4 (5.3) Sputter Deposition Processes 259 with E the projectile energy and Eavg the average energy of the recoils. The ratio E/Eavg gives the average number of recoils. The ratio between the projected range of the recoils Rpr and the projected range of the projectile Rpp gives the probability that the recoils are close enough to the surface to escape. Finally, the term {1/4} is the average probability that the recoils are moving toward the surface. Using approximations, the average recoil energy and the projected range can be calculated straightforwardly giving insight into the sputtering process. A good example of this is the calculation of the threshold energy. The physics behind the threshold energy is that the recoil atom has insufficient energy to overcome the surface energy barrier Us when its average energy Eavg is equal to or lower than the surface barrier energy. The average recoil energy in this simplified model is calculated as Eavg = Usln(γE/Us) (5.4) and the threshold energy E = Eth is therefore found by substituting Us for Eavg, and obtaining Eth = 2.72 Us/γ , which is similar to the formulae proposed by Yamamura et al. (see Table 5.1). Another approach to obtain a value for the sputter yield is to simulate the overall sputtering process (see [13] for an overview). The most commonly used simulation package is the well-known SRIM code [16] developed by Ziegler et al. and completely described in their book [17]. Figures 5.2 and 5.3 were simulated using this code. All features discussed in the previous paragraphs are present in both figures. SRIM is a static simulation code. That is, it does not account for target changes due to the ion bombardment itself. To account for these changes, Mo¨ller et al. developed a dynamic version of SRIM, i.e. TRIDYN [18]. All approaches also provide the angular and energy distribution of the sputtered particles. Thompson [19] showed that within a certain (low) ion energy range, the energy distribution follows the expression dY dE ∝ E(E + Us)3 (5.5) which is based on the assumption of a planar surface barrier for sputtered particles. This expression gives a peak at Us/2. Falcone [20] estimated the average energy of sputtered particles ¯E as ¯E = 2Us ( ln E Eth − 3 2 ) (5.6) Substituting reasonable values into Eq. (5.6) shows that the energy of the sputtered particles is at least one order of magnitude higher than the corresponding thermal evaporation energy for 260 Chapter 5 Figure 5.4: Comparison between the thermal energy distribution for copper evaporated at 1300K and the energy distribution of sputtered copper atoms. the same particle flux (Figure 5.4). Indeed, at 1000 K the thermal energy is only of the order of 0.1 eV. With a surface binding energy of a few eV (for Cu it is 3.5 eV [17]), the maximum in the energy distribution occurs at 1.8 eV, and using the SRIM calculated threshold energy, the average energy is 15.1 eV. If the pressure during deposition is low enough, the sputtered particles in the gas phase are ballistic and can reach the substrate with few or no collisions in the gas phase. In all of this work, one important concept is not addressed at all, i.e. the sputter yield of compound materials. The Yamamura formulae can be applied to multicomponent materials such as compounds and alloys using weighted average values for Z2, M2, and Us [13]. Surface binding energies, in particular, are difficult to obtain for oxides and nitrides. For metals, the surface binding energy is generally set equal to the vaporization enthalpy, but this approach is not applicable to oxides, nitrides, sulfides, etc. In the context of the preferential sputtering of oxygen from oxides during depth profiling of oxide thin films for analytical approaches, models have been proposed to estimate a value of the surface binding energy of the metal and the oxygen atoms. Using these values, the modification of the surface composition by ion bombardment is calculated and compared with experimental values. Results typical for this kind of study, including a model for the surface binding energy, are published by Malherbe et al. [21]. Some simulation codes, e.g. TRIDYN, use a different approach [18], where the effective surface binding energies of O and M are chosen to be dependent on the actual surface composition by use of a matrix method. The matrix elements of surface binding energies are SBVO–O, SBVO–M, SBVM–O, and SBVM–M. These elements are evaluated by Sputter Deposition Processes 261 the formulae SBVO−O = 0 SBVM−M = Us,M SBVO−M = 12Us,M + n + m 2nm �H f + n + m 4n �Hdiss (5.7) where n and m depend on the stoichiometry of the oxide MnOm. Us,M is the metal surface binding energy, �Hf denotes the formation enthalpy per molecule of the compound, and �Hdiss denotes the dissociation energy of the oxygen molecule. If the concentrations of O and M at the surface are CO and CM, respectively, we obtain the surface binding energy of M and O in the following way: SBE(M) = CO · SBVO−M + CM · SBVM−M SBE(O) = CM · SBVO−M + CO · SBVO−O (5.8) 5.3 How are the Energetic Particles Generated? Sputtering is initiated by the bombardment of energetic particles at the target. These energetic particles are generally ions. Two approaches can be followed to produce ions and sputter the target materials. The first is quite straightforward by using an ion source which is aimed toward the target. Collecting the sputtered particles on a substrate enables the deposition of a thin film. However, ion beam sputtering is not widely used for industrial large-scale applications. Ions guns are more often utilized in surface analytical techniques such as secondary ion mass spectrometry (SIMS) or to bombard the substrate during thin film deposition [22]. As such, these external sources of ions will not be covered in this chapter. A good overview can be found in [10]. Another source of ions is a plasma. By applying a high negative voltage to the cathode, i.e. the target, positively charged ions are attracted from the plasma toward the target. The ions gain energy in the electric field and bombard the target with sufficient energy to initiate sputtering. Sputtering was first discussed in the literature by W.R. Grove in 1852, who used this kind of set-up [23]. A good starting point to discuss plasma-based sputter deposition is using the simplest experimental arrangement. That is, a cathode and an anode are positioned opposed to each other in a vacuum chamber. Typically, the vacuum chamber is pumped by a combination of turbomolecular and rotary pumps, although a diffusion pump is still often used. After pumping 262 Chapter 5 Figure 5.5: The three primary regions of a gas discharge. The straight line is a typical load line. to a base pressure of the order of 1 × 10−4 Pa 1 or lower, a noble gas (usually argon) is introduced into the vacuum chamber, reaching a pressure between 1 and 10 Pa. When a high voltage difference in the range of 2000 V is applied between cathode and anode, a glow discharge is ignited. It is not in the scope of this chapter on sputter deposition to describe all details related to a glow discharge, but discussing a few can be instructive. To define a glow discharge, one can follow the current–voltage (I–V) characteristics. It is important to realize that these characteristics depend also on the pressure and the separation between cathode and anode. The main characteristics of the discharge, such as breakdown voltage, I–V characteristics, and structure of the discharge, depend on the geometry of the electrodes (cathode and anode) and vacuum vessel, the gas(es) used, and the electrode material. The I–V characteristics of such a discharge are illustrated in Figure 5.5 for a wide range of currents. Three general regions can be identified in the figure, the dark discharge region, the glow discharge, and the arc discharge. The electric circuit of the discharge gap also includes an external ohmic resistance R. In this case, Ohm’s law for the circuit can be written as EMF = V + RI (5.9) 1 Other commonly used pressure units are mtorr and mbar: 1 mtorr corresponds to 0.133 Pa; 1 mbar corresponds to 100 Pa. Sputter Deposition Processes 263 where EMF is the electromotive force and V is the voltage of the gas discharge. Equation (5.9) is usually referre