Substrate processing using plasma modulated with a DC magnetic field

Abstract

A method for plasma processing a substrate, the method comprising generating plasma in a plasma chamber in which the substrate is held during processing, wherein generating plasma means passing a discharge gas through the plasma chamber and coupling a radio frequency (RF) source signal to a first RF electrode, coupling means ionizing the discharge gas; coupling a bias signal to a second RF electrode, the bias signal being a periodic series of bias pulses, each period having a bias-on time and a bias-off time, with a bias voltage waveform applied during the bias-on time; generating a pulsed DC magnetic field in the plasma chamber by coupling a magnetization signal to an electromagnet, the magnetization signal being a periodic series of current pulses; and passing a DC magnetization current during the bias-on time by synchronizing a periodic series of current pulses with the bias signal before coupling the magnetization signal.

Description

[Technical Field]

[0001] Cross-reference of related applications This application claims the benefits of U.S. Nonprovisional Patent Application No. 18 / 241,008, filed on 31 August 2023, which is incorporated herein by reference in its entirety.

[0002] The present invention generally relates to plasma processing technology, and in particular embodiments to a system and method for processing a substrate with plasma modulated by a DC magnetic field. [Background technology]

[0003] An integrated circuit (IC) is a monolithic structure comprising a network of electronic components in a semiconductor substrate. An array of IC units is formed on each substrate as batches of substrates are processed through a series of patterning levels. At each level, patterned layers of various materials are formed using deposition, photolithography, and etching processes. Typically, the minimum half-pitch of the array of trenches or holes fabricated at a given patterning level defines its patterning capability. Approximately every two years, new IC nodes with more advanced technologies are introduced, enabling a doubling of component density, reducing the cost per unit while simultaneously improving performance and functionality. This doubling results from reducing the minimum feature size and deploying three-dimensional (3D) devices, a combination that requires fabricating high-aspect-ratio structures with controlled, small dimensional variations.

[0004] The first step in patterning a layer is printing, in which a pattern of chemical lines is transferred to a photoresist layer coated on the substrate. In a patterning process called single patterning, etching follows printing, transferring the resist pattern to an optional hard mask and then to a target layer. The patterning capability of single patterning is the resolution limit of the optical system, which is proportional to λ / NA, where λ is the radiation wavelength and NA is the numerical aperture. The minimum half-pitch decreased from 1 micrometer to 0.1 micrometers over 20 years, during which time λ decreased from 436 nm (ultraviolet (UV) mercury lamp) to 193 nm (deep ultraviolet (DUV) ArF excimer laser). The development of lithography systems with shorter λ = 13.5 nm (extreme UV (EUV) tin plasma) took approximately another 20 years. During this time, scaling was made possible by using techniques such as optical proximity correction (OPC), phase-shift photomasks, immersion lithography (193-i), and multiple patterning, but these came with higher processing costs and new patterning errors such as pitch walking. In recent years, 13.5nm EUV has been used for cost-effective manufacturing at the sub-10nm node, where lateral features of less than 20nm frequently appear in IC designs, and therefore the aspect ratio of the corresponding vertical structures has been pushed up to the point where a special etching process, generally called high aspect ratio contact (HARC) etching, is required.

[0005] Many deposition and etching processes in forming 3D devices (e.g., nanosheet transistors and vertical NAND (V-NAND) memory) are plasma processes, and some of these processes, such as HARC etching, may require specific plasma properties, such as supplying a highly collimated ion flux to the substrate. Further innovation is desired to achieve enhanced control over plasma properties such as plasma density, electron temperature, radical flux, ion flux, and ion angle and energy distribution, in order to meet the challenge of patterning high aspect ratio features at nanoscale pitch with high yield for mass production. [Overview of the project] [Means for solving the problem]

[0006] A method for plasma processing a substrate, the method comprising generating plasma in a plasma chamber in which the substrate is held during processing, wherein generating plasma means passing a discharge gas through the plasma chamber and coupling a radio frequency (RF) source signal to a first RF electrode, coupling means ionizing the discharge gas; coupling a bias signal to a second RF electrode, the bias signal being a periodic series of bias pulses, each period having a bias-on time and a bias-off time, with a bias voltage waveform applied during the bias-on time; generating a pulsed DC magnetic field in the plasma chamber by coupling a magnetization signal to an electromagnet, the magnetization signal being a periodic series of current pulses; and passing a DC magnetization current during the bias-on time by synchronizing a periodic series of current pulses with the bias signal before coupling the magnetization signal.

[0007] A system for plasma processing, comprising: a plasma chamber; a substrate holder configured to hold a substrate within the plasma chamber; a gas flow system configured to flow a discharge gas through the plasma chamber; a first radio frequency (RF) electrode; a second RF electrode configured such that the first RF electrode and the second RF electrode work together to generate plasma within the plasma chamber; an electromagnet configured to generate a pulsed DC magnetic field within the plasma chamber, wherein the pulsed DC magnetic field is configured to modulate the electron temperature profile of the plasma; a first electrical circuit configured to output an RF source signal; a second electrical circuit configured to output a bias signal; a third electrical circuit configured to output a magnetization signal; and a controller configured to transmit control signals to the first electrical circuit, the second electrical circuit, and the third electrical circuit in order to adjust and synchronize the RF source signal, the bias signal, and the magnetization signal.

[0008] A method for plasma treatment of a substrate, comprising: passing a discharge gas through a chamber; coupling a radio frequency (RF) source signal to a first RF electrode; coupling a bias signal to a second RF electrode, wherein the bias signal is a periodic series of bias pulses, each period being the sum of bias-on time and bias-off time, and a bias voltage waveform is applied during the bias-on time; generating a pulsed DC magnetic field in the chamber by coupling a magnetization signal to an electromagnet, wherein the magnetization signal is a periodic series of current pulses; and passing a DC magnetization current during the bias-on time by synchronizing a periodic series of current pulses with the bias signal before coupling the magnetization signal.

[0009] For a more complete understanding of the present invention and its advantages, please refer to the following description in conjunction with the accompanying drawings. [Brief explanation of the drawing]

[0010] [Figure 1A]Schematic diagrams of plasma processing systems according to several embodiments are shown. [Figure 1B] Schematic diagrams of plasma processing systems according to several other embodiments are shown. [Figure 2] A flowchart summarizing a method for plasma treatment of a substrate according to several embodiments is shown. [Figure 3A] The waveforms of the source signal, bias signal, and magnetization signal, as well as a graph of the time response of the mean electron temperature in the central region of the plasma chamber, according to one embodiment are shown. [Figure 3B] The waveforms of the source signal, bias signal, and magnetization signal, as well as the time response of the mean electron temperature in the central region of the plasma chamber, according to another embodiment, are shown. [Modes for carrying out the invention]

[0011] This disclosure describes embodiments of a method for processing a substrate with a plasma modulated by a pulsed DC magnetic field in a plasma chamber coupled to a gas flow system for passing a discharge gas through the chamber at low pressure. The plasma may be generated by coupling radio frequency (RF) power to ionize a gas and ignite and maintain a glow discharge plasma, and the pulsed DC magnetic field may be generated by an electromagnetized with a pulsed DC current. The electric and magnetic fields in the chamber are controlled by the effective electron temperature (T) of the plasma in the region of the chamber above the substrate, as will be described in more detail below. e ) may be configured to reduce T e The effect of reducing is to alter the ion velocity such that the dispersion of ion angle and ion energy in the joint distribution of ion angle and energy in the ion flux incident on the substrate is reduced. Thus, in embodiments of the methods described herein, a highly collimated ion flux with precisely controlled kinetic energy can be used to perform plasma processing.

[0012] Ion fluxes with very small ion angles and energy spreads are advantageous for anisotropic reactive ion etching (RIE) processes, especially when the RIE process is for etching high aspect ratio features, commonly known as high aspect ratio contact (HARC) etching. As the lateral dimensions of electronic components within an IC scale more rapidly than the vertical dimensions, the aspect ratio of vertical structures tends to increase at most patterning levels. For example, at the back-end of line (BEOL), the width of holes (e.g., for contacts and vias) and trenches (e.g., for metal lines) in the interlayer dielectric (ILD) layer shrinks more rapidly than the ILD thickness and trench depth. At the front-end of line (FEOL), the emergence of 3D field-effect transistors (FETs) (e.g., FinFETs, nanosheet FETs, forksheet FETs, and complementary FETs (CFETs)) and 3D memory (e.g., vertical NAND (V-NAND) memory) similarly results in vertical structures with increasing aspect ratios as they miniaturize. Self-aligned patterning (SAQP) using 193nm immersion (193-i) lithography was used to fabricate features with a minimum half-pitch of approximately 25nm at the 10nm node. However, IC designs targeting the sub-10nm node include features with a minimum half-pitch of less than 20nm in some patterning layers. For these, a more cost-effective 13.5nm EUV single exposure lithography is used to print the patterns.

[0013] After printing, the pattern is transferred to a hard mask layer using a resist as an etching mask, and the hard mask layer is used as an etching mask to pattern vertical openings (e.g., via holes) into a target layer (e.g., an ILD layer) using an anisotropic RIE process. For dimensions at the sub-10 nm node, commonly used vertical structures such as contacts and vias may have aspect ratios greater than 10, with the maximum aspect ratio being in the range of 50–100. Providing HARC etching for fabricating nanoscale-width openings with such high aspect ratios is not straightforward. In addition, to achieve high yield and reliability, the process may need to meet stringent specifications for sidewall profile control and local limiting dimensional uniformity (LCDU). To help HARC etching meet these requirements, as described below, plasma processing equipment and methods may need to generate a perpendicular downward ion flux relative to the substrate, in which case the ion distribution is controlled so that the ion angle and energy spread are very narrow. Due to its specificity, a 0-degree ion angle refers to the direction perpendicular to the ion's velocity when an ion collides with the main surface of the substrate in a direction perpendicular to the surface.

[0014] During HARC etching, it is undesirable for high-energy ions to collide with the sidewalls exposed by etching. Such collisions can cause undesirable effects such as line edge roughness (LER) and undercuts. Furthermore, inelastic collisions and ion absorption at the sidewalls can randomly reduce the availability of vertically oriented high-energy ions at the bottom of partially etched openings, correspondingly causing random fluctuations in the etching rate. As the bottom of a partially etched feature progresses deeper during anisotropic etching, the aspect ratio of the partially etched feature increases, and therefore the tolerance for ion angle fluctuations is narrowed to near 0 degrees. Clearly, reducing the spread of the angle distribution benefits the HARC etching process by reducing the likelihood of incident ions colliding with the sidewalls.

[0015] Ions in the plasma within the chamber collide with neighboring particles, which are primarily neutral particles. Generally, both the angle and kinetic energy of ions are affected by random collisions. Therefore, the dispersion of ion energy is related to the dispersion of ion angles. In this disclosure, plasma processing refers directly to plasma processing, and therefore, the ion angles and energies of interest are those of ions immediately before they collide with the surface of the substrate being processed, after being accelerated across a plasma sheath formed adjacent to the surface of the substrate. The sheath is a narrow, positively charged region of a high electric field directed toward the surface, formed along the periphery adjacent to the interface, whereas the bulk of a glow discharge plasma, called bulk plasma, is a charge-neutral region (i.e., the difference between positive and negative charge densities is negligible). Between the bulk plasma and the sheath is a quasi-neutral transition region called the pre-sheath. The ion velocity distribution in these three regions and the electron temperature (T) are used to determine the ion velocity distribution in the ion flux incident from the sheath to the surface. e The effects of ) are explained below.

[0016] The glow discharge plasma within the chamber contains two types of charged particles, namely ions and free electrons, along with predominantly neutral particles. For the sake of simplification, in the discussions of this disclosure, all neutral particles are grouped together and referred to as a single third species called the neutral species. Furthermore, ions are assumed to be a single chemical species with one unit of positive charge. These simplifications are for the purpose of clarifying the explanation and should not be construed as limiting the scope of the invention.

[0017] Particles are distributed across a wide range of energies. The energy distribution of various particles is given by exp(-E / k). B It can be approximated by the Maxwell distribution function proportional to T), where E represents the total energy of the particle (kinetic energy and potential energy), and k B is the Boltzmann constant, T represents the effective temperature, and ((3 / 2)k) is the average kinetic energy of the species particles. B This represents (equal to T). In the Maxwell approximation, kinetic energy is isotropically distributed, and therefore the directional (vertically downward) component of kinetic energy is (3 / 2)k B Note that this is not included in T. For example, ions in a strongly collimated ion flux colliding with a surface have a large directional component of kinetic energy, while the effective ion temperature within the sheath represents only the random isotropic component of ionic kinetic energy, which is much smaller.

[0018] The RF power that sustains the glow discharge increases the average kinetic energy of the gas mixture in the chamber beyond the thermal equilibrium value given by the ambient temperature. The excess kinetic energy is unevenly divided among the three species due to uneven energy absorption and distribution due to collisions. Thus, in thermal equilibrium, T is equal to the ambient temperature, but in non-equilibrium, the steady state can result in uneven effective temperatures for each species. The discharge plasma is not considered to be in thermal equilibrium because the potential energy from the externally induced RF field is converted into kinetic energy of free electrons in the plasma at a rapid rate. In general, the steady-state effective temperature of a species is the final result of the rate at which particles of that species gain or lose kinetic energy during the initial transient period as the plasma settles into a steady state.

[0019] Electrons, being extremely light compared to ions (typically with a mass ratio of several thousand), possess extremely high mobility. Due to this high electron mobility, free electrons rapidly acquire kinetic energy from the RF field in the bulk plasma (much faster than ions). Furthermore, free electrons elastically collide with other species (i.e., heavier ions and neutral species), but mainly due to their large mass ratio, very little momentum (and therefore kinetic energy) is transferred. When the rate at which kinetic energy is gained is much greater than the rate at which kinetic energy is lost, the average kinetic energy of free electrons is approximately 10,000 K or 1 eV. e The energy rapidly increases to a steady-state value corresponding to (i.e., approximately 30 to 200 times the thermal equilibrium value). High-energy free electrons undergo further ionization through inelastic collisions with neutral species, thus generating ion-electron pairs, which maintain the steady-state flux of ions and free electrons to the bulk plasma and the interface, including the chamber walls and the main surface of the substrate.

[0020] On the other hand, neutral species do not absorb energy from an electromagnetic (EM) field. Furthermore, as described above, collisions between neutral species and high-energy free electrons can be inelastic; in that case, instead of increasing the kinetic energy of the neutral species, kinetic energy is transferred from the free electrons, ionizing the neutral species. If the increase in kinetic energy is negligible, the effective temperature of the neutral species remains at ambient temperature (e.g., about 300K or 0.026 eV). In contrast, on average, the kinetic energy of ions increases slightly upon collisions with higher-energy electrons, while part of the gain is primarily lost through collisions with neutral species. The net result is that the effective temperature of ions (T i ) is slightly higher than the effective temperature of neutral species, which is at ambient temperature. However, as explained above, the increase in T i is smaller than the increase in T e . Typically, in bulk plasma, T i is several hundred degrees higher than ambient (e.g., about 350K to 900K, or 0.03eV to 0.08eV).

[0021] As described above, the initial transient after plasma ignition rapidly settles to a steady-state glow discharge plasma. In the present specification, the steady state is divided into three regions according to the movement of ions. Ions can be scattered by collisions and move with velocities distributed in all directions. At any location, the ion velocity distribution is approximated by a shifted Maxwell function comprising an isotropic portion superimposed on an anisotropic portion. The isotropic portion is a Maxwell function with T i as its temperature, and represents the random movement of ions at that location. The anisotropic portion accounts for the movement of ions directed perpendicular to the surface, and on average becomes a net directional component of ion velocity V d , which gives rise to the ion flux at that location. In other words, at any location, the average kinetic energy per ion is equal to (3 / 2)k B T i for the random component, and (1 / 2)M ion V d 2 for the directional component, where M ionThis is the mass of one ion.

[0022] The bulk plasma is the furthest from the surface, where the density of ions and free electrons is highest, the net charge and electric field are negligible, and there is a high collision frequency mainly with neutral species. As a result, the steady-state velocity distribution of ions and free electrons in the bulk plasma is almost perfectly isotropic, which is V d The fact that is almost zero suggests that the average kinetic energy per ion can be almost entirely attributable to random components. In a charge-neutral bulk plasma, V d Although the field strength is low, there is a slow-moving ion flux (and free electron flux) that "drifts" toward the surface. As ions approach the vicinity of the substrate, they enter a narrow sheath adjacent to the substrate surface. As mentioned above, the plasma sheath is a narrow region of high electric field where the electric field is directional, rapidly accelerating positively charged ions toward the surface to high speeds.

[0023] Ions emerging from the bulk plasma pass through the pre-sheath before entering the sheath. As mentioned above, the pre-sheath is the region where the properties of the steady-state plasma transition from their values ​​in the bulk plasma to significantly different values ​​at the edge of the sheath. Outside the bulk plasma, ion and free electron densities gradually decrease toward the surface, with free electron density decreasing more rapidly than ion density. This results in a final positive charge density profile that starts at negligibly low values, close to the charge-neutral bulk plasma, but gradually increases toward the surface through the pre-sheath, reaching much higher values ​​in the sheath. The pre-sheath region can be considered quasi-neutral compared to the high charge density inside the sheath. The direction of the electric field preferentially increases ion velocity toward the surface and thus increases the anisotropy of the ion distribution function. In fact, V d It starts increasing from almost zero near the bulk plasma, and after entering the pre-sheath for a short distance, the directional component of the kinetic energy ((1 / 2)M) ion V d 2) is the random component of the ionic kinetic energy ((3 / 2)k B T i ) becomes equal to ((3 / 2)k B T i ) far exceeds. In addition, V d As T increases, the average kinetic energy of ions scattered in ion-neutral species collisions in the pre-sheath region also increases, adding to the random movement of ions, so in the pre-sheath, T i The amount of T increases. However, as will be explained further below, inside the sheath i It may not increase any further, i.e., T i It reaches its highest value at the end of the sheath.

[0024] A sheath adjacent to the substrate surface is formed during the initial transient, when the rapid outward diffusion of free electrons from the plasma into the substrate causes the substrate to become negatively charged. The transient settles into a state where the negative charge of the substrate is balanced by the positive space charge within the sheath and the associated electric field profile that attracts ions and repels free electrons. This establishes a steady state including a potential profile and ion and free electron flux that satisfies the surface boundary conditions, e.g., an applied voltage if the substrate is biased, or a net-zero current requirement if the substrate is floating. As is known to those skilled in the art, in order to obtain a stable steady-state sheath, the directional average velocity of ions entering the sheath must exceed a minimum value called the Bohm velocity. The (as described above) quasi-neutral pre-sheath is subjected to V until it equals the Bohm velocity at a location defined as the sheath edge. d This is the region where the electric field increases. The pre-sheath is quasi-neutral, and most of the positive charge and potential drop occur in the narrow space between the sheath edge and the substrate surface, so the electric field reaches very high values ​​within the sheath.

[0025] Ions within the pre-sheath reach the edge of the sheath, where they randomly move into the adjacent sheath above the substrate at a random velocity V. o Let's assume it enters as follows: V oDepending on its size and direction, it has parallel components V (parallel and perpendicular to the surface of the substrate). x0 and vertical component V y0 It can be decomposed into: On average, at this interface (sheath edge), the magnitudes of the parallel and perpendicular components of the ion velocity are similar. When ions enter the sheath, they are accelerated toward the surface by the sheath electric field, gaining high kinetic energy. The sheath is a narrow region of high electric field, and ions are typically swept across the sheath with transport times much shorter than the mean free time between collisions. In the absence of collisions, the increase in ion velocity within the sheath is directed parallel to the field, i.e., perpendicular to the surface. In other words, if the velocity at which an ion exits the sheath and collides with the surface is V1, then its perpendicular component V y1 V y1 ≫V y0 It increases in such a way that its parallel component V x1 V remains unchanged, that is, x1 =V x0 This is the result. In addition, V x0 and V y0 Since their sizes are similar, generally V y1 ≫V x0 This is the result.

[0026] Note that increasing the ion velocity in the direction of the field reduces the ratio of the parallel to perpendicular components. This causes the ion angle to decrease, i.e., to approach the perpendicular. When an ion enters the sheath at the sheath end, the ion angle θ o ,=tan -1 (V ×0 / V y0 ) and the ionic angle θ1,=tan -1 (V x1 / V y1 ) is. V x1 =V x0 , and V y1 ≫V y0 Therefore, θ1≫θ o This is the result. Furthermore, from the inequality discussed above, the decrease in the ion angle (θ) o -θ1) is a larger angle θ oThis can be shown to be greater for ions entering the sheath. Therefore, acceleration within the sheath narrows the ion angular distribution and strengthens the directionality of the ion flux. The sheath electric field can be increased, for example, by applying a negative substrate bias. However, an increase in the average energy upon collision with the surface can lead to several undesirable effects such as surface damage, surface roughness, and heating.

[0027] As mentioned above, reducing the ion incidence angle is particularly important in etching high aspect ratio features. The ion angle is controlled by increasing the sheath electric field using a substrate bias, thereby reducing the perpendicular component of the ion velocity (V y1 ) to increase the horizontal component of the ion velocity in the pre-sheath (V x0 It can be reduced by either of the following: x0 The kinetic energy associated with it is due to random collisions, and therefore V x0 This is achieved by reducing the random component of the ionic kinetic energy, that is, T i This can be reduced by reducing (as described above). If there are no random collisions within the sheath, the random component of ionic kinetic energy remains unchanged from the component of ions in the pre-sheath region near the sheath end. If there is no further increase in the random component of ionic kinetic energy, the T inside the sheath i There was no further increase in T, that is, i It reaches its highest value at the end of the sheath.

[0028] As discussed above, the ion temperature (T i ) refers to the randomized average energy of ions, which are randomized by collisions with ions in a charge-neutral bulk plasma and a quasi-neutral pre-sheath. Therefore, T i This is the net result of the kinetic energy gained by the ion through a collision (e.g., an elastic collision with an electron) and the kinetic energy lost by the ion through a collision (e.g., a collision with a neutral species). iThe upper limit can be determined by considering the energy equilibrium condition in the steady state, where the ionic energy lost by collisions (mainly by neutral species) is, on average, balanced by the ionic energy gained by collisions (mainly by high-energy electrons). This is the ionic temperature T (which reaches its highest value at the sheath end). i to, electronic temperature T e To relate to this, as is known to those skilled in the art, an approximate formula can be derived using the theory of plasma physics, T i ∝T e 2 / 3 Therefore, from this analysis, T in bulk plasma and pre-sheath i It is clear that this reduction can be achieved by lowering the electron temperature there. As explained above, these conditions improve the ion incidence angle on the substrate during processing by achieving a more collimated ion flux incident on the surface.

[0029] Embodiments of this disclosure use a pulsed DC magnetic field in a chamber to reduce the ion angle and energy spread of ions in the ion flux used to process the substrate, thereby reducing the T in the plasma locally above the substrate. e This reduces the T of the plasma above the substrate in order to achieve a better controlled and highly collimated ion flux incident on the substrate. e While it is desirable to reduce T, in order to efficiently generate ions and radicals and maintain the desired plasma density in the plasma chamber, high T is necessary. e There are still cases where this is desirable. Various embodiments of the plasma processing systems and methods described herein use a pulsed DC magnetic field to process T e This offers the advantage of locally modulating the timing of the pulsed DC magnetic field, which will be explained in more detail below.

[0030] FIG. 1A is a schematic diagram of a plasma processing system 100 for direct plasma processing of a substrate 102 disposed on the upper insulating surface of a substrate holder 104 shown at the lower portion of a plasma chamber 110. The plasma chamber 110 has a central region around a central axis (shown by a dotted line in FIG. 1A) that vertically passes through the center of the substrate holder 104 on which the substrate 102 is held during processing, and an adjacent edge region surrounding the central region. The system 100 is configured to generate plasma and a pulsed DC magnetic field that modulates the electron temperature (T e ) profile of the plasma. Plasma may be generated by ionizing a discharge gas flowing through the chamber 110 using RF source power and pulsed bias power coupled to the gas by a first RF electrode 106 and a second RF electrode 108 of the system 100. The pulsed bias power is obtained from a bias signal, and the bias signal is a periodic series of bias pulses. The pulsed DC magnetic field may be generated by coupling a magnetization signal, which is a periodic series of current pulses, to an electromagnet (e.g., a planar electromagnet 112A schematically shown in FIG. 1A). The magnetization signal is synchronized to be tuned with the bias signal, that is, a DC magnetization current flows in each current pulse during the same time that a bias voltage waveform is applied in each bias pulse.

[0031] The DC magnetic field is a pulsed DC magnetic field, and the electron temperature profile changes over time. A first profile is generated during a time period when the DC magnetic field is present, and a second profile is generated during a time period when the DC magnetic field is not present. The average T in the central region of the plasma chamber 110 e is a first T for the first profile e (T e1 ) and a second T for the second profile e (T e2 ), and T e2 is greater than or equal to T e1 . As the DC magnetization current increases, the ratio T e2 / T e1 increases.

[0032] In the exemplary embodiment shown in Figure 1A, the substrate 102 is a semiconductor wafer in an intermediate stage of the IC fabrication flow, and the substrate holder 104 is the broad cylindrical upper portion of a multipurpose pedestal having a hollow stem (not shown) that extends through the floor of the chamber 110 and provides passages for various connectors coupled to equipment outside the chamber 104, such as wires, pipes for circulating liquid through cooling channels within the substrate holder 104, and openings for injecting gas into grooves on the upper surface through the pedestal to cool the back surface of the substrate 102. The substrate holder 104 may operate as an electrostatic chuck (ESC) by gripping the back surface of the substrate 102 with a DC electric field by positioning gripping electrodes 105 within a few millimeters of the upper surface, the DC electric field may be generated by coupling a DC voltage source to the gripping electrodes 105 using wires passing through the stem. Generally, the gripping electrode 106 comprises a conductive mesh embedded (to a depth of a few millimeters or less) within the body of the substrate holder 104, the conductive mesh comprising an insulator or a semi-insulating material coated with an insulator (e.g., ceramic), on which the substrate 102 is placed. In some embodiments, the bottom and sides of the substrate holder 104 may be insulator-coated conductors (e.g., ceramic-coated refractory metal) for use as electrodes, as will be further described below.

[0033] In some embodiments, a portion of the wall of the chamber 110 may include a conductor (e.g., aluminum) coated with an insulator (e.g., yttria), which is typically coupled to a reference potential commonly referred to as ground.

[0034] As described above, the RF source power and pulsed bias power can be coupled to the gas in the chamber 110 to generate a plasma. The RF source signal can be coupled to the first RF electrode 106, and the bias signal can be coupled to the second RF electrode 108.

[0035] In embodiments described herein (e.g., system 100 in Figure 1A), the first RF electrode 106 is an RF antenna or RF resonator positioned above a dielectric window 114 and configured to couple RF source power by inducing an EM field into the plasma within the chamber 110. The plasma generated by coupling EM power from an electrode outside the chamber 110 is generally referred to as inductively coupled plasma (ICP), and the plasma processing system (e.g., system 100) is said to be configured in ICP mode.

[0036] Those skilled in the art should note that embodiments using an ICP configuration (e.g., system 100 in Figure 1A) can be adapted to embodiments in which the system is configured in charge-coupled plasma (CCP) mode with RF source power coupled to an upper electrode exposed to the plasma.

[0037] The plasma can be sustained by the RF power received in the upper portion of the chamber 110 located vertically below the first RF electrode 106. However, the plasma extends horizontally to the side wall 116 and vertically from the ceiling to the substrate holder 104. In Figure 1A, a dashed rectangle is used to indicate the end of the sheath. Both the quasi-neutral pre-sheath and the charged neutral bulk plasma are within the dashed rectangle.

[0038] One advantage of configuring the chamber 110 to generate plasma using power coupled from two independent sources (RF source power and bias power) is that the plasma density, as well as the vertical potential and electric field profile, can be adjusted independently. Generally, the RF source signal is used to adjust the plasma density, and the bias signal is used to adjust the potential drop across the sheath and pre-sheath.

[0039] A magnetization signal containing a series of periodic current pulses can be coupled to an electromagnet (e.g., a plane electromagnet 112A) to generate a pulsed DC magnetic field.

[0040] In some embodiments (e.g., system 100), the first RF electrode 106 is an antenna in the shape of a planar coil, positioned above the central portion of the chamber ceiling, as shown in Figure 1A. The portion on which the antenna is positioned is often called a dielectric window 114, i.e., a dielectric transparent to RF radiation from the antenna. In the exemplary system 100, the planar coil comprises a single helical conductor extending across an area covering the substrate 102. However, it should be understood that conductors of various other shapes, such as resonators in the shape of a radiating structure with multiple arms connected to a central ring, can also be used as the first RF electrode 106.

[0041] In some embodiments, the chamber 110 has a conductive material in a portion of its wall (as described above). In that portion, the dielectric window 114 is removed. For example, in one embodiment, the chamber 110 may have a side wall 116, which includes a conductive material (e.g., aluminum) coated with an insulator (e.g., yttria) and coupled to ground, forming a grounding surface around the side of the plasma chamber 110.

[0042] In some embodiments, the second RF electrode 108 is part of the upper portion of a multipurpose pedestal, which is a substrate holder 104 in the system 100. In the embodiment shown in Figure 1A, the second RF electrode 108 includes a conductive plate positioned below the gripping electrode 105 and embedded within the body of the substrate holder 104. In other embodiments, where the bottom and sides of the substrate holder 104 are conductive and coated with an insulator (as described above), the conductive bottom and sides of the substrate holder 104 may be used as the second RF electrode 108.

[0043] In some embodiments, the electromagnet for generating a pulsed DC magnetic field within the chamber 110 may be a planar electromagnet 112A. The planar electromagnet 112A, schematically shown in Figure 1A, includes a conductor in the shape of a planar coil around the central axis of the chamber 110, with the inner diameter of the coil being greater than the diameter of the substrate 102. As shown in Figure 1A, the conductor is positioned above the ceiling of the chamber 110. In some other embodiments, for example, in the plasma processing system 101 schematically shown in Figure 1B, the electromagnet may be a solenoid electromagnet 112B. In Figure 1B, the solenoid electromagnet 112B includes a conductor in the shape of a helical winding around the side wall 116 of the chamber 110, with the coil positioned outside the chamber 110. Typically, the coils used in electromagnets (e.g., planar electromagnets 112A and solenoid electromagnets 112B) are constructed as multiple windings of a continuous conductive wire, each winding configured to carry an equal current. In some embodiments, the coil may comprise two or more continuous wires. In some embodiments, the current in one winding may differ from the current in another winding of the coil. Furthermore, in some embodiments, the electromagnet may be a configuration of multiple magnets. For example, in some embodiments, the electromagnet may comprise a configuration of a first electromagnet and a second electromagnet. A first magnetization signal may be coupled to the first electromagnet, and a second magnetization signal may be coupled to the second electromagnet.

[0044] The current flow in the plane electromagnet 112A and the solenoid electromagnet 112B is axially symmetric about the central axis of the chamber 110, and therefore the pulsed DC magnetic field created by the axially symmetric current flow is also axially symmetric. Due to the large inner diameters of the coils of the plane electromagnet 112A and the solenoid electromagnet 112B (larger than the diameter of the substrate 102), the DC magnetization current flows near the side walls 116 of the chamber 110. Considering that the strength of the DC magnetic field generated by the current in a wire decreases with increasing distance from the wire, the strength of the pulsed DC magnetic field in the chamber 110 is high near the side walls 116 and decreases with decreasing radial distance from the central axis. Therefore, on average, the magnetic field strength in the central region of the plasma chamber 110 above the substrate 102 is lower compared to the higher average magnetic field strength in the adjacent edge regions surrounding the central region. It should be noted that the planar electromagnet 112A of system 100 (Figure 1A), and the solenoid electromagnet 112B of system 101 (Figure 1B) are configured to generate, on average, stronger DC magnetic fields in the edge regions compared to the magnetic field in the central region where the substrate 102 is held during processing. Magnetic fields can be used to spatially confine charged particles. This confinement is more pronounced the lighter the particles are, and therefore pulsed DC magnetic fields are very effective in confining free electrons. This characteristic of pulsed DC magnetic fields is further described below, in the embodiments described in this disclosure, the mean electron temperature T in the central region of chamber 110. e It is used to reduce [something].

[0045] The DC magnetic field causes the mobile charged particles to rotate around the magnetic field lines, resulting in helical motion of free electrons along the magnetic field lines of the DC magnetic field in chamber 110. When the helical free electrons reach the sheath, the electric field within the sheath reflects the helical free electrons back into the bulk plasma. The stronger the DC magnetic field, the smaller the radius of the helix. The helical motion tends to confine the free electrons along the magnetic field lines. Because the DC magnetic field is a pulsed DC magnetic field, the T in chamber 110 e The average value changes over time.

[0046] The RF source signal, bias signal, and magnetization signal are output by the first electrical circuit 120, the second electrical circuit 130, and the third electrical circuit 140, respectively, as shown in Figure 1A.

[0047] The first electrical circuit 120 may include an RF oscillator for generating a continuous-wave (CW) RF signal and a power amplifier configured to supply sufficient power to the plasma. In a CW RF signal, an RF sinusoidal voltage waveform is continuously applied. In some embodiments, the RF source signal is a CW RF signal. Generally, the first electrical circuit 120 includes an impedance matcher between the power amplifier and the first RF electrode 106 for efficient power transfer from the first electrical circuit 120 to the first RF electrode 106.

[0048] In some embodiments, the RF source signal is a CW RF signal, while in other embodiments, the RF source signal is a periodic series of source pulses. Therefore, a chopper circuit may be included to enable the first electrical circuit 120 to output the RF source signal as either a CW RF signal or a periodic series of source pulses. In exemplary embodiments, namely system 100 in Figure 1A and system 101 in Figure 1B, the configuration of the first electrical circuit 120 can be selected by the controller 150 using a control signal to enable or disable the chopper circuit. The chopper circuit is configured to "chop" a continuous sinusoidal RF voltage waveform to generate a periodic series of RF pulses. In an RF pulse, the sinusoidal RF voltage waveform is applied only for a portion of the pulse period. Each source pulse in the periodic series of source pulses is an RF pulse, and the RF source voltage waveform is applied only during the source-on time, where the RF source voltage waveform is an RF sinusoidal voltage waveform. Each period of the periodic series of source pulses is the sum of the source-on time and the source-off time. When enabled by the controller 150, the chopper circuit works in conjunction with the RF oscillator to generate a periodic series of source pulses. The controller 150 is described in more detail below.

[0049] The bias signal output by the second electrical circuit 130 is a periodic series of bias pulses (as described above), where each period is the sum of the bias-on time and the bias-off time, and for each bias pulse, the bias voltage waveform is applied only during the bias-on time. The general architecture of the second electrical circuit 130 is similar to the architecture of the first electrical circuit 120 in that both circuits include a signal source, a chopper circuit, a power amplifier, and an impedance matcher. The difference between the first electrical circuit 120 and the second electrical circuit 130 depends on the difference between the bias voltage waveform and the RF source voltage waveform.

[0050] In some embodiments, the bias signal, which is a periodic series of bias pulses, is a periodic series of DC pulses. In each DC pulse, the bias voltage waveform applied during the bias-on time is a constant DC voltage level. Thus, a periodic series of DC pulses can be generated by "chopping" a continuous, constant DC voltage level output from a DC voltage source. The polarity of the DC voltage level is such that the vertically downward component of the electric field in the sheath above the substrate 102 in the chamber 110 increases.

[0051] In some other embodiments, a periodic series of bias pulses is a periodic series of DC burst pulses. In each DC burst pulse, the bias voltage waveform applied during the bias-on time is a plurality of DC pulses. In other words, if a periodic series of DC pulses is transmitted through another chopper, each DC pulse in the periodic series of DC pulses will be converted into a DC burst pulse. Therefore, a second chopper circuit may be included to enable the second electrical circuit 130 to output the bias signal as a periodic series of DC pulses or as a periodic series of DC burst pulses.

[0052] As discussed above, one embodiment of a second electrical circuit 130 that can be configured to output a bias signal as a periodic series of DC pulses or a periodic series of DC burst pulses comprises a DC voltage source, a first chopper circuit configured to generate a periodic series of DC pulses in coordination with the DC voltage source, a second chopper circuit configured to generate a periodic series of DC burst pulses from a periodic series of DC burst pulses generated by the first chopper circuit, a power amplifier, and an impedance matcher. The second electrical circuit 130 may be configured by a controller 150 using appropriate control signals to enable the first and second choppers, depending on which bias voltage waveform is selected to be applied during the bias-on time.

[0053] In some other embodiments, the periodic series of bias pulses is a periodic series of RF pulses. In such embodiments, the second electrical circuit 130 comprises an RF oscillator, a chopper circuit, a power amplifier, and an impedance matcher, similar to the first electrical circuit 120 described above.

[0054] The magnetization signal output by the third electrical circuit 140 is a periodic series of current pulses. To generate a periodic series of current pulses, the third electrical circuit 140 includes a DC current source and a chopper circuit. When the chopper circuit is activated by a control signal from the controller 150, it generates current pulses in coordination with the DC current source. The controller 150 may synchronize the timing of the control signal to the third electrical circuit 140 with the timing of the control signal to the second electrical circuit 130 in order to synchronize the timing of the current pulses with the timing of the bias pulses.

[0055] The controller 150 may, for example, comprise a processor and a memory. Instructions may be stored in the memory, and when executed by the processor, the instructions generate control signals for the first electrical circuit 120 (for selecting a source signal), the second electrical circuit 130 (for selecting a bias signal), and the third electrical circuit 140 (for selecting a magnetization signal). The control signals from the controller 150 may also be used to synchronize the pulse signal waveforms generated by the first electrical circuit 120, the second electrical circuit 130, and the third electrical circuit 140 to synchronously apply the source signal, the bias signal, the magnetization signal, and a reference potential for generating plasma within the chamber 110 according to the requirements of a plasma process.

[0056] In embodiments of the plasma processing system described in the present disclosure, since a DC magnetization current is only passed during a bias on-time, the magnetization signal is synchronized with the bias signal before being coupled to an electromagnet (e.g., the planar electromagnet 112A or the solenoid electromagnet 112B). The synchronization ensures that the electromagnet is magnetized during the bias on-time and demagnetized during the bias off-time. In general, application of bias power increases the potential drop and vertical electric field in the sheath, therefore, when bias power is present, the vertical velocity of ions colliding with the substrate 102 (V y ) increases. Due to the increased kinetic energy of incident ions, most anisotropic etching occurs during the bias on-time. During the bias off-time, processing proceeds by the ion flux having lower kinetic energy and radicals diffusing out from the plasma reacting with the substrate 102.

[0057] By synchronizing the current pulse with the bias pulse, the magnetic field is turned on only when the bias power is coupled to the plasma within the chamber 110 (i.e., during the bias on-time). In embodiments of the present invention, this timing of the pulsed DC magnetic field is utilized to achieve T e reduction when anisotropic etching is enhanced by the bias power, and to maintain a low T eThis allows it to return to its high value. During consecutive bias pulses (i.e., during bias-off time), the substrate 102 remains exposed to ion fluxes with lower kinetic energy and diffusive radical fluxes from the plasma, which continue to modify the surface. Using RF source power T e By returning to a higher value, the plasma density is restored to its initial value. T e This increases the T (and thus provides a high plasma density) during the enhanced anisotropic etching period. e By reducing (and thus providing a highly collimated ion flux), this technique is advantageous for etching high aspect ratio features.

[0058] During the time the DC magnetization current is flowing (i.e., the bias-on time), a magnetic field profile is established within the chamber 110. As described above, the electromagnets (i.e., electromagnet 112A in system 100 or electromagnet 112B in system 101) are configured such that the average magnetic field strength in the central region of the plasma chamber 110 above the substrate 102 is lower compared to the higher average magnetic field strength in the adjacent edge regions surrounding the central region. The DC magnetic field profile within the chamber 110 of system 100 (Figure 1A) differs from that of system 101 (Figure 1B) due to the different geometric shapes of the plane electromagnet 110A and the solenoid electromagnet 112B, but the DC magnetic field profiles in system 100 and system 101 have this characteristic that the magnetic field in the edge regions is stronger than the magnetic field in the central region. This spatial distribution of the DC magnetic field is the average T in the central region of the chamber 110 when no DC magnetic field is present, as will be described below. e (T e2 ) for the mean T in the central region of the chamber 110 on the substrate 102 when a DC magnetic field is present. e (T e1 The electronic temperature profile is changed accordingly.

[0059] Free electrons and ions, being charged particles, experience forces due to the DC magnetic field as they move randomly through a bulk plasma. This force is directly proportional to the product of particle velocity, charge, and DC magnetic field strength. On average, in a bulk plasma, heavier ions move much slower than much lighter free electrons. Therefore, on average, the force exerted on ions by the DC magnetic field is much weaker than that on free electrons. Furthermore, the acceleration due to the force is inversely proportional to mass. Thus, the effect of the DC magnetic field on ion motion can be ignored.

[0060] Consider free electrons in a bulk plasma in the presence of a DC magnetic field. According to electromagnetism theory, the force acting on a free electron due to the DC magnetic field (generally called the Lorentz force) is oriented perpendicular to both the direction of its velocity and the direction of the local magnetic field. This causes the free electron to orbit in a helical trajectory along the magnetic field lines of the DC magnetic field. The radius of the helix is ​​generally called the Larmor radius. Generally, for a particle with charge q and mass m moving at velocity v in a DC magnetic field of strength B, the Larmor radius r is given by r = (mv) / (qB). In other words, each particle is confined within a region that decreases with increasing magnetic field strength and decreasing particle mass.

[0061] Therefore, by applying an external pulsed DC magnetic field, light mass free electrons are confined very effectively, and a similar effect on heavier ions can be ignored. The DC magnetic field does not change the kinetic energy of the free electrons. However, as is known to those skilled in the art, magnetic confinement of free electrons affects collisions with electrons such that the transport of electrons by collision (represented by the diffusion coefficient) and the transport of random components of electron kinetic energy (represented by the thermal conductivity coefficient) are reduced in the direction along the magnetic field lines (i.e., perpendicular to the direction of the pulsed DC magnetic field). In exemplary embodiments (systems 100 and 101), the magnetic field lines in the plasma run roughly from the top to the bottom of the chamber 110, consistent with the configuration of electromagnet 112A in system 100 (see Figure 1A) and electromagnet 112B in system 101 (see Figure 1B). Considering that the RF power from the first RF electrode 106 is absorbed by the plasma in the upper portion of the chamber 110, which is vertically below the first RF electrode 106, the transport of electron kinetic energy is reduced, resulting in T in the bulk plasma and in the pre-sheath of the lower portion of the chamber 110, which is closer to the substrate 102. e It is clear that it will decrease.

[0062] Using a pulsed DC magnetic field, time-varying T in the plasma above the substrate e A general method 200 for achieving this is described below with reference to the flowchart shown in Figure 2. One embodiment of method 200, in which the RF source signal is a CW RF signal, is described with reference to Figure 3A. Another embodiment of method 200, in which the RF source signal is a series of periodic source pulses, is described with reference to Figure 3B.

[0063] Referring to Figure 2, the plasma treatment method 200 for a substrate (e.g., substrate 102 in Figure 1A) includes generating plasma in a plasma chamber by flowing a discharge gas through a chamber within the chamber (box 202 in Figure 2), ionizing the gas (box 204 in Figure 2), and forming a glow discharge. In an exemplary embodiment of the plasma treatment system 100, the substrate 102 is held by a substrate holder 104 in the central region of the chamber 110, as shown in Figure 1A.

[0064] A controlled gas flow can be maintained using a gas flow system coupled to the chamber 110 in an exemplary embodiment of the plasma processing system 100 shown in Figure 1A. As shown in Figure 1A, the discharge gas is introduced into the chamber 110 through a gas inlet 162 and exits through a gas outlet 164. The gas outlet 164 is coupled to a vacuum pump 166 of the gas flow system. The discharge gas may be a gaseous mixture containing reactants, diluents, and additives. The gas pumped out through the gas outlet 164 may further contain volatile by-products formed in the chamber 110 during processing. The components of the gas flow system maintain a controlled low pressure in the chamber 110 and a controlled flow rate of various gases in the gas mixture that enter through the gas inlet 162, flow over the substrate 102, and exit through the gas outlet 164. In addition to the gas inlet 162, gas outlet 164, and vacuum pump 166, the gas flow system may include a gas canister, flow paths, throttle valves, gas flow sensors, and controllers.

[0065] The gas can be ionized by coupling an RF source signal to a first RF electrode (e.g., the first RF electrode 106 in Figure 1A), as shown in box 204. In an exemplary embodiment of the plasma processing system 100, the planar coil antenna is the first RF electrode 106, as shown in Figure 1A. The RF source signal output by the first electrical circuit 120 can be CW RF or a series of periodic RF pulses, as described above.

[0066] As described above, and as shown in box 206 in the flowchart of method 200 shown in Figure 2, the bias signal is applied to the second RF electrode in one of several ways. In the example of Figure 1A, the second RF electrode 108 is a conductive plate embedded in the body of the substrate holder 104, and as described above, the bias signal from the second electrical circuit 130 is coupled to the second RF electrode 108 using a switching matrix 160 configured by the controller 150. The bias signal is a periodic series of bias pulses, and the bias voltage waveform is applied only during the bias-on time. The bias pulses can be DC pulses, DC burst pulses, or RF pulses, depending on which bias voltage waveform is selected by the controller 150.

[0067] In method 200, a pulsed DC magnetic field is generated to modulate the electron temperature profile of the steady-state plasma. In other words, T e The magnetic field changes not only with respect to position but also with respect to time by applying a pulsed DC magnetic field. The pulsed DC magnetic field is generated by coupling a magnetization signal to an electromagnet, as shown in box 208 in the flowchart of method 200 shown in Figure 2, where the magnetization signal is a series of periodic current pulses. The exemplary system 100 in Figure 1A uses a plane electromagnet 112A, and the exemplary system 101 in Figure 1B uses a solenoid electromagnet 112B.

[0068] As mentioned above, in order to reduce the spread of ion angles and energy distribution, the T of the plasma above the substrate is e A low T is desirable, but in order to efficiently generate ions and radicals to maintain the desired plasma density, a high T is required. e This is desirable. The time-varying T in the plasma above the substrate. e Therefore, both requirements can be satisfied. Accordingly, embodiments of the present disclosure, as described above, involve T in the plasma above the substrate. e Modulate the low T e1 (Average T in the central region during the period when bias power is present) e) provides, T e High T e2 (Average T in the central region during the period when bias power is absent) e ) is restored to ). Thus, as described above, before the magnetization signal is coupled, the magnetization signal is synchronized to tune with the bias signal (box 210). In exemplary system 100 (Figure 1A) and exemplary system 101 (Figure 1B), the third electrical circuit 140 is controlled by the controller 150 to generate current pulses such that the DC magnetization current is supplied only during the bias-on time, i.e., at the same time that the second electrical circuit supplies the bias voltage waveform.

[0069] Figure 3A shows the waveforms of the source signal, bias signal, and magnetization signal applied in one embodiment of Method 200, where the RF source signal is a CW RF signal. As described above, the bias signal and magnetization signal are in phase; that is, a DC magnetization current flows in each current pulse at the same time that the bias voltage waveform is applied in each bias pulse. The magnetization signal, which is a periodic series of current pulses, forms a pulsed DC magnetic field that modulates the mean electron temperature in the central region. Modulated in this way by the pulsed DC magnetic field, the mean electron temperature in the central region of the chamber changes over time. The electron temperature versus time graph included in Figure 3A shows the time response of this mean electron temperature to the applied source, bias, and magnetization signal waveforms. As described above, the mean electron temperature in the central region of the chamber above the substrate is higher when there is no DC magnetic field (during bias-off time) and lower when there is a DC magnetic field (during bias-on time).

[0070] Figure 3B shows the waveforms of the source signal, bias signal, and magnetization signal applied in another embodiment of Method 200, where the RF source signal is a periodic series of source pulses, and each source pulse is an RF pulse. Similar to the embodiments described above with reference to Figure 3A, embodiments using a periodic series of source pulses (shown in Figure 3B) as the RF source signal also synchronize the bias pulse of the bias signal to the current pulse of the magnetization signal. The source signal, bias signal, and magnetization signal have the same pulse frequency. However, as shown in Figure 3B, in this exemplary embodiment, the source signal (including the periodic series of source pulses) is not synchronized to the bias pulse and the current pulse. The time response of the mean electron temperature in the central region of the chamber 110 on the substrate 102 in this embodiment, where the source signal is a periodic series of RF pulses as shown in Figure 3B, is also included in Figure 3B. Similar to the embodiment shown in Figure 3A, it is observed that the mean electron temperature in the central region may decrease with the onset of the magnetic field. However, source power is required to restore the mean electron temperature in the central region to a higher value.

[0071] In some other embodiments, the phase of the source pulse can be selected such that the RF pulse of the source signal exists only during the bias-off time, i.e., the pulse of the source signal is "180° out of phase" with respect to the bias pulse. In other words, the source-on time coincides with the bias-off time, and the source-off time coincides with the bias-on time. Of course, with respect to the current pulse of the magnetization signal, the bias signal is in phase with the magnetization signal.

[0072] Please understand that other waveform sets may also be used in Method 200, which is summarized in the flowchart shown in Figure 2.

[0073] Exemplary embodiments of the present invention are described below. Other embodiments can also be understood from the entirety of this specification and the claims submitted herein.

[0074] Example 1. A method for plasma processing a substrate, the method comprising generating plasma in a plasma chamber in which the substrate is held during processing, wherein generating plasma means flowing a discharge gas through the plasma chamber and coupling a radio frequency (RF) source signal to a first RF electrode, coupling means ionizing the discharge gas and coupling a bias signal to a second RF electrode, the bias signal being a series of periodic bias pulses, each period having a bias-on time and a bias-off time, with a bias voltage waveform applied during the bias-on time and coupling a magnetization signal to an electromagnet, thereby generating a pulsed DC magnetic field in the plasma chamber, the magnetization signal being a series of periodic current pulses, and synchronizing a series of periodic current pulses with the bias signal before coupling the magnetization signal, thereby flowing a DC magnetization current during the bias-on time.

[0075] Example 2. The method according to Example 1, further comprising configuring an electromagnet to form a DC magnetic field profile, which generates a first magnetic field profile during the bias-on time and a second magnetic field profile during the bias-off time, before coupling a magnetization signal, wherein the second magnetic field profile is different from the first magnetic field profile.

[0076] Example 3. The method according to one of Examples 1 or 2, wherein the plasma has an average electron temperature in a plasma chamber, which is a first electron temperature during the bias-on time and a second electron temperature during the bias-off time, the second electron temperature being greater than or equal to the first electron temperature, and the ratio of the second electron temperature to the first electron temperature increases as the magnitude of the DC magnetization current increases.

[0077] Example 4. The method according to one of Examples 1 to 3, wherein the RF source signal is a series of periodic source pulses, each source pulse is an RF pulse, and each period of the series of periodic source pulses has a source-on time and a source-off time, and an RF sinusoidal voltage waveform is applied during the source-on time.

[0078] Example 5. The method according to one of Examples 1 to 4, wherein the source-on time coincides with the bias-off time, and the source-off time coincides with the bias-on time.

[0079] Example 6. The method according to one of Examples 1 to 5, wherein the periodic series of bias pulses is a periodic series of DC pulses, applied during the bias-on time of each DC pulse, and the bias voltage waveform is at a constant DC voltage level.

[0080] Example 7. The method according to one of Examples 1 to 6, wherein the periodic series of bias pulses is a periodic series of DC burst pulses, applied during the bias-on time of each DC burst pulse, and the bias voltage waveform is a plurality of DC pulses.

[0081] Example 8. The method according to one of Examples 1 to 7, wherein the periodic series of bias pulses is a periodic series of RF pulses, and the bias voltage waveform applied during the bias-on time of each RF pulse is an RF sinusoidal voltage waveform.

[0082] Example 9. The method according to one of Examples 1 to 8, further comprising coupling a conductive portion of the wall of a plasma chamber to a reference potential.

[0083] Example 10. A system for plasma processing, comprising: a plasma chamber; a substrate holder configured to hold a substrate within the plasma chamber; a gas flow system configured to flow a discharge gas through the plasma chamber; a first radio frequency (RF) electrode; a second RF electrode configured to coordinate with the first RF electrode and the second RF electrode to generate plasma within the plasma chamber; an electromagnet configured to generate a pulsed DC magnetic field within the plasma chamber, wherein the pulsed DC magnetic field is configured to modulate the electron temperature profile of the plasma; a first electrical circuit configured to output an RF source signal; a second electrical circuit configured to output a bias signal; a third electrical circuit configured to output a magnetization signal; and a controller configured to transmit control signals to the first electrical circuit, the second electrical circuit, and the third electrical circuit to adjust and synchronize the RF source signal, the bias signal, and the magnetization signal.

[0084] Example 11. The system according to Example 10, wherein the controller comprises a processor and a memory for storing instructions, and when an instruction is executed in the processor, a first electrical circuit generates control signals for selecting a source signal, a second electrical circuit for selecting a bias signal, and a third electrical circuit for selecting a magnetization signal, and the source signal, bias signal, magnetization signal, and reference potential are synchronously applied to generate plasma in a plasma chamber.

[0085] Example 12. The system according to one of Examples 10 or 11, wherein the first RF electrode is a planar coil-shaped antenna positioned above a portion of the ceiling of the plasma chamber, with a portion of it being a dielectric window.

[0086] Example 13. The electromagnet is a planar electromagnet having a conductor in the shape of a planar coil around a central axis perpendicular to the center of the substrate holder, the conductor is positioned above the ceiling of the plasma chamber, and the inner diameter of the planar coil is larger than the diameter of the substrate, as described in one of Examples 10-12.

[0087] Example 14. The system described in one of Examples 10-13, wherein the electromagnet is a solenoid electromagnet having a spiral-shaped conductor wound around the outer side wall of the plasma chamber.

[0088] Example 15. The electromagnet comprises an electromagnet component comprising a first electromagnet including a conductor in the shape of a first planar coil having a first diameter, and a second electromagnet including a conductor in the shape of a second planar coil having a second diameter, wherein the first diameter is greater than the second diameter, as described in one of Examples 10 to 14.

[0089] Example 16. The system according to one of Examples 10 to 15, wherein the first electrical circuit includes an RF oscillator configured to generate a continuous wave (CW) RF voltage waveform, a chopper circuit configured to generate a periodic series of RF pulses in coordination with the RF oscillator when activated by a control signal from a controller, a power amplifier, and an impedance matcher.

[0090] Example 17. The system according to one of Examples 10 to 16, wherein the second electrical circuit comprises an RF oscillator configured to generate a continuous wave (CW) RF voltage waveform, a chopper circuit configured to generate a periodic series of RF pulses in coordination with the RF oscillator when activated by a control signal from a controller, a power amplifier, and an impedance matcher, and the third electrical circuit comprises a DC current source and a chopper circuit configured to generate a periodic series of current pulses in coordination with the DC current source when activated by a control signal from a controller.

[0091] Example 18. The system according to one of Examples 10 to 17, wherein the second electrical circuit includes a DC voltage source, a first chopper circuit configured to generate a periodic series of DC pulses in coordination with the DC voltage source when activated by a control signal from a controller, a second chopper circuit configured to generate a periodic series of DC burst pulses from a periodic series of DC burst pulses generated by the first chopper circuit when activated by a control signal from a controller, and a power amplifier, and the third electrical circuit includes a DC current source, and a chopper circuit configured to generate a periodic series of current pulses in coordination with the DC current source when activated by a control signal from a controller.

[0092] Example 19. A method for plasma treatment of a substrate, comprising: flowing a discharge gas through a chamber; coupling a radio frequency (RF) source signal to a first RF electrode; coupling a bias signal to a second RF electrode, wherein the bias signal is a periodic series of bias pulses, each period being the sum of bias-on time and bias-off time, and a bias voltage waveform is applied during the bias-on time; generating a pulsed DC magnetic field in the chamber by coupling a magnetization signal to an electromagnet, wherein the magnetization signal is a periodic series of current pulses; and, prior to coupling the magnetization signal, synchronizing a periodic series of current pulses with the bias signal to pass a DC magnetization current during the bias-on time.

[0093] Example 20. The method according to Example 19, wherein the electromagnet comprises a first electromagnet and a second electromagnet, and coupling the magnetization signal to the electromagnets includes simultaneously coupling a first magnetization signal to the first electromagnet and a second magnetization signal to the second electromagnet.

Claims

1. A method for plasma treatment of a substrate, wherein the method is The process includes generating plasma in a plasma chamber in which the substrate is held during processing, and generating the plasma is The discharge gas is passed through the aforementioned plasma chamber, The process involves coupling a radio frequency (RF) source signal to a first RF electrode, wherein the coupling includes ionizing the discharge gas. The coupling involves coupling a bias signal to a second RF electrode, wherein the bias signal is a periodic series of bias pulses, each period having a bias-on time and a bias-off time, and a bias voltage waveform is applied during the bias-on time. The method involves generating a pulsed DC magnetic field in the plasma chamber by coupling a magnetization signal to an electromagnet, wherein the magnetization signal is a series of periodic current pulses. A method comprising synchronizing a series of periodic current pulses with the bias signal so that a DC magnetization current flows during the bias-on time, before coupling the magnetization signal.

2. Generating the aforementioned pulsed DC magnetic field means The method according to claim 1, further comprising configuring the electromagnet to form a DC magnetic field profile that generates a first magnetic field profile during the bias-on time and a second magnetic field profile during the bias-off time, before coupling the magnetization signal, wherein the second magnetic field profile is different from the first magnetic field profile.

3. The method according to claim 2, wherein the plasma has an average electron temperature in the plasma chamber, which is a first electron temperature during the bias-on time and a second electron temperature during the bias-off time, the second electron temperature being greater than or equal to the first electron temperature, and the ratio of the second electron temperature to the first electron temperature increases as the magnitude of the DC magnetization current increases.

4. The method according to claim 1, wherein the RF source signal is a periodic series of source pulses, each source pulse is an RF pulse, and each period of the periodic series of source pulses has a source-on time and a source-off time, and an RF sinusoidal voltage waveform is applied during the source-on time.

5. The method according to claim 4, wherein the source-on time coincides with the bias-off time, and the source-off time coincides with the bias-on time.

6. The method according to claim 1, wherein the periodic series of bias pulses is a periodic series of DC pulses, and the bias voltage waveform applied during the bias-on time of each DC pulse is at a constant DC voltage level.

7. The method according to claim 1, wherein the periodic series of bias pulses is a periodic series of DC burst pulses, and the bias voltage waveform applied during the bias-on time of each DC burst pulse is a plurality of DC pulses.

8. The method according to claim 1, wherein the periodic series of bias pulses is a periodic series of RF pulses, and the bias voltage waveform applied during the bias-on time of each RF pulse is an RF sinusoidal voltage waveform.

9. The method according to claim 1, further comprising coupling a conductive portion of the wall of the plasma chamber to a reference potential.

10. A system for plasma processing, Plasma chamber and A substrate holder configured to hold the substrate within the plasma chamber, A gas flow system configured to flow discharge gas through the plasma chamber, A first radio frequency (RF) electrode and A second RF electrode, configured such that the first RF electrode and the second RF electrode work together to generate plasma in the plasma chamber, An electromagnet configured to generate a pulsed DC magnetic field within the plasma chamber, wherein the pulsed DC magnetic field is configured to modulate the electron temperature profile of the plasma, A first electrical circuit configured to output an RF source signal, A second electrical circuit configured to output a bias signal, A third electrical circuit configured to output a magnetization signal, A system comprising: a controller configured to transmit control signals to the first electrical circuit, the second electrical circuit, and the third electrical circuit in order to adjust and synchronize the RF source signal, the bias signal, and the magnetization signal.

11. The aforementioned controller, Processor and The system according to claim 10, further comprising a memory for storing instructions, wherein when an instruction is executed in the processor, the first electrical circuit generates a control signal for selecting a source signal, the second electrical circuit generates a control signal for selecting a bias signal, and the third electrical circuit generates a control signal for selecting a magnetization signal, and the source signal, the bias signal, the magnetization signal, and a reference potential are synchronously applied to generate the plasma in the plasma chamber.

12. The system according to claim 10, wherein the first RF electrode is an antenna in the shape of a planar coil positioned on a portion of the ceiling of the plasma chamber, the portion of which is a dielectric window.

13. The system according to claim 10, wherein the electromagnet is a planar electromagnet having a conductor in the shape of a planar coil around a central axis passing perpendicularly through the center of the substrate holder, the conductor is positioned above the ceiling of the plasma chamber, and the inner diameter of the planar coil is larger than the diameter of the substrate.

14. The system according to claim 10, wherein the electromagnet is a solenoid electromagnet having a spiral-shaped conductor wound around the outer side wall of the plasma chamber.

15. The electromagnet includes an electromagnet component, and the component is A first electromagnet comprising a conductor having a first diameter and a shape like a first planar coil, The system according to claim 10, comprising a second electromagnet including a second conductor having a second diameter and having a shape like a second planar coil, wherein the first diameter is greater than the second diameter.

16. The first electrical circuit is, An RF oscillator configured to generate a continuous wave (CW) RF voltage waveform, A chopper circuit configured to generate a series of periodic RF pulses in cooperation with the RF oscillator when activated by a control signal from the controller, Power amplifier and The system according to claim 10, comprising an impedance matching device.

17. The second electrical circuit is, An RF oscillator configured to generate a continuous wave (CW) RF voltage waveform, A chopper circuit configured to generate a series of periodic RF pulses in cooperation with the RF oscillator when activated by a control signal from the controller, Power amplifier and Equipped with an impedance matching device, The third electrical circuit is, DC current source and The system according to claim 10, further comprising: a chopper circuit configured to generate a series of periodic current pulses in cooperation with the DC current source when activated by a control signal from the controller.

18. The second electrical circuit is, DC voltage source and A first chopper circuit, configured to generate a series of periodic DC pulses in cooperation with the DC voltage source when activated by a control signal from the controller, A second chopper circuit, which is configured to generate a periodic series of DC burst pulses from the periodic series of DC burst pulses generated by the first chopper circuit when activated by the control signal from the controller, Equipped with a power amplifier, The third electrical circuit is, DC current source and The system according to claim 10, further comprising: a chopper circuit configured to generate a series of periodic current pulses in cooperation with the DC current source when activated by a control signal from the controller.

19. A method for plasma treatment of a substrate, The discharge gas is passed through the chamber, The radio frequency (RF) source signal is coupled to the first RF electrode, The coupling involves coupling a bias signal to a second RF electrode, wherein the bias signal is a periodic series of bias pulses, each period being the sum of the bias-on time and the bias-off time, and a bias voltage waveform is applied during the bias-on time. The method involves generating a pulsed DC magnetic field in the chamber by coupling a magnetization signal to an electromagnet, wherein the magnetization signal is a series of periodic current pulses. A method comprising synchronizing a series of periodic current pulses with the bias signal before coupling the magnetization signal, thereby passing a DC magnetization current during the bias-on time.

20. The electromagnet comprises a first electromagnet and a second electromagnet component, The method according to claim 19, wherein coupling the magnetization signal to the electromagnets includes simultaneously coupling a first magnetization signal to the first electromagnet and a second magnetization signal to the second electromagnet.

Images