High efficiency plasma source with chemistry conversion

Abstract

An apparatus for plasma treatment comprises a plasma source and a treatment chamber, which is arranged below the plasma source and separated from the plasma source by a grid. The plasma source for plasma treatment comprises: an outer chamber for plasma generation formed between the outer dielectric wall and a sidewall of a dual gas injection insert, with one injection port at the top of the insert and the second injection port at the lower part of the insert; an RF antenna for the first gas plasma generation along the periphery of the outer chamber positioned between injection ports; a conversion zone below the second injection port near the bottom of the insert for conversion the first plasma into the second plasma; a plasma transfer chamber below the insert and above the grid. A method for efficient plasma and radicals generation using generation of a first gas plasma followed by its conversion into another gas plasma.

Description

High Efficiency Plasma Source with Chemistry Conversion PRIORITY CLAIM

[0001] The present allocation claims the benefit of priority of U.S. Provisional Patent Application No.63 / 654,964, entitled “High efficiency plasma source with chemistry conversion,” filed on June 1, 2024, which is incorporated herein by reference for all purposes. CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] The present disclosure relates generally to plasma generation for semiconductor plasma processing and, more particularly, to a plasma treatment and dry strip. PRIOR ART

[0003] K. Obuchi, “Apparatus for plasma treatment”, US 6,770,165, (2004)

[0004] V. Nagorny, C. Crapuchettes, "High efficiency plasma source", US 9,214,319, (2015).

[0005] V. Nagorny, W. Liu, R. George, "Plasma source and plasma processing apparatus thereof", US 11,658,006, (2023).

[0006] V. Nagorny, V. Vaniapura, V. Surla, "Validation of High Efficiency ICP Source Performance for Advanced Resist Ashing", Proc. of Advanced Semiconductor Manufacturing Conference ASMC 2015, 978-1-4799-9930-9 / 15 / $31.00 ©2015 IEEE BACKGROUND OF THE INVENTION

[0007] Plasma processing is widely used in the semiconductor industry for deposition, etching, mask removal, plasma treatment and related processing of semiconductor wafers and other substrates. Inductive plasma sources are often used for plasma processing to produce high density plasma and reactive species for processing wafers. For instance, inductive plasma sources can easily produce high density plasma using standard 13.56 MHz and lower frequency power generators.

[0008] For certain plasma processes such as a plasma treatment, or a dry strip it is not desirable to expose the semiconductor wafers directly to plasma ions. In these processes, a plasma source is used mainly as an intermediate for modification of a gas composition and to create some chemically active radicals for processing wafers. The plasma is formed remotely from the processing chamber and desired particles are channeled to the semiconductor waferfor example through a grid that is transparent to neutral particles and not transparent to charged particles. As long as one accepted the use of the grid, one can also use it for providing process uniformity on the wafer. So, here we will assume that one uses the grid and the process uniformity is not an issue.

[0009] Because rates of processes on a wafer are directly proportional to the rate with which these new species are created and delivered to the surface of the wafer, these processes typically require high RF power (e.g. about 3-10 kW) and in some cases high gas flows (e.g. about 5-25 slm) and high pressure (e.g. about 1 Torr or higher) and a large diameter of the plasma source. When gas and energy consumption demands are that high, the efficiency of the plasma source becomes increasingly important. Efficiency of the plasma source affects both capital and operational costs. In addition, large size at increased pressure creates additional problems for plasma ignition and sustaining. In order to overcome these problems earlier were suggested similar, but slightly different solutions (US 6,770,165, and US 9,214,319). The first one (US 6,770,165) suggested to use insert, to create a separate plasma generation region near antenna with reduced volume - that would increase the power density in that region and simplify both plasma ignition and sustaining. The second one (US 9,214,319) suggested that choosing a correct gap between the insert and the wall will make all the feed gas to flow through the region with high density of energetic electrons and then one can optimize the efficiency of radical generation by adjusting the length of the insert $L$ (position of the bottom of the insert) that controls the size of the active region. Typically that would require a positioning of antenna near the bottom of the insert (Nagorny, et al., ASMC2015).

[0010] Besides high efficiency one often needs a high power to reduce the process time, and the latter sometimes creates a different problem - high power at high gas pressure create an intense localized heat load on the wall near antenna (ICP coil), resulting in a very high temperature and temperature gradients of the wall near the coil. That makes a quartz, which has very low thermal expansion a preferred material for the external wall of high pressure ICP plasma sources. In any other material, like ceramics, thermal expansion causes so large stresses that they can survive only relatively low power, which strongly limits the process capabilities. The problem is that some process chemistries are incompatible with quartz. For example, light etch or some treatment processes require fluorine or hydrogen radicals that etch quartz and generate neutral contaminants. These contaminants react with the surface, modifying a film and affecting the film properties.

[0011] Thus, a need exists for an efficient and high power plasma source and methods that can minimize generation or arrival of undesired chemical species (contaminants) to the wafer while providing a large supply of desired reactive species. SUMMARY OF THE INVENTION

[0012] This invention relates to high pressure plasma sources, where plasma generation is localized in the close vicinity to the coil. The present invention focuses on maximizing generation of desired radicals while minimizing their reactivity with walls. This is achieved by splitting the process of radical generation in the outer chamber into two steps by creating first one kind of plasma in the plasma generating zone, and then converting that plasma into a second plasma rich on radicals in a plasma conversion or radical generation zone. The two zones are separated by a wall cooling area. Plasma, rich on desired radicals further enters the plasma volume, where all generation processes are weak, and the main process is transport of plasma species toward the processing chamber, separated from the plasma chamber by a grid.

[0013] In at least one embodiment a plasma source includes a first sidewall and a gas injection insert defining a plasma source interior volume, including plasma generation volume. The gas injection insert includes a first (top) peripheral gas injection port, a second (bottom) peripheral gas injection port, and an induction coil disposed around the first sidewall and between the first and the second gas injection ports. Both gas injection ports inject gases into different parts of the interior of the plasma generation volume between the insert and the sidewall. The plasma source further includes a radio frequency power generator coupled with the induction coil.

[0014] In at least one embodiment a radical source includes a first sidewall and a gas injection insert defining a plasma source interior volume, including plasma generation and plasma modification volume. The gas injection insert includes a first (top) peripheral gas injection port, a second (bottom) peripheral gas injection port, and a radio frequency antenna disposed around the first sidewall and between the first and the second gas injection ports. Both gas injection ports inject gases into different parts of the interior of the plasma generation volume. The plasma source further includes a radio frequency power generator coupled with the induction coil. The plasma source separated from the processing chamber by a grid.

[0015] The first feed gas (Gas1) may be one of inert gases like Ar orHe , so that both theplasma generation is very efficient and species from that gas do not react with the sidewalls even at high wall temperatures. The Gas1 passes the RF antenna, which efficiently generatesfirst plasma from this gas. The walls near antenna may become hot, but the first plasma species do not react with the walls. At some distance downstream below the antenna, where walls cooled down enough so that reactivity of desired radicals with walls drops to the level that doesn't result in the significant number of contaminants, the second gas (Gas2), which radicals is the purpose of the source, is introduced through the second port of the outer chamber into a plasma conversion or radical generation zone. Since injection is within this chamber, there will be high probability for Gas2 molecules to react with metastable and ion species created in the plasma generation zone. So, if metastable energy of the Gas1 plasma species is between ionization and dissociation energy of the Gas2 molecules, than these collision will result in a very efficient generation of desired Gas2 radicals and will not cause any undesirable wall reactions, because of the lower wall temperature around the radical generation / conversion region. In many situations ions of the Gas1 will exchange charges with Gas2 molecules, so at the exit from conversion zone one obtains mostly Gas2 ions and radicals.

[0016] This scheme makes it a very efficient and process safe source for radical generation. The source optimization requires adjusting the gap between the coil and the second gas injection port (wall cooling area) and the length of the main (conversion) zone - between the second injection port and the bottom edge of the insert. BRIEF DESCRIPTION OF DRAWINGS

[0017] The appended drawings only illustrate exemplary embodiments and are not to be considered limited of its scope.

[0018] Fig.1 is a schematic diagram of a plasma processing system, according to at least one embodiment.

[0019] Fig.2 is a schematic diagram, explaining the principle of the process safety with regard to etching the sidewall.

[0020] Fig.3 is a schematic diagram of a plasma processing apparatus, according to at least one embodiment.

[0021] Fig.4 is a schematic diagram of a plasma processing apparatus, according to at least one embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0022] There are two main features in this invention. The first is efficiency of radical generation and the second is a process safety with respect to etch product from the sidewall.

[0023] The efficiency is based on a very efficient process of plasma generation in noble gases and then efficient conversion of that plasma into the plasma with required radicals. The first feed gas (Gas1) is chosen from one of inert gases, like argon or helium, so that plasma generation is very efficient and species from that gas do not react with the sidewalls even at high wall temperatures. The Gas1 passes the RF antenna, which efficiently generates the first plasma from this gas. At some distance downstream below the antenna, where the energetic electrons density is significantly reduced, and the wall temperature is lower than near the antenna, the second gas (Gas2), which radicals is the purpose of the source, is introduced through the second port of the outer chamber into a plasma conversion or radical generation zone. Since injection is within this chamber, there will be high probability for Gas2 molecules to react with metastables and ion species created in the plasma generation zone. So, if metastable energy of the Gas1 plasma species is between ionization and dissociation energy of the Gas2 molecules, then these collision will result in a very efficient generation of desired Gas2 radicals and will not cause any undesirable wall reactions, because of the lower wall temperature around the radical generation / conversion region. In many situations ions of the Gas1 will exchange charge with Gas2 molecules, so at the exit from the conversion zone one obtains a second plasma with mostly Gas2 ions and radicals in the mixture of feed gases Gas1 and Gas2.

[0024] Since the main purpose of the source are radicals, the main requirement to the Gas1 is that excitation level of its metastables Em 1 is between ionization (Ii 2 ) and dissociation energies(E dis ) of the processing gas (Gas2): Edis < E m 1 < I i 2 . Often ionization energy of Gas1 Ii 1 is alsothan Ii 2 (Ii 1 > I i 2 ), so the chargereactions between ions of Gas1 and Gas2 (aswell asother ionic species) also takes place. For example, for argon Em 1 = 11.55 eVand Ii 1 = 15.755 eV , for hydrogen and oxygen dissociation energies are4.77eV ( H 2 ) ,5.19eV ( O 2 ) and ionization energies are15.5eV ( H 2 ) , 12.26eV ( O 2 ) . So, using argon as apriming gas (Gas1) will result in an efficient generation of all these radicals with complete conversion of argon plasma into the Gas2 (or mix Gas1-Gas2) plasma. Note, that it takes significantly less discharge power to ionize and sustain plasma in argon or helium than in any processing gas.

[0025] Let for example, choose Ar as the Gas1. As Ar passes the plasma generating region(coil), the Ar plasma is generated, and sustained in the vicinity of the coil where the mainreactions are electrons, ions and metastable generation, by collisions between hot electrons and argon atoms. e+ Ar → Ar+ +2 e , e + Ar → e + Ar *(all higher excited argon atoms quickly radiate and become metastables or lowest resonant states). Argon plasma especially at high pressure is very rich on Armetastables.

[0026] Argon metastables, colliding with argon atoms simply exchange their state without losing excitation Ar *+ Ar → Ar + Ar * ,they have very low diffusion coefficient, so they move with the flow and finally reach the second (conversion region), where the second gas is being introduced.

[0027] In the conversion region the main processes changing one particle to another are collisions between argon ions and metastables with Gas2 molecules. These are dissociation and charge exchange processes (for some gases). When colliding with most of the molecules M 2(like H 2 , O 2 , ...), argon metastables easily dissociate those molecules, and argon ionsexchange charge with Gas2 molecules: Ar*+ M + +2 → Ar + 2 M , Ar + M 2 → Ar + M 2 .

[0028] For example, if the second gas is hydrogen, then main reactions in the conversion zone are Ar * + H + +2 → Ar + 2 H , Ar + H 2 → Ar + H 2 ,hence, initially argon plasma is converted into mostly hydrogen plasma with argon atoms.

[0029] This scheme makes it a very efficient source for radical generation. The source optimization requires adjusting the gap between the coil and the second gas injection port (wall cooling area) and the length of the main (conversion) zone - between the second injection port and the bottom edge of the insert.

[0030] The second feature is a process safety with respect to etch products from the sidewall. The basis for this feature is the temperature dependence of a chemical reactivity r ws betweenwalls and species. Reactivity follows Arrhenius formula rws ∝ exp( − T a / T w ) , where weintroduced an activation temperature Ta = E a / R for reaction between the wall w and speciess , T w is the wall temperature, E a and R are activation energy for that reaction and the gasconstant. For example, for etching quartz reaction between quartz wall and hydrogen this activation temperature is 15396K. This temperature is much higher than any reasonable wall temperature Ta / T w >> 1 , which means that the reaction rate is very sensitive to the walltemperature. For example, if the temperature in the hottest point is Tw ,max = 1300 K , then atthe point where the wall temperature is dropped by only 211K (Tw = 1089 K ), the reactionrate with the wall will be reduced by a factor of 10 and only another 152K (Tw = 937 K ) willreduce it by another factor of 10. Thus, placing the second injection port at some distance from the coil, so that radical reactivity with walls is low enough one can always achieve acceptable etch conditions. The distance between the coil and the second injection port is the controlling factor for the process safety. Depending on the hottest wall temperature, the acceptable temperature where one can ignore reactions between walls and radicals, and cooling efficiency, this distance can be anywhere from 0.5 to 10cm. More details are explained in the Fig.2.

[0031] Of course, there is a loss of Gas1 metastables as one increases the gap between the plasma generation zone (coil) and the second injection port, which defines the beginning of the conversion / radical generation zone, so for any of Gas2 that does not react with the wall, the cooling zone must be minimized. For gases like H 2 or ammonia, generating a lot of hydrogenradicals the size of the cooling zone must be minimized with a condition describing maximum allowed plasma contaminants caused by the etching of the quartz sidewall.

[0032] Essentially, the plasma generation zone in this approach is a precursor generation zone, where the precursor is a metastable or lowest resonant species of Gas1 (e.g. Ar ), and the 2nd(conversion) zone is the main – radical generation zone. Only radicals and ground state atoms and molecules will pass the grid to the processing chamber and reach the wafer. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT

[0033] While one can come up with many embodiments (different coils, Faraday shield, ways of cooling O-rings, etc.), we will focus on detailed description of a few variations of the preferred one.

[0034] Fig.1 is schematics of the system 100, which has a plasma source 120 sitting on a top plate 110 and a processing chamber 150. A grid 151 separates the processing chamber 150 and the plasma chamber 130. The processing chamber has a pedestal 160 for holding a substrate, sidewalls and exhaust 155. Pedestal may have heating capabilities.

[0035] Plasma source 120 has a plasma chamber 130 confined by the dielectric chamber wall 122 from the outside, walls of the insert 140 and the grid 151. Plasma chamber consists of 2 parts - the outer chamber 131 between the dielectric chamber wall of the source and the sidewall of the insert 140 and the bottom chamber 132 - below the insert and above the grid 151. The material for the chamber wall without limitation may be, for example a fused quartz or ceramics.

[0036] The insert 140 has the first gas injection inlet 141 around the top of the outer chamber, the first gas distribution conduit 142 and injectors 143 from the conduit into the outer chamber. The second gas injection consists of the inlet 145, gas distribution conduit 146 and multiple injector channels 147. The number of injector holes of injectors 143 and 147 must be large enough to create more or less uniform gas injection. That number depends on the radius of the source, but the distance between adjacent injector holes 0.5cm-1cm, or 1cm-2cm should be good enough. The specific design of injectors 143 is not important. All it has to provide is a gas flow from the top of the outer chamber to the bottom chamber passing the induction coil. For example, instead of multiple holes 143 one may use a narrow slit between the dielectric sidewall and the insert as in US 9,214,319.

[0037] The outer chamber 131 has 2 zones - Gas1 plasma generation zone 133 around the coil 171, where the first gas coming through injectors 143 is ionized and the first gas plasma is sustained, the Gas2 radical generation or plasma modification zone 134, where the first gas metastables collide with Gas2 molecules and dissociate Gas2. The space L 1 between these twozones is a wall cooling area.

[0038] The plasma source has an induction coil 171 connected to an RF power supply 172 through a matching network 174, and a combination shield-Faraday shield 121, which is configured to operate as a Faraday shield near the coil and as a shield slightly below the coil. Openings in the Faraday shield part limit the length L 0 of the plasma generation zone 133, butthey have to be larger than the vertical size of the coil, to reduce losses in the shield.

[0039] The size of the cooling zone L 1 can be adjusted by moving the coil, or / and by changingthe vertical position of the second injection ports, which we will discuss later.

[0040] Material for the insert can be aluminum or any metal, which allows water cooling of the insert and improves efficiency of the plasma ignition. The water cooling can be embedded, for example, in the second gas injection plate 144.

[0041] Figs.2a-2c demonstrate the role of the second injection on the etching of the sidewall and its dependency on the size of the cooling area. Fig.2a shows qualitatively a normalized temperature profile Tn( x )= T ( x ) / T max ( x ) (dashed line) and normalized reactivity betweenradicals and walls rn( x )= r ws ( T ( x )) / r ws ( T max ( x )) (dotted line) along the vertical coordinate xparallel to the overall gas flow between the insert and the chamber wall in the vicinity of the coil. The plasma generation zone here is approximately (4.5-7.5) in arbitrary units (a.u.) shown as shaded area above the axis. For this plot we chose hydrogen for the processing gas, the maximum wall temperature 1260K and the activation temperature 15396K. Note, that the wall temperature profile is mainly determined by a plasma generation near the coil, so we will assume for simplicity that in case of a single or dual gas injections these values are the same, and thus reactivity, which follows the temperature is the same in both cases, so for the reactivity r ws we will use the same profile shown in Fig. 2a.

[0042] Effect of using a dual gas injection vs a single gas injection on the etching the sidewalls by radicals is demonstrated in Figs.2b, and 2c assuming the same temperature profile shown in Fig. 2a. The local etch rate ER ( x ) is proportional to the production of the radical flux toward tothe wall FH ( x ) and reactivity rws ( x ) : ER ( x )= FH ( x ) r ws ( x ) , so the total rate of generatingcontaminants P cont in the source isPcont = 2π R w ^ ER ( x ) dx ,where R w is the sidewall radius.for the local rate can berewritten as ER( x ) = ( FH ,max r ws ,max ) × ( F n ( x ) r n ( x )) , where Fn ( x ) is normalized profile of thehydrogen radicalsthe wall, so the Eq. (5) becomes Pcont = 2π R w ( F H ,max r ws ,max ) ^ F n ( x ) r n ( x ) dx .What follows from theand the production of radicals are the same, the total generation of contaminants will depends on the area under the Fn ( x ) r n ( x ) curve or on the overlapping of Fn ( x ) and rn ( x ) profiles.

[0043] In case of a single gas injection 141-143 the hydrogen radical flux to the wall Fn( x ) = F H ( x ) / F H ,max reaches saturation (maximum) in the area of maximum reactivity (Fig.2b), so the area under the production Fn ( x ) r n ( x ) curve and thus the etching of the wall arelarge. The 3 lines on this plot show the Fn ( x ) profile (dashed line), the normalized reactivityprofile (dotted line) and etch rate profile Fn ( x ) r n ( x ) (solid line), with rn ( x ) taken from the Fig.2a.

[0044] In case of the dual injection (Fig.2c) hydrogen radicals appear in plasma only after the second gas injection 147, where reactivity is significantly lower. For this plot we chose the position of the second injection ports 147 right below the coil (approximately 8 a.u.). The production Fn ( x ) r n ( x ) curve (solid line) in this case results in significantly smaller area. For theparameters we chose, the areas under these "production" curves, which characterize total production of contaminants, differ by a factor of larger than 20. The plasma conversion zone according to this figure should expand to at least from 8a.u. to 12 a.u. No more conversions occur beyond that point (12 a.u.). In reality one can experiment (or use simulations) to determine the size of the conversion zone that fits all processes planned for that tool.

[0045] Of course, plasma in different gases passing the coil generates different heat load on the wall which results in different maximum temperatures or temperature profiles, so the actual benefits of having dual gas injection may be somewhat different from those used in example shown in the Figs. 2a-2c. However the principle is correct and one can optimize the distance L 1for the acceptable radical production vs contamination level.

[0046] Fig. 3a shows a version of the insert, where the distance L 1 between the coil and thesecond injection port as well as the size of thegeneration zone L 2 can beeasily adjusted by using simple spacers 341, 340 and disk 342. In this picture the spacer 341 - is a simple spacer, the spacer 240 has machined grooves 147 on its surface, which form injection channels when 340 is pressed against the 341, and the disk 342 together with spacer 340 form the conversion zone.

[0047] Fig.3b shows details of this version of the injection ports 147. This design is much more economic and flexible way of making these injection ports, compared to drilling holes in the insert walls. One can easily change the number and the size of these ports by simply replacing one piece 340 by another. The grooves 147 can be made on both sides of the spacer 340, in which case the second gas injection will be more distributed along the Gas1 path further reducing the number of contaminants.

[0048] Fig.4 shows the source with exactly the same insert design, but uses coil with ferrite concentrators 471 to reduce the size of the plasma generation zone. In the case of ferriteantenna, the only limitation on the size of openings in the Faraday shield (if one is used) is that the length of openings must be larger than the vertical size of the ferrites.

[0049] The distance L 1 between the coil and the second injection port has to be adjustedbased on the temperature in the hottest spot near the coil (T max ) and the acceptabletemperature for second injection point (T 2 ) so that the integral under the ER ( x ) curve (seeFigs.2b-2c) is small enough to ignore reactions between Gas2 radicals and the walls. That distance of course is affected by the value ofT max , the wall cooling rate, which determines thetemperature profile along the wall and the allowed level of contaminants for that particular process. Accordingly, L 1 can be between 0 and 10cm, including between 0 and 3cm, 2cm and5cm.

[0050] The length L 2 of the conversion zone must be adjusted because there could besignificant difference in the flow pattern between metastables of the Gas1 and molecules of the Gas2. For example, if the first gas is argon and the second one is hydrogen, the diffusion coefficients of Ar * metastables is about 10 time smaller than that of H 2 in the same, mostlyargon gas. The only way to force them to be together until most of argon metastables are converted to hydrogen radicals is to increase the distance where they flow together. But, of course, that property depends on the combination of Gas1 and Gas2. Accordingly, one can expect L 2 to be anywhere between 0.5 cm and 7 cm, including 0.5 cm and 3 cm, 2 cm and 7 cm.

[0051] Although the initial purpose for this invention was to protect the most thermal stress resistant material for the plasma source sidewall - fused quartz or alike materials from a chemical etch by plasma products in the area of plasma generation, the dual gas injection scheme also helps protecting more chemical resistant but less thermal stress resistant materials like alumina, or other ceramics for the main source wall. For example, ceramics (alumina ( Al 2 O 3 ), yttria (Y 2 O 3 ), zirconia (ZrO 2 ) and others) is almost the only choice to withstandfluorine chemistry, which etches quartz even at low temperature. However, discharges in noble gases generate lower and less localized heat load on the sidewall, resulting in much smaller thermal stress. In addition, generation of the same amount of radicals F in this case) requiresmuch less power, which results in additional reduction of the heat load and thermal stress on the sidewall.

[0052] While we have considered here only noble gases ( Ar , He ) for the Gas1 in some casesone can replace a noble gas with the nitrogen for the Gas1 if used with process gases having low dissociation (and maybe ionization) energy. For the processing feed gas Gas2 one may choose any of H 2 ,O 2 , H 2 O ,CH 4 , N 2 , NH 3 ,CF 4 , NF 3 ,C 2 F 4 ,CHF 3 , SF 6 and many othermolecular gases.

Claims

AMENDED CLAIMS received by the International Bureau on 15 September 2025 (15.09.2025)1. A plasma source comprising: a first dielectric sidewall forming a cylinder; a gas injection insert disposed within the cylinder forming a second sidewall of the diameter smaller than the first sidewall, and located at an upper end of the cylinder, the gas injection insert and the first sidewall defining a plasma source interior volume, including a plasma generation and a plasma conversion volumes between the first and the second sidewalls, the first sidewall cooling area between the first and the second sidewalls, separating plasma generation and conversion zones; the gas injection insert comprising: a first gas injection port coupled to the top part of the insert, the second gas injection port disposed at the bottom part of the insert between the sidewall cooling area and the plasma conversion zone; and a radio frequency antenna disposed around the first sidewall, and located between the first and the second gas injection ports.

2. The plasma source of claim 1, wherein the radiofrequency antenna is an inductive coil.

3. The plasma source of claim 2, wherein the first sidewall cooling area between the bottom of radio frequency antenna and the second injection port is in the range of about 5mm to about 100mm.

4. The plasma source of claim 1, wherein the plasma conversion zone between the second injection port and the bottom edge of the insert sidewall is in the range of about 5mm to about 50mm.

5. The plasma source of claim 2, wherein the inductive coil uses magnetic field concentrators.

6. The plasma source of claim 2, wherein the Faraday shield limiting the plasma generation zone is placed between the first dielectric sidewall and the inductive coil.

7. The plasma source of claim 1, wherein the gas injection insert comprises metal.

8. A plasma source comprising: a first sidewall forming a dielectric cylinder; a gas injection insert disposed within the cylinder and located at an upper end of the cylinder; the gas injection insert and the first sidewall defining a plasma source interior volume around the insert and inside the first sidewall, including a process plasma generation volume between the first sidewall and the insert sidewall;the process plasma generation volume comprising four sequential zones - a gas injection zone starting from a first gas inlet at the top of the plasma generation volume, followed by a first gas plasma generation zone, followed by a first sidewall cooling zone and finally a plasma conversion zone starting from the second gas inlet and ending at the bottom edge of the insert; a radio frequency antenna disposed around the first sidewall, located between the first and the second gas injection ports and defining the first gas plasma generation zone;9. The plasma source of claim 8, wherein the radiofrequency antenna is an inductive coil.

10. The plasma source of claim 9, wherein the first sidewall cooling area between the bottom of radio frequency antenna and the second injection port is in the range of about 5mm to about 100mm.

11. The plasma source of claim 8, wherein the plasma conversion zone between the second injection port and the bottom edge of the insert sidewall is in the range of about 5mm to about 50mm.

12. The plasma source of claim 9, wherein the inductive coil uses magnetic field concentrators.

13. The plasma source of claim 9, wherein the Faraday shield limiting the plasma generation zone is placed between the first dielectric sidewall and the inductive coil.

14. The plasma source of claim 8, wherein the gas injection insert comprises metal.

15. A method for generating a plasma for processing a substrate in a plasma source having (with) a gas injection insert that divides a plasma chamber into an outer chamber defined by a dielectric sidewall of the plasma source and a sidewall of the insert for generation of the processing gas and a bottom chamber defined by the dielectric sidewall of the plasma source below the insert and above a grid separating the plasma chamber from a processing chamber comprising: generation of processing gas in the plasma chamber in two steps, wherein the first step comprises making a downstream gas flow in the outer chamber by injecting a first gas into the outer chamber through the top injection port above a radio-frequency antenna and energizing that antenna to generate the first gas plasma in the plasma generation zone of the outer chamber adjacent to the antenna; and the second step comprises injecting a second gas into a plasma conversion zone of the outer chamber for converting the first gas plasma into the second gas plasma; said second gas injection port is positioned downstream from the coil / first plasma generation area at a wall cooling distance.

16. The method of claim 15, wherein the radiofrequency antenna is an inductive coil.

17. The method of claim 15, wherein the first gas injected through the first injection port is mainly a noble gas, like argon and the second gas injected through the second port is a processing feed gas.

18. The method of claim 15, wherein the processing gas includes some of H2,N2, NH3, CH4, o2, H2O, CHF3, CF, , NF3, SF6.

19. The method of claim 15, wherein the plasma conversion zone between the second injection port and the bottom edge of the insert sidewall is in the range of about 5mm to about 50mm.

20. The method of claim 15, wherein the first sidewall cooling area between the bottom of radio frequency antenna and the second injection port is in the range of about 5mm to about 100mm.

21. A method of processing substrate comprising: placing a substrate on a substrate holder in a processing chamber, separated from the plasma chamber by a grid and generating the processing plasma using the method of claim 15.Statement under Article 19(1)I appreciate a great work of the Examiner of the ISA, with very good examples demonstrating ambiguities and other problems with my initial set of Claims. Following his criticism, I'd like to cancel ALL my previous Claims and replace them with new sets (14 Claims for the Apparatus and 7 Claims for the Method).

Images