PLASMA DENSIFICATION METHOD

Abstract

The plasma is formed between electrodes to be energized from an electric power source, containing a partially ionized mass having a luminescence region including neutral atoms (NA), primary electrons (PE), secondary electrons (SE), and ions.

The method comprises the interspersed steps of: accelerating the primary electrons (PE) toward one of said electrodes (10) polarized by a short, positive, high voltage pulse, impacting primary electrons (PE) against said electrode (10) and ejecting secondary electrons (SE) from it; subsequently, accelerating the secondary electrons (SE) toward the luminescence region by polarization of said electrode (10) by a negative voltage with a lower voltage pulse colliding the secondary electrons with neutral atoms (NA) and producing positive ions (PI) and derived electrons (DE); the negative pulse must have a period of time sufficient to accelerate the positive ions (PI) of the luminescent region towards the electrodes 10, striking the surface of said electrodes; repeating the previous steps in order to obtain a steady state plasma with a desired degree of ionization. The control of the intensity and the period of the positive and negative pulses allow the control of the degree of ionization and the volume of the luminescent region of the plasma.

Claims

1. A plasma densification method of the type formed between electrodes (10) to be energized from an electric power source (20), said plasma containing: a partially ionized mass, which may have a region of luminescence, and including neutral atoms (NA) of an ionizable fluid, primary electrons (PE), positive ions (PI) and possible negative ions (NI); and with the method being characterized by the following intercalated steps: accelerating the primary electrons (PE), contained in the partially ionized mass, towards the surface of one of said electrodes (10) by means of a positive high voltage short pulse applied to the electrode (10), with energy enough to impact at least part of the primary electrons (PE) against the surface of said electrode (10) and to eject secondary electrons (SE); accelerating the secondary electrons (SE), and any other electrons (PE) in the vicinity of said electrode (10) toward the luminescence region, by biasing said electrode (10) with a longer and lower negative voltage bias pulse, with energy enough to collide them with neutral atoms (NA) and produce positive ions (PI) and derived electrons (ED); and, continuously repeating the previous biasing steps over the electrodes (10) with positive and negative bias pulses, in order to generate a stable plasma, in steady state, with a desired degree of plasma density, whose degree of ionization is controlled by the intensity and period of applied voltage pulses.

2. A method, according to claim 1, characterized by the fact that the positive bias pulse (30) has a voltage and sufficient duration to kinetically energize at least part of the primary electrons (PE) from the partially ionized mass, colliding them with the electrode (10), but with insufficient power to increase the kinetic energy of the positive ions (PI), not moving them away from the region in which they are in the ionized mass, keeping them near to the electrode 10, while the positive bias pulse is applied, or immediately after.

3. A method, according to claim 2, characterized by the fact that the positive bias pulse (30) has a voltage in the range of hundreds up to thousands of volts and a duration of nanoseconds up to tens of microseconds.

4. A method, according to claim 1, characterized by the fact that the negative bias pulse (40) has a voltage and duration sufficient to kinetically energize the secondary electrons (SE), ejected from the electrode (10), and any other electrons (PE) in the region of luminescence of the plasma, producing positive ions (PI) and derived electrons (ED) by collision with neutral atoms (NA) in the region of luminescence, while the negative bias pulse is applied, or immediately after.

5. A method, according to claim 4, characterized by the fact that the negative bias pulse (40) has a voltage in the range of tens up to hundreds of volts and a duration of microseconds, but greater than that of the positive bias pulse.

6. A method, according to claim 4, characterized by the fact that the negative bias pulse (40) has sufficient duration to accelerate the positive ions (PI), produced in the region of plasma luminescence, toward the electrodes (10), promoting ionic bombardment of the positive ions (PI) against the electrodes (10).

7. A method, according to claim 1, characterized by the fact that any one of the positive (30) or negative (40) polarizations is initiated at the end of the other polarization.

8. A method, according to claim 7, characterized by the fact of comprising two or more subsequent positive bias pulses (30), wherein between each two adjacent positive bias pulses an intermediate negative bias pulse (45) is provided, having a duration time similar to that of said positive bias pulses (30).

9. A method, according to claim 8, characterized by the fact that the positive bias pulses (30), the intermediate negative bias pulses (45) and the negative bias pulses (40) are interleaved with a null time interval between them.

10. A method, according to claim 7, characterized by the fact of comprising two or more subsequent positive bias pulses (30), wherein between each two positive bias pulses (30), a time interval (50) of null voltage is provided with enough time to allow the secondary electrons (SE), ejected from the electrode (10) during the immediately preceding positive bias pulse, (30) to reach the ionized mass without significant deceleration.

11. A method, according to claim 1, characterized by the fact that the electrodes (10) are contained in a reaction chamber (RC), containing a partially ionized mass defined in a gaseous or liquid medium.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0025] From now on, the invention will be described with references to the accompanying drawings, given by way of example, in which:

[0026] FIG. 1 is a graph illustrating a typical voltage waveform of commercially known power supply sources for asymmetrical bipolar mode, for the generation of plasma applied to film deposition by sputtering, processes of surface treatment and etching, among others;

[0027] FIG. 2 represents schematically and partially a plasma chamber, containing a negatively polarized electrode for attracting the ions contained in the ionized mass, according to a known technique;

[0028] FIG. 2A shows a voltage versus time plot for the negative bias polarization of the electrode, as shown in FIG. 2;

[0029] FIG. 2B is a schematic representation similar to that of FIG. 2, but illustrating ions impacting the negatively polarized electrode and causing the emission of secondary electrons and eventually spraying neutral atoms or ions from the surface of said electrode;

[0030] FIG. 3 represents schematically and partially a plasma forming chamber containing a positively polarized electrode for attracting the primary electrons contained in the ionized mass, according to the first step of the method in question;

[0031] FIG. 3A shows the voltage versus time graph for the positive bias of the electrode, as shown in FIG. 3;

[0032] FIG. 3B shows a schematic representation similar to that of FIG. 3, but illustrating primary electrons bombarding and impacting the positively polarized electrode and causing the emission of secondary electrons from the surface of said electrode;

[0033] FIG. 4 represents schematically and partially the plasma forming chamber of FIGS. 2 and 3 with the electrode negatively polarized to accelerate the secondary electrons and the other electrons not absorbed by the electrode towards the ionized mass;

[0034] FIG. 4A shows the voltage versus time graph in a new and subsequent negative bias of the electrode, as shown in FIG. 4;

[0035] FIG. 4B is a schematic representation similar to that of FIG. 4, but illustrating secondary electrons and the primary electrons not absorbed by the electrode, colliding with neutral atoms and producing derived ions and electrons, according to the second step of the method in question;

[0036] FIG. 5 shows a voltage versus time graph illustrating a sequence of alternating negative and positive bias steps of the electrode, with zero time interval between said steps;

[0037] FIG. 6 is a graph similar to that of FIG. 5, but illustrating the provision of two subsequent positive bias pulses and one negative bias pulse between them; and,

[0038] FIG. 7 is a graph similar to that of FIG. 6, but showing the provision of a zero power time interval between the two subsequent positive bias pulses.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0039] As already discussed previously, the present invention relates to a novel method for increasing the density of a plasma, generally, but not necessarily, formed within a normally grounded reaction chamber RC and in which a mass of ionized elements, with a region of luminescence and including neutral atoms NA of an ionizable fluid in the gas or liquid phase and primary electrons PE and positive ions PI formed in an initial phase or in a starting ionization phase of the ionizable fluid within the reaction chamber RC of a suitable reactor (not shown). It should be understood that the electrodes may or not be contained in a reaction chamber RC, and if contained, one of the electrodes may or may not be defined by the wall (not shown) of said reaction chamber RC.

[0040] In brief, the method consists of controlling the waveform of the voltage applied to the electrodes and, consequently, controlling the energy and the intensity of secondary electrons SE, that will be described further. The increase in the rate of emission of secondary electrons, formed in one step of the method, allows the control of the generation of positive ions PI in the plasma.

[0041] As schematically illustrated in FIGS. 3 and 4, the method in question aims at the densification of a plasma formed, for example, inside a reaction chamber RC containing an ionized mass defined by a typical region of luminescence and including neutral atoms NA contained in an ionizable fluid and also primary electrons PE and positive ions PI.

[0042] Normally, the system is compounded by a grounded reaction chamber RC that houses at least one electrode 10 to be energized from an external electric power source 20 (as indicated in FIG. 4), designed to provide an energizing electrical voltage for the electrode 10 with pulses of high power positive polarization, and negative polarization pulses of low or also high power, depending on the method used.

[0043] The method of the invention comprises the steps of: biasing the electrode 10 with an external power source 20 by means of a positive bias pulse 30 with high power and during a first time duration (FIG. 3), positively polarizing the electrode 10 and generating a first electric field E1 that accelerates the primary electrons PE contained in the ionized mass in a direction opposite to the first electric field E1 and towards the surface of the electrode 10.

[0044] Each primary electron PE accelerated by the positive bias of the electrode 10 and impacting the electrode 10 may eject secondary electrons SE from the surface of the electrode 10, as shown in the sequence of FIGS. 3, 3A and 3B.

[0045] After the duration of the positive bias pulse has elapsed, the electrode 10 is biased with a negative bias pulse 40, shown in FIGS. 3A and 4A, for a second time duration, generating a second electric field E2 and accelerating, in the direction of the ionized mass, the secondary electrons SE and the primary electrons PE not absorbed by the electrode 10.

[0046] The secondary electrons SE and other electrons located in the vicinity of the electrode 10, and accelerated by the negative polarization of the electrode 10, collide with neutral atoms NA, producing positive ions PI and derived electrons DE, resulting from the ionization process.

[0047] The positive and negative bias steps with the respective positive 30 and negative 40 biasing pulses of the electrode 10 are repeated continuously, forming a periodic pulse train. Adjusting the intensity and the period of the respective pulses leads to the achievement of a desired degree of ionization of the ionized mass. The process can start with either a positive pulse or a negative pulse.

[0048] The voltage waveform used for plasma generation is an important aspect for the application of the method in question. The voltage waveform should contain positive bias pulses 30 and negative bias pulses 40, as shown in FIGS. 5, 6 and 7, respectively.

[0049] FIG. 2 highlights two phenomena that occur during the application of the negative bias pulses 40 on the electrode 10: initially, the positive ions PI from the ionization of the neutral atoms NA of the ionizable fluid are accelerated by the second electric field E2 toward the negative polarized electrode 10. After a period of time, these positive ions PI reach the surface of the electrode 10 and eject secondary electrons SE from that surface. Other phenomena may also occur, such as, for example, the sputtering of neutral atoms NA from the surface of the electrode 10. The emission efficiency of secondary electrons SE by the bombardment of positive ions PI is low (γ=(secondary electrons SE)/incident positive ions PI), being generally much smaller than the unit, that is, the emission of a secondary electron SE occurs (statistically) after the incidence of several positive ions PI. In general, the value of γ varies from 0.01 up to 0.5, depending on the energy and mass of the ions, the nature of the plasma's neutral atoms NA, the physicochemical properties of the electrode 10, in addition to its surface temperature.

[0050] FIG. 4A shows a voltage versus time plot for the negative bias of the electrode 10, as shown in FIGS. 4 and 4B.

[0051] FIG. 3B shows the time interval subsequent to that of FIG. 3 in which the electrode 10 is biased by a short high power positive pulse 30. During this positive bias pulse 30, an electric field E1 is generated, with direction pointing out of the electrode 10. This positive bias pulse 30 has a short duration, since the objective, at this moment, is to bombard the electrode 10 with primary electrons PE and any other electrons that are under the influence of the electric field generated. As electrons have very small mass (compared to the mass of the ions), this time is sufficient to accelerate the primary electrons PE and any other electrons in the direction of the electrode 10 (in the opposite direction of the electric field E1).

[0052] At the initial instants of this positive polarization pulse 30, the primary electrons PE and any other electrons near the electrode 10 reach the surface of the electrode 10, with sufficient energy for ejecting secondary electrons SE (see FIGS. 3, 3A and 3B). Secondary electron emission SE, due to the electronic bombardment of the electrode 10, is very efficient, and may have a d yield greater than 1, as can be seen in Table 1. This yield is defined as: d=(emitted secondary electrons SE)/(incident electrons).

[0053] During this short, high power, positive bias pulse 30 the secondary electrons SE initially move away from the surface of the electrode 10 and then tend to return to the surface, as shown by the curved arrows in FIG. 3B. This deceleration occurs due to the performance of the first electric field E1 on these secondary electrons SE. In a short period of time, typically in the range of nanoseconds up to microseconds, this process (emission and re-absorption of secondary electrons SE) is established on the electrode 10, with secondary electrons SE being emitted and reabsorbed by the surface of the electrode 10. These reabsorbed secondary electrons may also produce more secondary electrons from the electrode surface, if they have enough energy to do so. Also, within this interval of time, there is the possibility of some secondary electrons SE reaching the luminescent region of the plasma.

[0054] When the polarization is reversed by applying a negative bias pulse 40 to the electrode 10, as shown in FIGS. 4, 4A and 4B, the second electric field E2 reverses in direction with respect to the first electric field E1 and the electrons (both the secondary electrons SE and the primary electrons PE that are in the vicinity of the electrode 10) are accelerated away from the surface of the electrode 10. In the first moments of time, in this new polarization the electrons emitted from the electrode 10, and other electrons in the vicinity, are accelerated in the opposite direction to the second electric field E2, towards the plasma's luminescent region. During the application of the negative pulse, the positive ions PI (already existing and/or produced in the ionized mass) near the electrode are accelerated in the same direction of the second electric field E2, that is, from the plasma's luminescent region to the electrode 10. Similarly, but in the opposite direction, the eventual negative ions NI formed in the process are accelerated by the second electric field E2, from the electrode 10 to the plasma's luminescent region. Neutral atoms NA of the ionizable fluid (e.g., gas) do not undergo any force due to the electric field, since they are electrically neutral.

[0055] The secondary electron SE population, (amplified by electronic bombardment during each immediately preceding positive bias pulse 30) and other electrons located in the vicinity of the electrode 10, reach the luminescent region of the plasma with high speed (due to the acceleration carried out by the second electric field E2) and collide with the neutral atoms NA of the ionizable mass. In this event, the accelerated and colliding electrons (secondary electron SE and eventual primary electron PE) can remove derived electrons DE from these neutral atoms NA and generate new positive ions PI, as shown in FIG. 4B and, thus, significantly increasing the population of positive ions PI in the plasma.

[0056] Another effect caused by the acceleration of the secondary electrons SE, which reach the luminescent region of the plasma, is the heating (increase of kinetic energy) of the primary electrons PE already present in the plasma. This heating is by “collision” (coulombian electrical interaction) between these electrons. A portion of these electrons can acquire energy greater than the ionization potential of the neutral atoms NA and thus increase the production of positive ions PI in the plasma in subsequent collisions with neutral atoms.

[0057] Since the number of secondary electrons SE is amplified due to the positive bias pulse 30 applied in the previous step, the probability of generating positive ions PI in the plasma becomes very large. In this way, the density of positive ions PI in the plasma increases significantly, being able to reach values up to tens of times greater than the ionic density obtained in plasmas generated by traditional sources of continuous DC or pulsed DC voltages, with similar waveform, for example, to that shown in the FIG. 1 graph.

[0058] As previously mentioned, the voltage waveform should contain positive bias pulses 30 and negative bias pulses 40, as shown in FIGS. 5, 6 and 7, respectively and defined below: [0059] 1—Positive bias pulses 30, short and with high power (with voltage on the order of hundreds or thousands of volts, of short duration, on the order of nanoseconds up to microseconds), periodic, interspersed with negative bias pulses 40 (with a voltage ranging from tens up to hundreds of volts, or thousands of volts), or with time intervals with zero voltage, and with complementary duration to the previous period (see FIG. 5). In this form of operation, during the negative bias pulses 40, the positive ions PI are accelerated towards the electrode 10, whereas the electrons (both secondary electrons and other electrons located near the electrode 10) are accelerated in the opposite direction, promoting ionization of neutral atoms NA in the luminescent region, that is, producing positive ions in the plasma; [0060] 2—A sequence of at least two short positive bias pulses Vp 30 between which a reverse negative bias pulse 45, having a voltage intensity denoted by Vr (see FIG. 6), is applied. Said reverse negative bias short pulse 45 accelerates the secondary electrons SE and other electrons in the vicinity of the electrode 10 into the luminescent region and, by consequence, further increases the rate of plasma ionization; and [0061] 3—A sequence of at least two short, high power, positive bias pulses 30 separated by a zero voltage time interval 50, followed by a negative bias pulse 40. In that configuration, a short time interval between the high power positive bias pulses 30, allows the secondary electrons SE, generated during the first positive bias pulse 30, to reach the luminescent region of the plasma without being decelerated by the electric field E1. This occurs because the electric field is zero in the interval between the two pulses of positive polarization 30.

[0062] In this way, it is possible to control the generation of positive ions PI (by electronic impact) allowing to amplify the density of the plasma by a factor up to tens of times.

[0063] Other combinations with variances of the intensity and configuration of the short positive and long negative polarization pulses can be constructed, where it is possible to provide a plasma with desired characteristics (such as temperature, ionization rate, chemical reactivity, volume, etc.) to a given application.

[0064] Although FIGS. 1, 2A, 3A, 4A, 5, 6 and 7 illustrate (ideally) the positive and negative polarization pulses, by means of schematic graphic representations with rectangular waveforms, with rectilinear boundaries, it should be understood that disturbances and non-linearities may occur due to the parasitic inductances and capacitances of the cables and the electric supply circuit, which may generate distortions in these pulses.

[0065] Thus, the plasma generated has the following characteristics: it can have high intensity (large quantity of ions and electrons), being possible to achieve a substantial increase in the density of ions in the plasma, of up to 1000% in relation to continuous DC or pulsed DC plasmas; high stability, with suppression of voltaic arcs; it can be produced over a wide working gas pressure range, from 10-3 Torr up to ambient pressure (760 Torr); the volume of the plasma's luminescent region can be controlled by the adjustment of the positive short pulses (since the volume can be restricted by the contour of the electrode or fill up the entire volume of the discharge chamber), and may be produced in a liquid medium; it can be maintained with negative pulses in a wide range of electrical voltage values (from tens of volts up to thousands of volts). This allows an extensive power control, controlling the energy and ion density, allowing better plasma fit for different technological applications; and has flexibility to adjust parameters of the plasma, broadening its spectrum of technological applications.

[0066] The method in question is based on the different physical properties of electrons and ions: these particles have the same value of electric charge, but with opposite signals (charge=1.6×10.sup.−19 C; positive for the ion and negative for the electron). However, the mass of the electron is much smaller than the mass of the ion, for example, the ratio between the mass of an electron and the mass of an argon ion is approximately (m.sub.e/m.sub.Ar)=9.11×10.sup.−31 kg/6.65×10.sup.−26 Kg=1.37×10.sup.−5. Because of their tiny inertia, the electrons undergo much greater acceleration than the ions when subjected to an electric field (electric potential difference ΔV), since the acceleration can be calculated by the equation:

a=1/m(e.Math.ΔV)/x

[0067] Where, “a” is the acceleration, “e” is the elementary electric charge (charge of an electron), ΔV is the electric potential drop in the electrode sheath (sheath is the space between the surface of the electrode and the plasma's luminescent region), and “x” is the thickness of the sheath (note: this equation is valid only for the ideal case, when there are no collisions between the particles).

[0068] Thus, the short-duration positive bias pulse (for example, half microsecond) has sufficient time interval to promote electronic bombardment (primary electrons PE from the luminescent region and the sheath) over the electrode 10 and, consequently, significantly increase the emission rate of secondary electrons SE. In this time interval, the velocity of the positive ions PI is less affected, so that the plasma ion velocity distribution does not undergo significant changes. Therefore, in this time interval, a large production of secondary electrons SE occurs without significantly affecting the ion speed distribution. Another factor that makes the emission of secondary electrons SE more efficient when bombardment is done with other electrons is that the transfer of linear momentum is maximized when the collision occurs between bodies with equal masses. Therefore, the electron-electron collision is much more efficient (than the ion-electron collision) for momentum transfer and hence for the ejection of electrons from the surface.

[0069] After the emission of the secondary electrons, they are accelerated by the field E2 (due to inversion of polarity), pushed to the luminescent region with high energy. As a consequence, denser plasma with a higher density of positive ions PI and derived electrons is obtained. The degree of ionization, temperature and volume of the plasma (luminescence region volume) can be easily controlled by the intensity, duration and sequence of positive bias pulses 30. In addition, the positive bias pulses 30 interspersed with negative bias pulses act very efficiently in suppressing voltaic arcs between the electrodes.