throbber
Plasma Physics Reports, Vol. 21. No. 5, 1995, pp. 400 - 409. Translatedfrom Fizilm Plazmy, Vol. 21, No. 5. 1995, pp. 422 . 433.
`Original Russian Text Copyright © 1995 by Mozgrin, Ferirov, Khodachenka
`
`LOW-TEMPERATURE PLASMA
`
`High-Current Low-Pressure Quasi-Stationary Discharge
`in a Magnetic Field: Experimental Research
`
`D. V. Mozgrin, I. K. Fetisov, and G. V. Khodachenko
`Moscow Engineering Physics Institute, Kashirskoe sh. 31 , Moscow, 115409 Russia
`Received October 22, 1993; in final form, July 12, 1994
`
`Abstract -— The possibility of realizing several types of high—power quasi—stationary low-pressure discharge in
`a magnetic field was shown. Two noncontracted discharge regimes in crossed E and H fields were studied.
`These discharges had much higher cathode current densities than those of other known discharge types. Their
`parameter ranges were determined, and their operating regimes were investigated. The voltage for the high-volt-
`age form discharge ranged from 450 to 1000 V; the discharge current amounted to 250 A, and cathode current
`density reached 25 A/cmz. A low-voltage discharge form was first observed: voltage ranged from 75 to 120 V;
`discharge current amounted to 1800 A, and cathode current density reached 75 A/cmz; lifetime was about 1.5 ms.
`The ion density was 1.5 X 1015 cm’3 in argon discharges and amounted to S x 10‘4 cm'3 in He~H2-niixture dis—
`charges, while the electron temperature was about 3 — 8 eV. The properties of both discharge types are expected
`to open up new fields of application in technology.
`
`1. INTRODUCTION
`
`Low—pressure discharges in a magnetic field attract
`much attention due to their wide use in technological
`magnetron devices, closed—electron—drift plasma accel—
`erators, and, as plasma emitters in electron or ion injec-
`tors.
`
`Stationary regimes of the discharges in planar mag-
`netrons of technological use are characterized by p a
`10‘4 - 5 x 10“2 torr operating pressure and 300 — 1000 G
`magnetic field at the cathode surface [1, 2]. Their AV
`characteristic is described bythe formula 1,, = kUZ,
`where 1,, is the discharge current, and Ud is the dis-
`charge voltage. The quantities k and n depend on the
`device geometry, working gas type and pressure, and
`magnetic field strength. The condition n > 1 holds, if the
`cathode current density jc does not exceed 0.03 A/cmz.
`In this case the discharge voltage amounts to 400 — 600 V,
`the plasma density n; ranges from 108 to 1011 cm‘3, and
`electron temperature Te reaches 20 eV. If the current
`density is higher, a transition of the discharge into the
`arc regime is observed.
`Because of the need for greater discharge power and
`plasma density, pulse or quasi—stationary regimes
`appear to be of interest. Some experiments on magneo
`tron systems of various geometry showed that dis-
`charge regimes which do not transit to arcs can be
`obtained even at high currents. For example, a super—
`dense glow discharge, realized in a device of “reversed—
`magnetron” type of coaxial geometry, exhibited the fol—
`lowing parameters: about 70 A discharge current Id,
`400 V discharge voltage Ud, 60 us pulse duration, and
`1012 cm”3 plasma density n,- [3]. A pulse duration decrease
`down to 100 us, which was performed in a planar magne-
`tron discharge in Ar, N2, or H2 at 10'3 - 5 X 10“2 torr
`
`pressures and 1.0 — 3.0 kG magnetic field strength, per-
`mitted a 1000 A current value to be obtained in the non—
`contracted regime, at 300 — 500 V discharge voltage,
`with about 50 J of total energy deposition [4]. In both
`examples, the discharge current—voltage characteristic
`increased and then became constant with the increase
`in the discharge currents. A further increase in the dis—
`charge currents caused the discharges to transit to the
`arc regimes, with voltage not higher than 50 V under
`those conditions.
`
`Our previous experiments demonstrated the possi-
`bility of realizing several stable discharge regimes in
`devices with closed electron drift [5 — 7]. Among these
`regimes which differed from the arcs, was an interme—
`diate low—voltage regime (Ud z 100 V, 1d _<_ 1.5 kA)
`of longer than l-ms pulse duration (hereafter called
`a “high—current diffuse regime”).
`The main purpose of this work was to study experi-
`mentally a high-power noncontracted quasi—stationary
`discharge in crossed fields of various geometry and to
`determine their parameter ranges. We investigated the dis—
`charge regimes in various gas mixtures at 10‘3 — 10 torr,
`BO S 1000 G, and pulse durations exceeding 1 ms.
`Such regimes can be useful in generating large-volume
`dense plasmas and intense flows of charged particles.
`Furthermore, we consider qualitatively the mechanism
`of low—voltage high—current discharge formation.
`
`2. EXPERIMENT
`
`To study the high—current forms of the discharge,
`we used two types of devices: a planar magnetron and
`a system with specifically shaped hollow electrodes.
`The planar magnetron (Fig. 1)
`involved a plane
`cathode 120 mm in diameter and a ring—shaped anode
`
`1063-780‘ ’95/2105-0400$l0.00 © 1995 MARK Hayxa y’lnterpericdica Publishing
`
`INTEL 1203
`
`TSMC-1403
`
`TSMC v. Zond, Inc.
`
`Page 1 of 10
`
`TSMC-1403
`TSMC v. Zond, Inc.
`Page 1 of 10
`
`

`

`HIGH—CURRENT LOW—PRESSURE QUASI-STATIONARY DISCHARGE
`
`401
`
`160 mm in diameter, The electrodes were immersed in
`
`a magnetic field of annular permanent magnets. Mag—
`netic circuits were used to vary the size of the region of
`the large magnetic—field radial component and mag-
`netic field inhomogeneity degree. To control the mag—
`netic field strength at the cathode surface, We displaced
`the magnetic system along the axis 2 (Fig. l) and used
`two types of permanent magnets made of SmC05 and
`NdFeB. The discharge had an annular shape and was
`adjacent to the cathode. The maximum of the magnetic
`field radial component B, at the cathode surface was
`800 G for the SmCo5 magnet or 1200 G for the NdFeB
`magnet. The cathodes we used were made of Cu, Mo,
`Ti, Al, or stainless steel. The cathode was placed on a
`cooled surface. The anodes were made of aluminum or
`stainless steel.
`
`The system with shaped electrodes involved two
`hollow axisymmetrical electrodes 120 mm in diameter,
`separated by about 10 mm, and immersed in a cusp—
`shaped magnetic field produced by oppositely directed
`multilayer coils. The discharge volume bounded by the
`electrodes was about 103 cm3. The ratio of the maximal
`
`magnetic field at the axis of symmetry Bmax(z, O) to the
`maximal magnetic field at
`the plane of symmetry
`Bmax (0, r) was about 3. The values ofBmax were controlled
`by coil current variation to range from O to 1000 G.
`The electrode shapes followed the magnetic line pro—
`file, which enabled the electric field to be perpendicular
`to the magnetic field along the cathode surface. Such
`a field configuration allowed us to combine a high-cur—
`rent magnetron discharge with a hollow—cathode dis-
`charge.
`
`The gas from the discharge volume was pumped out;
`minimal residual gas pressure was about 8 X 10'6 torr.
`It was possible to form the high--current quasista—
`tionary regime by applying a square voltage pulse to the
`discharge gap which was filled up with either neutral or
`pre—ionized gas. Estimates were made to determine
`both the quasi— stationary plasma density and its build-
`ing-up time [5 7]. The necessary pre— ionized plasma
`density n- turned out to be 107— 109 cm‘3 for argon
`In addition,
`the estimates determined the shape Dand
`parameters of the voltage pulse. The pre-ionization
`could be provided by RF discharge, anomalous glow or
`magnetron discharge, etc.
`
`Figure 2 presents a simplified scheme of the dis—
`charge supply system. The supply unit
`involved
`a pulsed discharge supply unit and a system for pre—
`ionization. The quasi— stationary discharge-s—upply unit
`consisted of a long line of W: 5.5 k] maximal energy
`content, a switch, and a matching unit. The pre- ioniza—
`tion system provided direct current up to 0. 3 A and volt—
`age up to 3 kV.
`The frequency parameters of the pulsed supply unit
`were chosenin accordance with theincrease in time of
`the quasi-stationary plasma density formation and the
`times of the ionization instability and ionization—over-
`heating instability development. Designing the unit, we
`
`PLASMA FHYSlCS REPORTS
`
`Vol. 21
`
`No. 5
`
`1995
`
`m
`
`
`Fig. 1. Discharge device configurations: (a) planar magne—
`tron;
`(b)
`shaped—electrode configuration.
`(1) Cathode;
`(2) anode; (3) magnetic system.
`
`took into account the dependences which had been
`obtained in [8] of ionization relaxation on pre—ionization
`parameters, pressure, and pulse voltage amplitude.
`In addition, we allowed for the fact that the development
`time for the ionization—overheating instability was about
`10‘3 — 3 X 10‘3 s in the pressure range up to 0.5 torr [9].
`Thus, the supply unit was made providing square volt-
`age and current pulses with raise times (leading edge)
`of 5 - 60 its and durations of as much as 1 .5 ms. Short-
`circuit current amplitude was up to 3 kA; no—load volt—
`age was as much as 2.4 kV.
`
`For pre—ionization, we used a stationary magnetron
`discharge; the discharge current ranged up to 300 mA.
`We measured the discharge current—voltage characteris-
`tics (CVC) in a 10‘3 — 10 torr pressure range and plasma
`parameters of the discharge at the symmetry center of
`the shaped—electrode system using a probe technique.
`We found out that only the regimes with magnetic field
`strength not
`lower than 400 G provided the initial
`plasma densityin the l09 10” cm 3range This initial
`density was sufficient for plasma densitybto grow when
`the square voltage pulse was applied to the gap. So we
`chose these regimes as pre—ionization regimes.
`
`TSMC-1403 I Page 2 of 10
`
`TSMC-1403 / Page 2 of 10
`
`

`

`402
`
`MOZGRIN et a1.
`
`
`High-voltage
`
`supply unit
`
`
`
`
`Stationary
`
`discharge
`
`supply unit
`
`
`Fig. 2. Discharge supply unit.
`
`3. QUASI—STATIONARY DISCHARGE REGIMES
`
`We studied the high—current discharge in wide
`ranges of discharge current (from 5 A to 1.8 kA) and
`operating pressure (from 10‘3 to 10 torr) using various
`
`(a)
`
`1
`
`2a 2b
`
`3
`
`(b)
`
`
`
`1
`
`2a 2b
`
`3
`
`Fig. 3. Oscillograms of (a) current and (b) voltage of the
`quasi-stationary discharge (50 us per div., 180 A per div.,
`180 V per div.).
`
`U, V
`500
`
`400 W
`
`300
`
`1
`
`200
`
`O
`
`
`l_t_LLLU11!J_LLLU11_J_|—LUJJJJ
`LJ.LLUJJJ.1
`0.1
`1
`10
`100
`1000
`I, A
`
`Fig. 4. Current—voltage characteristic of the quasi—stationary
`discharge with shaped electrodes in argon, p = 0.1 torr; B =
`0.4 kG.
`
`gases (Ar, N2, SF6, He, and H2) or their mixtures of vari-
`ous composition (argon percentage in Ar—N2 and Ar—SF6
`composition ranged from 10 to 90%; He : H2 = 1 : 1).
`We investigated the planar-magnetron and cusped-mir—
`ror configurations varying the magnetic field strength.
`We obtained current—voltage characteristics of the dis—
`charge, time—integrated photographs of the discharge
`glow, and probe characteristics of the discharge plasma.
`We detected the particle flux from the plasma and mea-
`sured their intensities. As a result, we found out that
`a variety of regimes differing in discharge voltage, cur-
`rent range, and discharge space structure occurred.
`
`Figure 3 shows typical voltage and current oscillo-
`grams of the quasi—stationary discharge. Part I in the
`voltage oscillogram represents the voltage of the sta-
`tionary discharge (pm-ionization stage). Part 2a dis—
`plays the square voltage pulse application to the gap.
`At this stage, the plasma density grows and reaches its
`quasi-stationary value (part 2b); the discharge current
`also grows, and then both the discharge current and
`voltage attain their quasi—stationary values (part 3).
`The time it takes for the plasma density to reach its
`quasi-stationary value corresponds to the ionization
`relaxation time. For example, for argon, discharge when
`pus—ionization plasma density is about 109 — 10“ cm“3
`this time is about 50 us. Each point of the discharge
`characteristic represents a pair of voltage and current
`oscillograrns. We detected inhomogeneity of the dis—
`charge plasma or cathode spots visually, using filters,
`or by photographing the discharge.
`
`The current—voltage characteristic of the low—pres-
`sure quasi—stationary discharge in a magnetic field had
`four different parts corresponding to stable forms of the
`discharge. Figure 4 shows a typical CVC of the dis-
`charge in argon at 10'| torr pressure and 0.4 kG mag-
`netic field. One can differentiate two parts: part 1 cor-
`responds to the magnetron discharge with current up to
`0.2 A and voltage range from 260 to 280 V; part 4 cor-
`responds to the high-current low-voltage arc discharge
`of current greater than 1 kA and 10 - 30 V voltage with
`a cathode spot. In addition, we found out experimen-
`tally that two other stable forms of quasi-stationary dis—
`charge conld exist, Both the plasma and cathode layer
`had a diffuse character at cathode current density much
`higher than that of typical magnetron discharge. If the
`discharge current ranged from 0.2 to 15 A, a high—cur-
`rent magnetron discharge having initial discharge char-
`acteristics was observed (part 2 of the oscillogram).
`
`PLASMA PHYSICS REPORTS
`
`Vol. 21
`
`No, 5
`
`1995
`
`TSMC-1403 I Page 3 of 10
`
`TSMC-1403 / Page 3 of 10
`
`

`

`HIGH—CURRENT LOW—PRESSURE QUASI—STATIONARY DISCHARGE
`
`403
`
`In this case the discharge voltage was rather high,
`approximately 350 — 500 V. If the current was increased
`and ranged from 15 to 1000 A, a diffuse regime of high-
`current discharge was observed (part 3); its CVC was
`a straight line parallel to the current axis. The discharge
`voltage was about 90 V over
`the current
`range.
`The cathode current density was about 50 A/cmz.
`It should be noted that the boundaries of regimes
`could vary depending on the discharge conditions, e.g.,
`on pressure, magnetic field strength, etc. Then, we
`studied regimes 2 and 3 separately to determine the
`boundary parameters of their occurrence, such as cur-
`rent, voltage, pressure, and magnetic field.
`We studied the regimes both in the planar magne-
`tron and shaped—electrode system geometries and
`found out that both regimes could occur regardless of
`the type or particular parameters of the discharge con—
`figuration.
`Figure 5a exhibits representative CVC of the high-
`current magnetron discharge. They were measured in the
`discharge in Ar and N2, as well as in or Ar—Nz (10 — 90%
`of argon) or He : H2 = 1
`: 1 mixtures at 10‘3 — 10 torr
`pressure range and 0.4 — 1.0 kG magnetic field. The
`cathodes we used were made of Cu, Ti, Al, M0, or
`stainless steel. To reduce the effect of cathode surface
`quality on the discharge parameters,
`the electrodes
`were preconditioned by multiple discharges or cleaned
`by glow discharge in argon. The dependence Ud(Id)
`remained qualitatively the same for all values of were
`the pressure p, transverse magnetic field BL, sort of the
`gas, cathode material, electrode configuration and dis-
`charge size. The discharge voltage increased monotoni-
`cally with current up to a maximum U?“ = 500 — 1100 V
`depending on the magnetic field strength, sort of the
`gas, and cathode material. Then the discharge trans—
`ferred to regime 3 or to the arc regime. If the voltage
`pulse duration ’5 was less than 20 ms, the current of
`transition amounted to 250 A, which corresponded to
`25 A/cm2 cathode current density j. A decrease in mag-
`netic field strength resulted in an increase in the dis-
`charge voltage UZ‘“(BL) up to some value U31 depend-
`ing only on the cathode material and sort of the gas.
`A further decrease in B, caused the discharge to transit
`to a high-voltage regime which was characterized by
`a steep CVC and low discharge current (about 1 A).
`As the decreasing magnetic field approached the
`value of the discharge transit
`to the high-voltage
`regime, the discharge voltage increased smoothly, and
`the discharge current decreased.
`We measured the CVC of the high—current magne-
`tron discharge for two different discharge diameters.
`The CVC turned out to be independent of the diameter
`in the max B, region. It should be noted that, being
`transferred to the high-current regime, the discharge
`expands over a considerably larger area of the cathode
`surface than it occupied in the stationary pre-ionization
`regime. In the case of the planar magnetron, the dis—
`
`PLASMA PHYSIC.) REPORTS
`
`Vol. 21
`
`No. 5
`
`1995
`
`
`
`200
`
`1
`)
`(
`Z 2
`o 3 M
`. 4 W
`100 m.
`
`b
`
`
`l
`|
`_l
`10
`100
`1000
`I, A
`
`0
`
`Fig. 5. (a) High-current magnetron discharge: (1) planar
`magnetron, Cu, p = 5 X 10’3 torr, Ar; (2) planar magnetron,
`Ti, [1 = 5 x 10‘3 torr, Ar : N2 = 4 : 1; (3) planar magnetron,
`Ti, p = 10‘2 torr, N2; (4 and 5) shapedelectrode system, Cu,
`p = 5 x 10‘2torr, He 2 H2=1 : 1, and Cu,p =10‘1torr,Ar.
`(b) High-current diffuse regime: (1 and 2) shaped—electrode
`system, Cu,p=1torr, He: H2: 1
`: l and Cu,p =10‘1torr,
`He : H2: 1 : l; (3) planar magnetron, Cu,p = 10“1 torr, Ar;
`(4) planar magnetron, Cu, p = 10'1 torr,Ar: SF6 =4 : 1.
`
`charge occupied either the ring area beyond the circular
`region of max B, or the disk area bounded by the circle
`of max 3,; the area depended on the magnetic field con—
`figuration. Because the discharge current was the same
`in both cases, the current densities differed consider—
`ably, but the CVCs were similar. The current density
`values characteristic of these regimes for the argon dis-
`charges werej = 4 A/cm2 (Ud = 540 V, 1,, = 225 A) and
`j = 25 A/cm2 (Ud = 500 V, 1,, = 218 A).
`
`The roughness of the cathode surface was not
`important for the occurrence of regime 2, though the
`probability of discharge transit to the arc discharge was
`greater for the cathodes with rougher surfaces.
`
`A feature of the shaped—electrode discharges in the
`He—H2 mixture was a second form of high-current
`magnetron regime at a 400 — 650 V discharge voltage
`that was independent of discharge current until trans-
`ferring to regime 3.
`
`Regime 2 was characterized by an intense cathode
`sputtering due to both high energy and density of ion
`flow. To study the Sputtering, we used a probecollector
`placed 120 mm from the cathode. The pulsed deposition
`rate of cathode material (copper was used) turned out to
`be about 80 urn/min in the argon discharge, 1,, = 65 A,
`Ud = 900 V. The current pulse duration was 25 ms, and
`
`TSMC-1403 I Page 4 of 10
`
`TSMC-1403 / Page 4 of 10
`
`

`

`404
`
`MOZGRIN et a].
`
`the repetition frequency was 10 Hz, which corre-
`sponded to =20 urn/min averaged deposition rate.
`We used a scanning electron microscope REM-101
`(Russian trade mark) to measure the thickness of
`deposited layers.
`
`We measured the plasma density n,- in the region
`near the collector by applying to the collector a pulse of
`biased voltage with respect to the anode. The density
`turned out to be about 3 x 1012 cm“3 in the regime of
`Id =60A and Ud= 900V.
`
`Figure 5b presents typical CVCs of high—current dif—
`fuse discharge measured at various pressures, gases,
`cathode materials, magnetic fields, and pre-ionization
`parameters. Analyzing the CVCs, We found out that the
`discharge voltage weakly depended on the magnetic
`field geometry and its strength, and on the cathode
`material; the constant voltage turned out to range from
`70 to 140 V as the current ranged from 5 to 1800 A.
`The voltage was slightly (within 50 V) changed from
`gas to gas. Transferring to regime 3, the discharge occu—
`pied a significantly larger cathode surface than in the
`stationary regime.
`
`The parameters of the shaped-electrode discharge
`transit to regime 3, as well as the condition of its transit
`to are regime 4, could be well determined for every
`given set of the discharge parameters. The point of the
`planar-magnetron discharge transit to the arc regime
`was determined by discharge voltage and structure
`changes; the structure changes Were recorded by opti-
`cal diagnostics. To study the structure of the discharge
`in regime 3, we photographed it using neutral light fil—
`ters of various attenuation factors. The filters and expo-
`sure times were chosen so that the pre—ionization dis-
`charge glow was not recorded. One can see from the
`photographs presented in Fig. 6 that the discharge was
`spatially uniform even at about 1 kA discharge cur—
`rents. If the current was raised above 1.8 kA or the
`pulse duration was increased to 2 — 10 ms, an instability
`development and discharge contraction was observed.
`The planar-magnetron discharge transfer to regime 3
`resulted in a smearing of the annular structure of the
`pre—ionization discharge: the discharge plasma and cur—
`rent area were seen to expand and cover the whole cath—
`ode surface (Fig. 6). If the discharge current or pulse
`duration were increased, the instability development
`accompanied by the plasma column contraction and the
`occurrence of one of several cathode spots were also
`observed in the planar magnetron.
`
`Chemical analysis of the collector surface layer was
`done;
`the cathode material was not detected there.
`Hence,
`there was no cathode sputtering in these
`regimes.
`
`We elaborated on a pulsed probe technique specially
`designed to measure the plasma parameters in regime 3.
`The technique provided probe characteristics to be
`measured in =10 its time intervals and allowed the
`probe current to amount to 50 A [10].
`
`We measured the parameters of pulsed high—current
`quasi-stationary discharge in a cusp magnetic-field
`configuration with B ranging from zero to l kG in var-
`ious gases. The pressure ranged from 10‘1 to 1 torr; the
`discharge current ranged up to 1500 A. The pulse volt—
`age applied to the probe was 100 — 500 us delayed with
`respect to the discharge current pulse, i.e., TE and n,-
`were measured after the establishment of the quasi-sta—
`tionary regime of the high-current discharge.
`
`The plasma parameters were determined from the
`probe measurements. Ion density measured at the sys-
`tem center in regime 3 in argon increased almost lin-
`early with the discharge current at various pressures
`and magnetic field strengths. The density ranged from
`(2 — 2.5) X 10)4 cm“3 at 360 - 540 A current up to
`(1 - 1.5) X 1015 cm“3 at 1100 - 1400A current. The max—
`imal plasma density of high-current diffuse discharge
`in argon was measured to be In = 1.5 X 1015 cm‘3, while
`the electron temperature Te was 4 - 6 eV, the discharge
`current was 1100 A, magnetic field strength B was
`0.8 kG, and the pressure p was about 0.2 torr. The ion
`saturation current of the probe jSm was about 11 A/cmz.
`Ion density increased with pressure;
`the density
`increase was accompanied by a decrease in the electron
`temperature.
`
`The plasma density in He—H2 discharge also
`increased with the discharge current. However, the max-
`imum of ion density was n; z 2.4 X 1014 cm“3 at the con‘
`ditions similar to those mentioned above: p = 1.5 torr,
`B = 0.8 kG, Id 21100 A.
`
`4. DISCUSSION
`
`We obtained a generalized CVC of the quasi—station-
`ary low-pressure discharge in a magnetic field (Fig. 7)
`based on a variety of measured AV discharge character—
`istics under various conditions. Parts 1 and 4 correspond
`to stationary magnetron and are discharges, respectively.
`They were inherent in the discharge throughout the pres-
`sure and magnetic field ranges. These two regimes were
`comprehensively described in [1, 11].
`Part 2 pertains to the high-current magnetron dis—
`charge regime occurring in the 0.2 — 250 A current
`range. The voltage increased with current up to some
`critical value of current and then became constant.
`The discharge voltage was rather high — up to 1.2 kV.
`The discharge had a greater probability of being real—
`ized if the pressure ranged from 2 X 10‘3 to 10‘1 torr.
`We suggested that this discharge was structurally
`very close to the high—current discharge described in [4].
`The reasons are the following: both the pressure and
`magnetic field ranges were almost the same, the dis—
`charge did not exhibit contraction, and their CVCS
`were very similar. However, the discharge we dealt
`with had a higher discharge voltage (500 - 1200 V) than
`the 300 — 500 V discharge described in [4}. Hence, one
`could expect
`the cathode sputtering to have more
`importance.
`
`PLASMA PHYSICS REPORTS
`
`Vol. 21 No.5
`
`1995
`
`TSMC-1403 I Page 5 of 10
`
`TSMC-1403 / Page 5 of 10
`
`

`

`HIGH—CURRENT LOW—PRESSURE QUASI—STATIONARY DISCHARGE
`
`405
`
`
`
`Fig. 6. High-current quasistationary discharge regimes. (a) planar mlragnetron (I) high-current magnetron regime (p: 5 x 103 torr,
`Ar, Id: 70A Ht): 900 V); (2) high—current diffuse regime (p=10‘l torr Ar Id: 700A Ud-— 80 V); (3) are regime (p-_ 10‘110rr,
`AI, Id: 1000A Ud-45 V) (b) Shapcd--electrode system: (1) high—current diffuse regime (p=10I torr Ar, [d- 1000 A U01-- 90 V);
`(2) contracted arc regime (p-_ 10’ torr, Ar, 11!—— 1500 A, Ud- 50 V).
`
`PLASMA PHYSICS REPORTS
`
`V01. 21
`
`N0. 5
`
`i995
`
`TSMC-1403 I Page 6 of 10
`
`TSMC-1403 / Page 6 of 10
`
`

`

`MOZGRIN et al.
`
`406
`
`U, v
`500 - 1000
`
`70 - 170 -------------- L-—-———3-—————.
`
`0
`
`15 — 225
`
`1000 — 1800 1,?
`
`Fig. 7. Generalized ampere-voltaic characteristic CVC of
`quasi-stationary discharge.
`
`We estimated the Steady—state average density of the
`cathode material atoms n5 in the discharge plasma; the
`estimates were based on measured current density and
`initial working gas density ”3‘ Using a one—dimensional
`continuity equation, it is possible to derive the follow—
`ing equation for nc:
`
`t
`
`
`an,
`at +I’l':
`
`“ S
`-—
`
`
`
`V3
`15
`v;
`B
`g-C gng<vc>+z Cichnc(Vc)i
`
`(1)
`
`where Sg-C and SPC are sputtering and self-sputtering
`factors of the cathode, respectively; V; and V: are ther-
`mal velocities of working gas atoms and cathode mate-
`rial atoms, respectively;
`(V0)
`is the angle-averaged
`emitted atom velocity—component perpendicular to the
`cathode surface; 68 and [36 are ionization degrees of gas
`atoms and cathode material atoms, respectively.
`The steady—state solution of (1) is as follows:
`
`
`V;
`1
`ZSg—Efig"g<vc>
`”c ? —“‘1—“‘T
`1— 152—ch <Vc>
`
`As an example, we considered the copper-cathode
`argon discharge at 10'2 torr, Id = 65 A, and U,1 = 900 V
`
`and estimated the copper fraction in its plasma. The
`fraction turned out to be about 30% and increased with
`ionization degree.
`
`Table 1 presents the parameter ranges correspond—
`ing to regime 2. The presented parameters are limiting
`values that could be independently realized.
`
`If the discharge current ranged from 10 to 1800 A,
`high-current diffuse regime 3 occurred. The voltage
`ranged from 70 to 140 V depending on the working
`gas sort. Regime 3 could be primarily realized in the
`10‘2 - 5 torr pressure range no matter What the dis-
`charge electrode configuratiou, sort of working gas, or
`cathode material. Moreover, pre-ionization was not
`necessary; however, in this case, the probability of dis-
`charge transferring to are mode increased. The cathode
`current density of the high—current diffuse discharge
`amounted to 75 A/cmz.
`
`Table 2 presents the parameter ranges correspond-
`ing to regime 3 for both electrode configurations: vari-
`ous cathode materials, and various pressures and sorts
`of gases.
`
`It should be emphasized that the limiting values of
`regime 2 and regime 3 parameters were highly depen-
`dent on current pulse duration. The transit current val-
`ues increased with a decrease in the pulse duration.
`
`The process of the high~current magnetron dis-
`charge passing to a low—voltage regime seemed likely
`to resemble the anomalous glow discharge at moderate
`or high current densities [11]. In this case both the gas
`heating and electron density increase were important;
`these factors promoted both a partial equilibrium in the
`discharge plasma, and the transfer of the discharge to
`a regime of lOWer discharge voltage.
`The simplified quasi—stationary equation for gas
`temperature of regime 2 has the form [9]:
`
`(Tg— T0)MngvT = Pg,
`
`(2)
`
`where PE is the power consumed to heat the gas; Tg is
`the established gas temperature; T0 is the electrode tem—
`perature; Mg is the mass of the gas in the discharge vol-
`ume; Cp is the heat capacity at constant pressure; VT =
`
`Table 1. Existence conditions and regimes of high-current magnetron discharge: (1) shaped-electrode system; (II) planar magnetron
`
` Discharge de—
`
`vice type
`
`Hell-12
`50/50%
`Ar
`Ar
`
`
`
`
`Voltage, V
`Pressure, torr
`Current, A
`
`
`10-2 — 10-1
`260 — 990
`0.2 _ 15
`
`5 ><10-2 — 1
`320 — 950
`0.2 — 36
`ire/H2
`
`
`
`50/50%
`
`5 x10’2 ~1
`
`
`
`
`
`0.2 — 72
`540 - 990
`10-3 - 5 x 10—1
`
`
`
`0.2 — 250
`540 - 1100
`10—3 - 5 ><10‘l
`
`
`
`0.2— 180
`540-720
`10-3.5><10-l
`N2
`
`
`
`10*3 — 5 x 10—1
`540 - 900
`0.2 - 140
`Ar/Nz
`
`
`
`90 ~ 10% Ar
`
`
`400 - 650
`
`9 — 120
`
`
`
`PLASMA PHYSICS REPORTS
`
`V0121 No.5
`
`1995
`
`TSMC-1403 I Page 7 of 10
`
`TSMC-1403 / Page 7 of 10
`
`

`

`HIGH—CURRENT LOW—PRESSURE QUASI—STATIONARY DISCHARGE
`
`407
`
`Table 2. Existence conditions and regimes of high—current diffuse discharge: (1) shaped—electrode system; (H) planar magnetron
`
`
`
`
`lelfigrt‘ggge- Cathorqpllmate-
`Gas
`B, kG
`Pressure, tort
`Voltage, V
`Current, A
`
`
`
`
`
`Cu
`Ar
`10'2 - 5
`80 — 110
`15
`— 1500
`Cu
`Ar/SF6
`0.4 — 1.0
`2 x 10*2 A 3
`8O - 120
`7 — 1200
`
`
`30/20%
`ire/H2
`
`50/50%
`
`Ar
`
`1
`Ar
`1
`Ar/SF6
`1
`80/20%
`1
`51:6
`1
`Ar/SF6
`1
`80l20%
`
`3
`
`0.8 - 1.0
`
`5 ><10‘2 — 3
`
`0.3 - 0.7
`0.3 — 0.7
`0.3 — 0.7
`
`0.3 » 0.7
`0.3 ~ 0.7
`
`10-2 - 1
`5 x 10-2 — 1
`10-2 - 1
`
`10-2 — 5 ><10'1
`10-2 — 1
`
`
`
`
`
`
`
`
`
`
`70 - 120
`
`65 - 90
`65 — 90
`90 - 135
`
`90 -135
`90 — 135
`
`7 - 1800
`
`4 - 800
`4 ~ 1200
`18 — 1800
`
`4 ~ 1600
`18 ~ 1200
`
`
`
`Cu
`
`Cu
`Mo
`Mo
`
`Mo
`stainless
`steel
`
`1
`
`'
`
`
`
`n8 > 12?. The value of n:' can be determined experi-
`mentally; it depends on the excitation energy of the first
`excited level.
`
`If the cylindric layer is considered, the averaged rate
`of charged particle rate due to diffusion can be
`described as follows [9]:
`
`(
`
`9—1-2—
`at
`
`an
`
`DA
`2 DAAn = ~an 2: —X§n,
`
`(5)
`
`where VD is the diffusion frequency; A is a character—
`istic length of diffusion depending on the discharge
`geometry. In the presence of a magnetic field, A is
`dictated by the highest of the frequencies determin-
`ing the particle loss at
`the electrodes VD, =
`(T, + T1) (veu +vu.)
`rt
`——-2—————~—————— (—)
`mwe + (Veg + Vet") [11'an L
`
`In and at the lateral sur—
`
`2
`
`face of the discharge volume V DS =
`
`(Te+ T1) 2.4)2
`uiuVia
`R
`[12]. Here, R is the discharge radius; we is the electron
`cyclotron frequency; veg, vet, and V1,, are the frequencies
`of electron—atom, electron—ion, and ion—atom colli-
`sions, respectively; pm is the reduced mass. One can
`compare the frequencies of diffusion, ionization, and
`recombination in argon discharge and, then, fmd out of
`establishing a detailed balance the possibility in the dis-
`charge plasma.
`The estimates of the plasma density, in the case T =
`T, =T,- = 1.1 eV, show that, according to [9], the follow—
`irig relation is valid:
`
`8U
`
`062
`gi Zflm 3/2 (kDS/z
`_
`expt—kT'}.
`1—0’2 — gg(
`’12)
`
`ng+ni
`
`<6)
`
`where g,- = 6 (Ar) and gg = l are the statistical weights
`of ions and atoms, respectively; U,- is the argon ioniza-
`tion potential. The ionization degree or = 118 /(ng + m)
`ranges from (X = 1 (p = 0.01 torr) to 01 z 0.7 (p =1 torr),
`
`TSMC-1403 I Page 8 of 10
`
`x/A2T is the frequency characteristic of the heat evacu-
`ation; X = k/(Mnng) is the temperature conductivity;
`AT is a characteristic heat evacuation length; l is the gas
`thermal conductivity at the established temperature;
`M is atomic mass; and :1g is the gas density. The gas
`energy balance equation takes into account the dis-
`charge geometry, so the actual electrode profile was
`substituted for a plane one with area S and interelec-
`trode distance L corresponding to the operating area in
`the regimes involved. As far as the plane discharge
`layer of 2L thickness is concerned, the value of AT is
`about L /71:. The power consumed for gas heating can be
`evaluated from the energy balance equation:
`
`(3)
`
`I —
`
`‘1,
`ml
`
`Pg = IdUd—
`
`where IdUd is the total power released in the discharge
`volume; Id/ en is the power consumed for ionization;
`e is the electron charge; 1/1] is the Stoletov constant
`corresponding to the value of E/n characteristic of the
`cathode layer.
`The discharge conditions typical for the discharge tran—
`sit from regime 2 to regime 3 are as follows: Id 2 15 A;
`'d = 300 V; p = 1 torr; B = O; argon was used as a work—
`ing gas; L = 1 cm; and S = 60 cm2. Under these condi—
`tions, the gas temperature could increase to Tg = 1.1 eV.
`It follows from (2) that
`
`Id
`1 L [dud _ a V
`P
`
`Tg ~ M C v ‘ n2 S
`l
`,
`817T
`
`(4)
`
`i.e., that the gas temperature does not depend on gas
`density. The action of the magnetic field serves only to
`limit the electron thermal conductivity and to provide
`collisions sufficient for efficient energy transfer from
`electrons to heavy particles.
`According to [11], one can evaluate the ionization
`degree from the Saha equation if ambipolar diffusion (A)
`is negligible, i.e., if plasma density is sufficiently high:
`
`PLASMA PHYSICS REPORTS
`
`Vol. 21
`
`No. 5
`
`1995
`
`TSMC-1403 / Page 8 of 10
`
`

`

`408
`
`MOZGRIN et a1.
`
`the plasma density in the discharge passing to
`i.e.,
`regime 3 exceeds r16 2 5 X 1013 cm"?
`
`plasma. Then, the positive space charge layer thickness
`at the cathode su

This document is available on Docket Alarm but you must sign up to view it.


Or .

Accessing this document will incur an additional charge of $.

After purchase, you can access this document again without charge.

Accept $ Charge
throbber

Still Working On It

This document is taking longer than usual to download. This can happen if we need to contact the court directly to obtain the document and their servers are running slowly.

Give it another minute or two to complete, and then try the refresh button.

throbber

A few More Minutes ... Still Working

It can take up to 5 minutes for us to download a document if the court servers are running slowly.

Thank you for your continued patience.

This document could not be displayed.

We could not find this document within its docket. Please go back to the docket page and check the link. If that does not work, go back to the docket and refresh it to pull the newest information.

Your account does not support viewing this document.

You need a Paid Account to view this document. Click here to change your account type.

Your account does not support viewing this document.

Set your membership status to view this document.

With a Docket Alarm membership, you'll get a whole lot more, including:

  • Up-to-date information for this case.
  • Email alerts whenever there is an update.
  • Full text search for other cases.
  • Get email alerts whenever a new case matches your search.

Become a Member

One Moment Please

The filing “” is large (MB) and is being downloaded.

Please refresh this page in a few minutes to see if the filing has been downloaded. The filing will also be emailed to you when the download completes.

Your document is on its way!

If you do not receive the document in five minutes, contact support at support@docketalarm.com.

Sealed Document

We are unable to display this document, it may be under a court ordered seal.

If you have proper credentials to access the file, you may proceed directly to the court's system using your government issued username and password.


Access Government Site

We are redirecting you
to a mobile optimized page.





Document Unreadable or Corrupt

Refresh this Document
Go to the Docket

We are unable to display this document.

Refresh this Document
Go to the Docket