RU2176681C2 - Method of making coats in vacuum, device for making such coats and method of manufacture of said device - Google Patents

Abstract

FIELD: making of coats from metals, alloys, dielectrics, semiconductors and superconductors in electronic industry, space and atomic engineering, as well as in mechanical engineering and optical industry. SUBSTANCE: firing the discharge in generation zone is effected by means of defocused laser beam whose duration τ_gen is determined from relationship τ_gen≫ τ_las and by matching directional pattern of laser plasma flow with area of forming electrode gap break-down channel followed by superimposing magnetic fields on generation, transportation and condensation areas and distributing magnetic fluxes is generation, transportation and condensation areas in such way that anode part of discharge is used for partial closing of magnetic flux at which positive jump of optimal potential is formed; magnetic flux through profiled surface radially bounding the transportation area is equal to zero and magnetic flux in condensation area is equal to longitudinal magnetic flux in transportation area; ion component of plasma is directed to condensation surface and is separated from drop phase at transportation state in Langmuir layer area. Device for realization of said method is additionally provided with ion separator by transverse momenta and synchronizing unit; firing unit is provided with diaphragm having hole lying in one optical axis with end working surface of consumable cathode and input port at distance from lens which is equal to focal distance of focusing system lens. Plasma conduit body is made in form of screw cylinder formed by motion of ring over helical line on cylinder; profile of inner surface of body is triangular in shape. Electromagnetic system has two coils: one coil is connected in series in pumping circuit of laser radiator and second coil is connected in series in plasma generator discharge circuit. Device is provided with additional plasma ion separator by transverse momenta located in area of outlet end face of plasma conduit, additional plasma generator, plasma conduits, firing units, ion mixing chambers by transverse moment and synchronizing units. Method of manufacture of device for making coats realizes optimal conditions of construction of helical plasma-optical system and mixing chamber. EFFECT: enhanced efficiency. 23 cl, 32 dwg

Description

 The invention relates to the field of production of coatings from metals, alloys, dielectrics, semiconductors, superconductors and can be used in technical physics, electronic, space and nuclear technology, as well as in the engineering and optical industries.

 The aim of the invention is to improve the quality of multicomponent coatings and the expansion of technological capabilities.

 The method of producing coatings in vacuum was carried out using devices whose designs are presented in FIG. 1-32, and a method for manufacturing the device is shown in FIG. 25-32.

области стыковки соленоида анода и винтового соленоида, а также вариант изменения Ф_А при регулировании расстояния l_АП между ними; на фиг. 20 схематично изображена топология магнитного поля

в области стыковки соленоида анода и винтового соленоида, а также вариант изменения Ф_А при изменении плотности намотки в области торцевой рабочей поверхности; на фиг. 21 представлена конструкция устройства с плазмооптической системой в виде эквидистантных многозаходных винтовых цилиндров с камерой смешения со стороны выходных торцов и соленоидов, расположенных с тыльной стороны подложки; на фиг. 22 представлено устройство для получения многокомпонентных пленок в виде винтового плазмовода с камерой смешения на входном торце. На фиг. 23 схематично показана элементарная ячейка слоистого высокотемпературного сверхпроводника Y₁Ba₂C₃O₇; на фиг. 24 - временная диаграмма формирования как элементарной ячейки, так и тонкой пленки данного сверхпроводника. (Для фиг. 23 направление роста пленки совпадает с кристаллографической осью с; d_c - размер элементарной ячейки в направлении с; В - подрешетка ВаС; А - подрешетка У; на фиг. 25 схематично изображен корпус тороидального плазмовода, на фиг. 26 - способ его изготовления (прототип); на фиг. 27 и 28 - то же, но для корпуса винтового плазмовода (предлагаемый способ), а на фиг. 29 - схема расчета расстояния l_К между нормалями последовательных К сечений прямой трубы, а также схема расчета угла поворота γ_к между двумя последовательными сечениями сегмента; на фиг. 30 схематично показан корпус камеры смешения для N = 2 источников плазмы в плоскости осей выходного и входного сегмента; на фиг. 31 схематически показан вид сбоку на камеру смешения в плоскости, перпендикулярной плоскости осей выходного и входного сегмента; на фиг. 32 схематично показан вид сверху на камеру смешения в плоскости, перпендикулярной оси выходного сегмента.In FIG. 1 schematically shows the design of a pulsed plasma generator based on a DC vacuum arc with a laser ignition unit; in FIG. 2 shows a device for producing coatings with a screw plasma duct and an additional screw housing; in FIG. 3 shows a screw plasma duct; in FIG. 4 shows the ideal profile of the inner surface of the plasma duct; in FIG. 5 is a real profile of the inner surface of the plasma duct; in FIG. 6 shows another profile view of the inner surface of the plasma duct body; in FIG. 7 shows the profile of the inner surface of the plasma duct body and the distribution of potential U _cmt under the condition t _p > l _Lengm ; in FIG. 8 - the same, but under the condition t _p <l _Langm ; in FIG. 9 schematically shows the design of a device with a plasma-optical system in the form of equidistant screw plasma ducts with a mixing chamber; in FIG. 10 is a cross-sectional view along AA of FIG. 9; in FIG. 11 is a diagram for determining the minimum radial dimensions of the device (section by a plane perpendicular to the helical axis of the helical cylinder in the system N of multi-pass helical plasma ducts); in FIG. 12 shows a scan of the cylindrical surface of a straight cylinder passing through the helical axis of the system N of multi-pass helical plasma ducts (scheme for determining the minimum longitudinal dimensions of the device); in FIG. 13 schematically shows the design of the ion separator for transverse momenta (longitudinal section); and in FIG. 14 is a design of an ion separator for transverse momenta (cross section); in FIG. 15 shows a device for applying coatings in the form of two identical symmetrically located devices with opposite torsion of equidistant multi-start helical plasma ducts and with the consonant inclusion of their solenoids; in FIG. 16 shows a device for depositing coatings over large areas; in FIG. 17 shows a cross section of a deposition chamber; in FIG. 18 shows the design of a plasma generator with an electromechanical system in the laser ignition unit, as well as expansion diagrams of the plasma of the working substance with two zones ω _e τ _ei <1 and ω _e τ _ei > 1 and magnetic flux distribution Ф _К , Ф _А and Ф _Т ; in FIG. 19 schematically shows the topology of the magnetic field

the area of joining of the anode solenoid and the screw solenoid, as well as the variation of Ф _А when adjusting the distance l _АП between them; in FIG. 20 schematically shows the topology of the magnetic field

in the area of joining of the anode solenoid and the screw solenoid, as well as the option of changing Ф _А when the winding density changes in the region of the end working surface; in FIG. 21 shows the design of a device with a plasma-optical system in the form of equidistant multi-start helical cylinders with a mixing chamber on the side of the output ends and solenoids located on the back side of the substrate; in FIG. 22 shows a device for producing multicomponent films in the form of a helical plasma duct with a mixing chamber at the inlet end. In FIG. 23 schematically shows a unit cell of a layered high-temperature superconductor Y ₁ Ba ₂ C ₃ O ₇ ; in FIG. 24 is a timing diagram of the formation of both a unit cell and a thin film of a given superconductor. (For Fig. 23, the film growth direction coincides with the crystallographic axis c; d _c is the unit cell size in the direction c; B is the BaC sublattice; A is the Y sublattice; Fig. 25 schematically shows the body of the toroidal plasma duct, Fig. 26 is a method its manufacture (prototype); in Fig. 27 and 28 it is the same, but for the screw plasma duct housing (the proposed method), and in Fig. 29 is a diagram for calculating the distance l _K between the normals of consecutive K sections of a straight pipe, as well as a diagram for calculating the angle rotation γ _to between two consecutive sections of a segment; in Fig. 30 with schematically shows the housing of the mixing chamber for N = 2 plasma sources in the plane of the axes of the output and input segment; Fig. 31 schematically shows a side view of the mixing chamber in a plane perpendicular to the plane of the axes of the output and input segment; Fig. 32 schematically shows a top view of mixing chamber in a plane perpendicular to the axis of the output segment.

 A device for producing coatings in a vacuum contains pulsed plasma generators of solid substances, each of which contains connected to the capacitive storage 1 tires 2, forming a discharge circuit, a hollow anode 3 and a consumable cathode 4 with an end working surface 5 located inside the anode, an ignition unit 6 of the discharge, a solenoid 7, covering the anode, and a curvilinear plasma duct adjacent to the output end of the anode, made in the form of a hollow body 8 (Fig. 2, 9, 15, 16) connected to the capacitance Cn 9 (Fig. 2), bias circuit, and coil 10 ( Fig. 9, 15-16, 19-22), placed on the housing with a threaded profile 11 (Fig. 2, 4-9, 14, 19-22) on the inner surface.

 The ignition unit 6 in plasma generators (Fig. 1, 18) contains a laser emitter 14 located on the optical axis, oriented to the end face 5 of the cathode 4, a laser radiation input window 15, a focusing lens 16 located between the radiation input window 15 and the laser emitter 14, as well as a diaphragm 17 located on the optical axis between the end working surface 5 of the cathode (target) 4 and the laser input window at a distance equal to the focal length of the lens, as well as a longitudinally transparent transparent screen 18 (Fig. 1), fixed in the feed mechanism, moved by the drive 19 to move the cathode 4.

 In the second embodiment, in the area between the diaphragm 17 and the input window 15, a normally open shutter 12 is arranged in the form of a lever with a return spring 13. A smaller shutter lever is connected to the armature of the electromagnetic system, including the magnetic circuit 20 and two coils 21 and 22. One of the coils is turned on sequentially in the circuit 23 of the discharge of the pump lamps of the emitter 14, and the second according to the first and sequentially in the discharge circuit to the tires 2 of the plasma generator.

 At least one additional housing 25 (Fig. 2) is located inside the main screw housing, the outer and inner surfaces of which are profiled similarly to the main body of the plasma duct, and also parallel to it at each point of the cross section.

 One of the variants of the device for producing multicomponent plasma flows of solid substances (Fig. 9) contains three plasma generators G1, G2, and G3 equipped with screw plasma ducts with 10 solenoids located on hollow housings 8 of the plasma ducts made in the form of equidistant multi-start screw cylinders connected to mixing chamber 26 (Fig. 9, 21) containing solenoids (not shown in the drawing) and autonomous power sources of plasma generators in the form of capacitive storage devices Sn and a capacitive storage device of the potential of displacement of the camera body with Addressing and housings plasma ducts (not shown).

 The mixing chamber 26 is a continuation of the plasma ducts, while its body is formed by combining one end of the direct output cylinder and N input lines, so that the axes of the input cylinders extend along the generatrices of the cone facing the apex to the output end of the mixing chamber. Moreover, each of the input cylinders is surrounded by a solenoid, and the output cylinder is surrounded by N (N =! 3) solenoids, forming a magnetic system of the mixing chamber, each of the windings of the solenoid of the input cylinder is connected according to one of the windings of N solenoids of the output cylinder and to the device’s magnetic system.

 On the back side of the substrate, N (N = 3) solenoids are mutually coaxial and coaxial to the N solenoids of the output cylinder of the mixing chamber, the windings of which are connected in accordance with and sequentially in the device’s magnetic system, the length and diameter of the solenoids being equal to the length and diameter of the solenoids of the mixing cylinder output cylinder.

 An impulse ion separator (Fig. 13-15) coaxially adjacent to the output end of the plasma duct (or mixing chamber), including conductive walls 27 forming longitudinal channels, a substrate 28 located perpendicular to the longitudinal channels connected to the substrate bias source (not shown), moreover the substrate and the conductive channels are surrounded by a solenoid 29, the magnetic field lines of which are parallel to the channel walls and perpendicular to the substrate.

Moreover, the device for producing coatings contains N = 3 solenoids 31 on the back side of the substrate (see Fig. 21) or two similar devices (see Fig. 14), if their number is more than two (see Fig. 15) located symmetrically relative to the plane of the substrate 28, and the torsion directions of the system of multi-pass helical plasma ducts in opposite devices are opposite, the solenoids 10 of the plasma ducts are included according to, and the distance L _П∥ in the deposition chamber 30 (Fig. 15, 16, 21) (the substrate is the output ends of the devices) the second case is determined and inequality L _П L ≥ V _PL∥ • τ _П , where V _PL∥ is the longitudinal velocity of plasma flows, τ _П is the time of radial relaxation of the plasma density.

The device operates as follows. The power source 1 of the plasma generator is charged to a voltage of ~ 600 V (total energy ~ 10 ³ J), which is applied to the anode 2 and cathode 3 (the target being consumed). Then, the synchronization unit according to a given program according to the distribution of substances in the unit cell of the substance sets the synchronization mode of ignition of discharges, as well as the inlet of reactive gas (see Fig. 23). The operating voltage is applied to the laser emitter, as a result of which a laser pulse is generated of a duration of ~ 20 ns and an energy of 10 ^-1 J in the direction of the optical axis. The laser radiation is focused by the lens 16, then introduced into the plasma generator (in the region of the end face of the sacrificial target) through the window 15, the transparent screen 13 and the diaphragm 17, for interaction with the target material. A laser plasma stream is injected from this zone into the interelectrode gap and a pulsed vacuum-arc discharge of a direct current of a cathode form is excited. The erosion products of the target (droplet phase, ionized and neutral components) flow out through the output end of the anode into the transport region. The process of charging a capacitive storage device 1, ignition of the discharge and its burning is set by the synchronization unit. During the process, the end working surface 5 of the target 4 is produced. As it develops, the zone of exposure to laser radiation moves along the end working surface of the target. The sacrificial cathode 4 as it is generated is periodically moved in the longitudinal direction by the feed mechanism 19 to the output end of the generator. Thus, by periodically supplying the sacrificial cathode, the initial geometry of the interelectrode gap is maintained. As plasma flows are transported in plasma ducts, they are separated from the droplet phase and high-energy ions and enter the mixing chamber, where they are mixed and enter the condensation region (onto the substrate) through the output end of the mixing chamber as a multicomponent plasma stream of a given chemical composition. When deposited on profiled substrates after mixing, the flows are transported through the longitudinal channels of the plasma ion separator by transverse momenta, and then condense.

 Three options for implementing the method in time are possible: synchronous (which is, in fact, mixing) or asynchronous excitation of spatially separated vacuum-arc discharges, as well as a combination of these modes.

In the method of producing coatings in vacuum, two mixing options are possible according to the criterion of the initial phase state: only solid-phase substances or solid-phase and gaseous substances are mixed. The latter option is implemented by injecting reactive gas (O ₂ ; N ₂ ) into the generation region pulsed and synchronously with the excitation of vacuum arcs.

обусловливает два возможных режима реализации смешения. В первом случае (r_i < D_с) наблюдается однородное радиальное распределение химических элементов (ионов) в области смешения (и покрытии). Во втором случае (r_i <D_с) смешение затруднено, так как ограничено магнитным полем радиальное перемещение ионов, ситуация аналогична, если λ_ii < D_c. В варианте (когда r_i>D_с) возможны два случая его реализации. В первом случае сначала смешивают многофазные потоки плазмы, а затем многофазный и многокомпонентный поток плазмы сепарируется путем транспортирования в криволинейной (винтовой) плазмооптической системе. Подобная последовательность операций не исключает массообмен по капельной фазе между пространственно разделенными вакуумными разрядами (с катодами различного химического состава каждого смешиваемого потока, а следовательно, и покрытия. Во втором случае исходные многофазные потоки сначала сепарируют от капельной фазы, а затем смешивают, что исключает массообмен по капельной фазе между пространственно разделенными вакуумными разрядами (катодами), поэтому химический состав покрытия будет определяться однозначно величинами зарядов, протекающих в каждом из разрядов - следовательно, второй случай является оптимальным, что, в сущности, обусловлено различной геометрической вероятностью р пролета капельной фазы вдоль профилированной поверхности, ограничивающей зону транспортирования. В способе величина эродирующей массы единичного импульса в импульсном дуговом разряде определяется равенством: M= k•q где k - коэффициент эрозии, q - заряд, протекающий в цепи разряда. Так как q = C•U, то эродируемой массой можно управлять как изменением емкости С, так и напряжением U в широких пределах, что позволяет при наличии нескольких пространственно-разделенных вакуумных дуг и временной последовательности их возбуждения формировать в одном цикле элементарную ячейку со сложной структурой, а многократное повторение таких циклов позволяет формировать пленку заданной толщины без существенных структурных неоднородностей в направлении роста. Таким образом состав покрытия в способе задают пространственным разделением вакуумных дуг и регулированием режима их синхронизации на этапе возбуждения, а также регулированием величины заряда электричества, протекшего в разряде за счет изменения напряжения и емкости, причем возбуждение вакуумных дуг осуществляют циклически во времени синхронно или последовательно путем инжекции лазерной плазмы, транспортирование осуществляют в пространственно разделенных областях, а их смешение в области пространственного совмещения магнитных полей областей транспортирования, при этом состав многокомпонентного покрытия задают из условия N_A/N_B/N_C=K_At_A C_HAU_A/K_Bt_BC_HBU_B/K_Ct_CC_HCU_C, где N_A, N_B, N_C - концентрация химических элементов A, B, C в многокомпонентном покрытии, K_A, K_B, K_C - коэффициент эрозии плазмообразующих веществ A, B, C, t_A, t_B, t_C - коэффициент транспортирования потоков плазмы A, B, C, C_HA, C_HB, C_HC..., U_A, U_B, U_C - величины емкости (Ф) и изменений напряжения (В) емкостных накопителей, причем последовательность импульсов задают, исходя из порядка расположения подрешеток (слоев) формируемой кристаллической элементарной ячейки слоистого вещества покрытия (например, сверхпроводника (сэндвич- структуры), число импульсов плазмы n_A, n_B ... в последовательности для каждой подрешетки (слоя) определяют из выражения

где ρ_А,В,С - плотность наносимого вещества (химического элемента) А, В, С в покрытии, кг/м³; S_т - площадь поперечного сечения А, В, С области транспортирования (смешения), м²; d_A,B,C - толщина подрешетки (слоя) А, В, С в элементарной ячейке (сэндвич- структура), м; q_A,B,C - заряд, протекший в цепи разряда, Кл, а число циклов m последовательности импульсов плазмы определяются из равенства d_я•m= l, где d_я - размер элементарной ячейки (сэндвич-структуры) в направлении роста покрытия, м; l - толщина покрытия (сэндвич-структуры), м.The presence of a magnetic field in the mixing region

determines two possible modes of realization of mixing. In the first case (r _i <D _s ), a uniform radial distribution of chemical elements (ions) in the mixing region (and coating) is observed. In the second case (r _i <D _s ), the mixing is difficult, since the radial movement of ions is limited by the magnetic field, the situation is similar if λ _ii <D _c . In the variant (when r _i > D _c ) two cases of its implementation are possible. In the first case, multiphase plasma flows are mixed first, and then the multiphase and multicomponent plasma flows are separated by transportation in a curvilinear (screw) plasma-optical system. Such a sequence of operations does not exclude droplet phase mass transfer between spatially separated vacuum discharges (with cathodes of different chemical composition of each mixed flow, and hence the coating. In the second case, the initial multiphase flows are first separated from the droplet phase and then mixed, which excludes mass transfer by drop phase between spatially separated vacuum discharges (cathodes), therefore, the chemical composition of the coating will be determined uniquely by the magnitudes of the charges, p swelling in each of the discharges - therefore, the second case is optimal, which, in essence, is due to the different geometric probability p of the passage of the droplet phase along the profiled surface that limits the transportation zone. In the method, the value of the eroding mass of a single pulse in a pulsed arc discharge is determined by the equation: M = k • q where k is the erosion coefficient, q is the charge flowing in the discharge circuit, since q = C • U, the eroded mass can be controlled both by a change in capacitance C and voltage U in wide which makes it possible, in the presence of several spatially separated vacuum arcs and a time sequence of their excitation, to form a unit cell with a complex structure in one cycle, and repeated repetition of such cycles allows you to form a film of a given thickness without significant structural heterogeneities in the growth direction. Thus, the coating composition in the method is determined by the spatial separation of the vacuum arcs and the regulation of their synchronization mode at the stage of excitation, as well as by the regulation of the charge of electricity flowing in the discharge due to changes in voltage and capacitance, and the excitation of vacuum arcs is carried out cyclically in time synchronously or sequentially by injection laser plasma, transportation is carried out in spatially separated areas, and their mixing in the field of spatial combination of magnetic transport areas, and the composition of the multicomponent coating is specified from the condition N _A / N _B / N _C = K _A t _A C _HA U _A / K _B t _B C _HB U _B / K _C t _C C _HC U _C , where N _A , N _B , N _C - concentration of chemical elements A, B, C in a multicomponent coating, K _A , K _B , K _C - erosion coefficient of plasma-forming substances A, B, C, t _A , t _B , t _C - transport coefficient plasma flows A, B, C, C _HA , C _HB , C _HC ..., U _A , U _B , U _C are the capacitance (Ф) and voltage changes (V) of capacitive storage, and the pulse sequence is set based on the order of arrangement of sublattices (layers) of the formed crystal tally unit cell of a layered coating material (for example, a superconductor (sandwich structure), the number of plasma pulses n _A , n _B ... in the sequence for each sublattice (layer) is determined from the expression

where ρ _{A, B, C} is the density of the applied substance (chemical element) A, B, C in the coating, kg / m ³ ; S _t is the cross-sectional area A, B, C of the transportation (mixing) area, m ² ; d _{A, B, C} is the thickness of the sublattice (layer) A, B, C in the unit cell (sandwich structure), m; q _{A, B, C} is the charge flowing in the discharge circuit, C, and the number of cycles m of the sequence of plasma pulses is determined from the equality d _i • m = l, where d _i is the size of the unit cell (sandwich structure) in the direction of coating growth, m; l - coating thickness (sandwich structure), m

 The volumetric growth mechanism does not provide conditions for the formation of layered structures due to the fact that the height of the nucleus is greater than the monolayer (sublattice). This mechanism contributes to the chaotic distribution of layered structure elements. With two-dimensional (flat) film growth (one, at most three atoms - one nucleus), all conditions for the formation of films with a layered structure are satisfied and they can be provided with two characteristics of the condensed medium: the density and kinetic energy of ions. The density of condensed particles in the pulsed mode increases by 3-6 orders of magnitude, which in turn provides a reduction in the size of the critical nucleus by approximately 3-6 times.

Such an increase in the density of the vapor (condensed medium) provides an increase of 10 ³ - 10 ⁶ times the probability of interaction between the ion and the three-dimensional nucleus, as a result of which the latter is destroyed, since the binding energy of the adatoms in the nucleus in the growth direction (fractions - units of eV) is significantly less than the kinetic energy of adsorbed ions (tens of eV). All this together ensures the formation of films with a layered structure at the level of a monolayer.

However, excess ion energy can impair the quality of the coating. The formation of radiation defects on the condensation surface is inherent in all plasma condensation methods and is due to the presence in the fraction of the ionic component of the energy greater than the energy (W _def ) of the formation of the defect on the condensation surface. The method has two sources of radiation defects: at the stage of ignition, laser plasma and at the generation stage, plasma of a vacuum arc. To eliminate the influence of these sources, before condensation, it is necessary to separate ions from the plasma whose energy is greater than the energy of defect formation, and to leave ions for which the inverse relation is true. The maximum energy in this method is determined by two factors: firstly, the magnitude of the potential on the wall, limiting the transport region and, secondly, the magnitude of the bias potential U _{see the} surface of the condensation (substrate), which depends on the conductivity of the substrate.

 Two cases are possible when the substrate is made of a dielectric or semi-insulating semiconductor or the substrate is made of metal or a highly alloyed semiconductor.

In the first case, it is not possible to adjust the value of U _cm.s. with the aim of limiting the maximum ion energy in the condensation region, since a floating potential equal to (eU _cm.s. ≃ 5Te) is established on the dielectric substrate. Therefore, limiting the maximum energy of a condensed ion is possible only by changing the value of the potential on the wall that bounds the transport zone, i.e. eU _{cm t} + 5Te) <W _def. In the second case, one can also change the value of U _cm.s. , giving a positive or negative bias potential so that eU _cm.s. ± eU _cm.s. • Z <W _def .

The cases considered are similar if the condensed film is a dielectric or a semi-insulating semiconductor, metal or a highly alloyed semiconductor. At the transportation stage, it is necessary to ensure the exit of all plasma ions (laser and vacuum arc) to the wall of the plasma duct body, where they will be separated by energies at the potential barrier U _cm . If their energy exceeds the value of the potential barrier, then this part of the ions goes to the wall, and the remaining part, for which the inverse relation holds, goes into the condensation region. The exit of all ions to a potential barrier is observed if two conditions are satisfied: first, if λ _ii > D _T otherwise (if λ _ii <D _T ) the ion bypasses the surface of the transport region due to collisions in the transport region. The second necessary condition for the energy separation of ions is observed if r _i > D _t . Otherwise (if r _i <D _t ) due to the magnetization of the ionic component in the volume of the transport region, all ions pass the surface of the transport region. The indicated disequilibrium in the distribution is maintained at the mean free path of the ion λ _ii , therefore, the condensation surface is set at a distance L _k less than the mean free path of the ions (L _k <λ _ii ). Otherwise (if L _k > λ _ii ) due to ion-ion collisions, the nonequilibrium is eliminated and again high-energy ions appear in the plasma stream.

 The initial chemical composition of the plasma-forming substance may vary due to a number of reasons, which in turn are determined by the component composition of the working substance. Thus, the initial composition of a one-component plasma-forming substance changes due, firstly, to the use of material of ignition electrodes with a chemical composition different from the chemical composition of the cathode material (at the stage of excitation of the main discharge), and secondly, to the development of anode thermal instability as at the stage of plasma generation, and at the stage of its transportation and mixing. Due to the peculiar circulation of the introduced impurity (formation ---> condensation on the substrate and on other structural elements ---> re-evaporation from these structural elements), it must be eliminated at each stage, otherwise, an impurity will inevitably appear at all stages of the method, and also in a thin film of condensate on a substrate.

 Changes in the chemical composition of the coating during the ignition phase can be minimized by using laser ignition.

 In laser firing, there are a number of reasons for the introduced chemical impurities. The first is caused by the process of self-cleaning (transparency regeneration) of the working surface of the optical input window from the condensation products of the plasma components of the vacuum arc (droplet phase, ionized and neutral components).

 In addition, the level of chemical purity in the laser-ignited solid-state plasma generator can be violated due to erosion of the anode material and the plasma duct case, provided that the laser radiation from the end working surface of the cathode is partially reflected on the anode (plasma duct body).

, где α - коэффициент поглощения, a - температуропроводность рабочего вещества.Another variant of changing the chemical composition of a multicomponent working substance is possible, due to the duration of τ _l of the ignition stage, since two limiting variants are possible: τ _l ≪ τ _g and τ _l ≥ τ _g . In the case of τ _l ≥ τ _g , cathode spots are localized in the zone of interaction of laser radiation with the end working surface, which determines the localization of heat sources (laser radiation, cathode spots), their melting isotherms overlap and the formation of a continuous fusion zone (excess liquid phase), which stimulates the process of sublimation of the low-melting components of the working substance, and also increases the energy consumption for excitation. Thus, the mode τ _l ≪ τ _{g of} modulated Q factor is most optimal when excited, when the condition

where α is the absorption coefficient, a is the thermal diffusivity of the working substance.

Another source of impurities in the method is the thermal anode instability, which occurs due to the nonuniform distribution of the electric field near the microinhomogeneities of the surfaces of the structural elements (anode, plasma duct housing, mixing chamber housing at l _leng. > H _n (where l _leng. Is the Langmuir layer size, h _H - height of irregularities) occurs spatial redistribution of current transfer from the plasma to the surface due to the local increase in electric field that causes high heterogeneity tep ovyh sources and their location in said inhomogeneities Due hindered heatsink with microroughness when h _n> d _n (where h _n and d _n - height and diameter of the microscopic irregularities of structural elements, respectively). they are exposed to intense thermal action, accumulating energy, and development time thermal anode instabilities melt (explode), creating a stream of impurities.

Thus, in order to reduce the level of impurities from the generation region due to thermal anode instability, it is necessary to minimize the anodic potential jump ΔU _a (i.e., ΔU _a ) = 0), or to limit the time of plasma flow generation (τ _g ) by the condition τ _g <τ _tan where τ _{T.AN is the} time of the development of thermal anode instability. The first option (ΔU _a = 0) is unacceptable in the method, since there is no potential barrier near the anode and the ions go to the anode, reducing the efficiency of plasma generation. Therefore, Δ U _a > 0 and amounts to several ionic temperature values, and elimination of the impurity at the generation stage is achieved by limiting its duration.

In order to reduce impurities from the transport region (the inner surface of the plasma duct body) in the case of thermal anode instability, it is not possible to fulfill the condition U _cm = 0 (since to increase the efficiency of transporting the plasma (ion) flux through the plasma duct, a positive bias potential is always applied to its body U _{see t} > 0, equal in value to several values of ionic temperatures), therefore, it is necessary to limit the transportation time τ _{T by the} condition τ _T <τ _T.AN.
The nature of the formation of introduced chemical impurities at the stage of mixing the plasma flows (in the mixing chamber) is similar to the nature of the formation of chemical impurities at the stage of transportation (in the plasma duct), therefore, the condition for their absence can be written down: τ _C <τ _T.AN.S , where τ _C is the time mixing plasma flows.

 The use of a multicomponent substance as a plasma-forming substance causes, in addition to thermal anode instability, a number of other reasons for the composition violation, which are fundamentally inherent only to such a plasma-forming substance and which (reasons) are observed at all stages of the method.

At the stage of generating a multicomponent plasma flow, there is a mechanism that causes a change in the chemical composition directly in the region of plasma formation — cathode spots of a vacuum arc that differ in lifetime. From the point of view of changing the chemical composition in the plasma of a multicomponent working substance, the cathode spots are unequal, since the fractional sublimation of the fusible components of the plasma-forming substance most effectively occurs in associations (a group of cathode spots of the second kind). “Technological” cathode spots are spots of the second kind, since the energy spectrum of their ions is more optimal W _i <W _def , therefore, the discharge duration is determined from the condition τ _Г ≤ τ _II , where τ _II is the lifetime of the cathode spots of the second kind. The latter circumstance, in addition to lowering the level of fractional sublimation of the fusible components of the substance, helps to reduce the level of the droplet phase generated from the cathode spots, due to the smaller volumes of the liquid phase in the region of plasma formation.

, а также скрещенного с ним электрического поля

может стать причиной пространственного перераспределения ионов в многокомпонентном потоке по их массам M_i, m_i и кратности ионизации (Z) даже при наличии в исходном потоке после этапа генерирования пространственной однородности по этим параметрам.At the stage of transportation, in the case of multicomponent plasma flows, there are a number of mechanisms that determine the change in the chemical composition due to the presence of ions of chemical elements with different masses M _i , m _i in the plasma, as well as differences in the ionization ratio (Z =! 1, 2, 3. ..) ions, where M _i , m _i are the mass of the heavy and light ions, respectively. The presence in the method of a curved and inhomogeneous magnetic field

as well as an electric field crossed with it

can cause a spatial redistribution of ions in a multicomponent stream over their masses M _i , m _i and ionization multiplicity (Z) even if there is spatial homogeneity in these parameters in the initial stream after the stage of generation.

полях и, во-вторых, относительно оси винтового магнитного поля.Redistribution can be of two types: radial and longitudinal. The radial redistribution can be caused by the drift of curvature, the gradient drift of ions of different masses and the multiplicity of ionization Z, as well as by centrifugal forces due to the rotation of the plasma, firstly, in crossed

fields and, secondly, relative to the axis of the helical magnetic field.

электрон-ионных столкновений на длине L_т зоны транспортирования, равной

где λ_ei - величина расстояния электрон-ионных столкновений. Обычно L_T= (5-10)λ_ei, поэтому градиентный дрейф и дрейф кривизны не являются в способе лимитирующими факторами перераспределения ионов плазмы по M_i, m_i и Z. Эффект радиального перераспределения ионов за счет вращения плазмы в

полях имеет максимальную эффективность при замагниченности ионной компоненты, так как в этот вид движения вовлечена непосредственно ионная компонента плазмы. В случае (когда r_i>D_т) непосредственно в

вращении участвует только электронная компонента плазмы, так как она замагничена (r_i<<D_т ω_eτ_ei> 1 , где ω_eτ_ei> 1 - параметр Холла электронов), а вовлечение ионной компоненты возможно за счет электрон-ионных столкновений, т.е. аналогично градиентным дрейфам. Обычно L_т ≤ 10² см и определяется сверху условиями запирания капельной фазы.Since the ionic component of the plasma at the stage of transportation is not magnetized (r _i > D _t ), therefore, these drifts can be realized due to

electron-ion collisions over a length L _{t of} the transport zone equal to

where λ _ei is the distance of electron-ion collisions. Usually L _T = (5-10) λ _ei , therefore gradient drift and curvature drift are not in the method the limiting factors of redistribution of plasma ions over M _i , m _i and Z. The effect of radial redistribution of ions due to plasma rotation in

In fields, it has maximum efficiency with magnetization of the ionic component, since the plasma ion component is directly involved in this type of motion. In the case (when r _i > D _t ) directly in

only the electron component of the plasma is involved in the rotation, since it is magnetized (r _i << D _t ω _e τ _ei > 1, where ω _e τ _ei > 1 is the electron Hall parameter), and the involvement of the ion component is possible due to electron-ion collisions, those. similar to gradient drifts. Usually L _t ≤ 10 ² cm and is determined from above by the locking conditions of the droplet phase.

The method for redistribution is limited by centrifugal forces due to the rotation of the plasma relative to the axis of the screw cylinder, since ions are directly involved in this type of motion even when r _i > D _t . This mechanism is absent if:
1. N _rep. > N _volume. those. λ _ii ≥ D _T , where N _rep. , N _volume. - the number of collisions of ions with the surface bounding the transport zone, and in the volume between themselves, respectively.

, когда λ_ii< D_т, где V_i - скорость иона, n - концентрация плазмы, r_т - радиус кривизны зоны транспортирования, Т_i - ионная температура.2.

when λ _ii <D _t , where V _i is the ion velocity, n is the plasma concentration, r _t is the radius of curvature of the transport zone, T _i is the ion temperature.

3. τ _ave <τ _sec. when λ _ii <D _t , where τ _ave , τ _sec. - time of flight and separation, respectively.

In the first case, the volume redistribution caused by the volume source due to directionally selective collisions of ions and different masses M _i , m _i is destroyed (randomized) when the ions collide with the surface, since it is not selective in direction.

вследствие инерционности процесса разделения (τ_разд.) радиальное перераспределение не успевает реализоваться за время пролета (τ_пр.) плазмой области транспортирования в области конденсации (подложки), однако установится стационарное радиальное перераспределение, если τ_разд.≤ τ_пр. Величина стационарного (т. е. при τ_пр.> τ_разд.) коэффициента разделения β химических компонент в поле центробежных сил равна

(при λ_ii< D_Т ), где ΔM=M_i-m_i - разница в массах ионов, V_∥П - скорость потока плазмы, r_т - радиус кривизны зоны транспортирования, Т_i - ионная температура.In the second version, even with λ _ii <D _t , the decisive factor is the chaotic movement of ions, since the pressure of the chaotic movement of ions is much greater than the centrifugal pressure on the plasma, so the radial redistribution is not realized, since at any moment it is destroyed by the chaotic movement of ions. In the third case, even when conditions are met

due to the inertia of the separation process (τ _sec. ), the radial redistribution does not manage to be realized during the time of flight (τ _ave. ) by the plasma of the transport region in the condensation region (substrate), however, a stationary radial redistribution is established if τ _sec. ≤ τ _ave. The value of the stationary (i.e., at _ave. > Τ _aux. ) Separation coefficient β of chemical components in the field of centrifugal forces is

(at λ _ii <D _Т ), where ΔM = M _i -m _i is the difference in ion masses, V _{∥ П} is the plasma flow rate, r _t is the radius of curvature of the transport zone, and T _i is the ion temperature.

, чем в тороидальных, выражен эффект радиального перераспределения многокомпонентных потоков плазмы в поле центробежных сил.Therefore, avoid radial redistribution by
transportation stage is possible if λ _ii > D _T , and with increasing plasma density, when λ _ii <D _T only under the condition τ _sec. > τ _ave. To a lesser extent in helical magnetic fields

than in toroidal ones, the effect of radial redistribution of multicomponent plasma flows in the field of centrifugal forces is expressed.

и электрического поля

и вариантом исполнения источника магнитного поля

на этапе транспортирования. Электрическое поле создается прямым подключением корпуса плазмовода к положительному полюсу емкостного накопителя, т.е. оно не модулируется во времени. Возможны три варианта создания импульсного магнитного поля: сторонним источником с длительностью импульса тока τ_ЕH= τ_Г+τ_ПР., непосредственно током вакуумной дуги, длительностью τ_ЕН= τ_Г и сторонним источником, длительностью τ_пр. на заднем фронте тока вакуумной дуги: τ_ЕН= τ_Г+τ_пр., т.е. смешанным способом. Влияние продольного перераспределения на состав пленки отсутствует, если, во-первых, толщина конденсируемой пленки за импульс плазмы меньше или равна толщине монослоя

(что обычно выполняется для импульсных генераторов плазмы твердых веществ) и, во-вторых, когда магнитное поле

создается сторонним источником тока, длительностью τ_ЕН= τ_Г+ τ_пр. или смешанным способом. Для создания

обычно используют только сильноточечные (~5•10³А) цепи вакуумной дуги, что позволяет снизить энергозатраты и синхронизировать моменты появления плазмы и магнитного поля на этапе транспортирования. В этом случае на заднем фронте потока плазмы на время τ_пр. в момент погасания вакуумной дуги отсутствуют условия транспортирования плазмы по магнитному полю, поэтому плазма, обогащенная тяжелой ионной компонентой, конденсируется в зоне транспортирования (на поверхности плазмовода), обусловливая избыток легкой ионной компоненты в зоне подложки (в пленке).At the transportation stage, a longitudinal redistribution of chemical components in the plasma stream (and film) is possible, due to different speeds V _{i of} ions with different masses (M _i , m _i ) at the same ion temperature T _i and which in turn is determined by a number of factors: the thickness of the film condensed in one plasma pulse by the time τ _EH of the magnetic

and electric field

and an embodiment of a magnetic field source

at the transportation stage. The electric field is created by directly connecting the plasma duct body to the positive pole of the capacitive storage device, i.e. it is not modulated in time. Three options for creating a pulsed magnetic field are possible: an external source with a current pulse duration τ _ЕH = τ _Г + _{τПР .} , directly by the vacuum arc current, duration τ _ЕН = τ _Г and a third-party source, duration τ _av at the trailing edge of the vacuum arc current: τ _ЕН = τ _Г + τ _av , i.e. in a mixed way. There is no influence of longitudinal redistribution on the composition of the film if, firstly, the thickness of the condensed film per plasma pulse is less than or equal to the thickness of the monolayer

(which is usually done for pulsed solid-state plasma generators) and, secondly, when the magnetic field

created by an external current source, duration τ _ЕН = τ _Г + τ _пр. or in a mixed way. For creating

usually only high-strength (~ 5 • 10 ³ A) vacuum arc circuits are used, which allows reducing energy consumption and synchronizing the appearance of plasma and magnetic field at the stage of transportation. In this case, there are no conditions for transporting the plasma over the magnetic field at the trailing edge of the plasma flow for a time τ _av at the time of extinction of the vacuum arc; therefore, the plasma enriched in the heavy ion component condenses in the transport zone (on the surface of the plasma duct), causing an excess of the light ion component in the area of the substrate (in the film).

 The uniformity of the thickness of the films during their deposition from plasma in longitudinal magnetic and radial electric fields depends on a number of factors: the degree of plasma ionization, the magnitude and topology of the magnetic field, and also the magnitude of the electric field. If the degree of plasma ionization is less than unity, then the film profile consists of the spatial distribution of the neutral component, which is independent of external fields, as well as the spatial distribution of the ionized component, which is determined by the topology of the magnetic field in the volume and the electric field on the surface.

полях и получают плазменный поток со степенью ионизации, равной единице. В данном случае профиль пленки определяется только топологией магнитного поля в объеме плазмы и радиальным электрическим полем в ленгмюровском слое (l_ленгм.). Топология магнитного поля у поверхности конденсации определяется распределением напряженности магнитного поля зоны транспортирования. При этом величина неоднородности толщины пленки

в данном случае определяется величиной неоднородности магнитного поля

. Последнее справедливо только в объеме плазмы, где отсутствует электрическое поле, которое экранируется на поверхности (в толщине слоя l_ленгм плазмы.To eliminate this drawback, the plasma stream is transported in crossed

fields and receive a plasma stream with a degree of ionization equal to unity. In this case, the film profile is determined only by the topology of the magnetic field in the plasma volume and the radial electric field in the Langmuir layer (l _Langm. ). The topology of the magnetic field at the condensation surface is determined by the distribution of the magnetic field strength of the transportation zone. The magnitude of the heterogeneity of the film thickness

in this case is determined by the magnitude of the inhomogeneity of the magnetic field

. The latter is true only in the plasma volume, where there is no electric field that is screened on the surface (in the layer thickness l _{is the} plasma _langm .

при истечении плазмы на малогабаритную подложку из одного выходного торца плазмовода (камеры смешения). На практике возможен вариант истечения на крупногабаритную подложку (>Ф 200 нескольких пространственно разделенных потоков плазмы из ряда выходных торцов плазмоводов (камер смешения)). В данном случае равномерность покрытия определяется еще двумя факторами: расстоянием L_П∥ между выходными торцами устройств и подложкой и нестабильностью генерирования плазмы в каждом из генераторов. Первая причина лимитируется временем τ_П радиальной релаксации плотности плазмы. Оптимальной величиной расстояния L_П∥ является величина, определяемая из неравенства L_П∥≥ V_ПЛ∥•τ_П , где V_ПЛ∥ - продольная скорость потока плазмы. Однако это условие не является достаточным, так как равномерность покрытия на большей площади может нарушаться вследствие нестабильности возбуждения вакуумных дуг в генераторах плазмы. В данном случае неравномерность

покрытия и нестабильность

процесса возбуждения дуговых разрядов в генераторах Г1, Г2... Г18 (см. фиг. 19) связаны соотношением

, причем

зависит от числа N генераторов в устройстве, т.е.

. Поэтому чем больше число генераторов N в устройстве, тем больше равномерность покрытия по толщине. Конструкции устройств для получения покрытий позволяют сформировать продольный общий поток плазмы из нескольких индивидуальных потоков плазмы. Однако возможен и другой вариант формирования из нескольких радиальных индивидуальных потоков плазмы для осаждения покрытий на подложки в виде тела вращения или на подложки другой формы, расположенных на карусели. В данном случае для получения равномерного по толщине покрытия и соблюдения химического состава необходимо соблюдение некоторых требований. Цилиндрическая подложка или карусель с подложками может быть расположена за пределами зоны перекрытия индивидуальных импульсных потоков.The indicated technical solutions allow solving the issue of uniformity of coatings

when plasma flows onto a small-sized substrate from one output end of the plasma duct (mixing chamber). In practice, it is possible to discharge onto a large-sized substrate (> Ф 200 several spatially separated plasma flows from a number of output ends of the plasma ducts (mixing chambers)). In this case, the uniformity of the coating is determined by two more factors: the distance L _П∥ between the output ends of the devices and the substrate and the instability of plasma generation in each of the generators. The first reason is limited by the time τ _{P of} radial relaxation of the plasma density. The optimal value of the distance L _П∥ is the value determined from the inequality L _П∥ ≥ V _PL∥ • τ _P , where V _PL∥ is the longitudinal velocity of the plasma flow. However, this condition is not sufficient, since the uniformity of the coating over a larger area may be violated due to the instability of the excitation of vacuum arcs in plasma generators. In this case, the unevenness

coatings and instability

the process of excitation of arc discharges in the generators G1, G2 ... G18 (see Fig. 19) are related by the relation

, and

depends on the number N of generators in the device, i.e.

. Therefore, the larger the number of N generators in the device, the greater the uniformity of the coating over the thickness. The designs of the devices for producing coatings make it possible to form a longitudinal total plasma flow from several individual plasma flows. However, another variant is possible of forming from several radial individual plasma flows for deposition of coatings on substrates in the form of a body of revolution or on substrates of another shape located on the carousel. In this case, to obtain a uniform coating thickness and compliance with the chemical composition, certain requirements must be observed. A cylindrical substrate or a carousel with substrates may be located outside the overlapping zone of the individual impulse flows.

где V_∥П, V_⊥П - продольная и поперечная составляющая скорости иона соответственно; j_∥, j_⊥ - плотность потока ионов в продольном и поперечном направлении соответственно; n - концентрация ионов в плазме.In this case, the uniformity of the thickness will be violated at the boundary of the spatial alignment of two (three) flows in the region of their Langmuir layers. The condition for coating uniformity in this region (combining Langmuir layers) is the location of the substrate at a distance from the plasma boundary equal to l _Langm (L _{k, r} ≥l _Langm ). In this case, the Langmuir layers are summed up, as a result of which their total concentration profile is equal to its value in the plasma volume. "Dusting" of the vertical walls of profiled surfaces, which is due to the presence of the transverse component of the ion velocity V _{⊥ П in the} plasma ion component. The ratio of film thicknesses on the vertical wall and on the surface of the substrate can be estimated from the expression

where V _∥P , V _⊥P - the longitudinal and transverse components of the ion velocity, respectively; j _∥ , j _⊥ - ion flux density in the longitudinal and transverse directions, respectively; n is the concentration of ions in the plasma.

, и оставить в потоке ионы, для которых

. При этом величина расстояния от начала зоны сепарации до поверхности конденсации (подложки) должна быть меньше величины V_∥П•τ_ii, равной длине свободного пробега ионов в потоке плазмы, где

, τ_ii - время между ион-ионными столкновениями в плазме; τ_ei - время между электрон-ионными столкновениями; m_e, m_i - масса электрона и иона соответственно; Т_e, Т_i - температура электронов и ионов соответственно. За время пролета τ_c плазмой сепарации длиной h_c ионы с компонентой V_⊥П должны быть отсепарированы, поэтому внутренний объем зоны сепарации должен быть разделен продольными каналами с поперечным размером канала, равным V_⊥•δ_c, где V_⊥ - поперечная скорость иона после сепарации (V_⊥<< V_⊥П). Это условие является только необходимым, но не достаточным. Достаточным условием является отсутствие положительного потенциала на стенках каналов. Для устранения ухода от стенки ионов с компонентами скоростей

необходимо, чтобы подвижности электронов и ионов в поперечном направлении были приблизительно одинаковы. Поэтому на зону сепарации и конденсации накладывают продольное магнитное поле, что необходимо для повышения производительности и однородности

толщины покрытий с другой стороны, причем его величину определяют из неравенства r_i> V_⊥•τ_c>> r_e, где V_⊥•τ_c= b - поперечный размер канала; r_e, r_i - ларморовский радиус электрона, иона соответственно; V_⊥П - поперечная скорость иона до сепарации;

- время пролета плазмой зоны сепарации.To eliminate the “dusting” of the vertical walls of the profile, it is necessary to separate ions, for which

, and leave ions for which

. In this case, the distance from the beginning of the separation zone to the condensation surface (substrate) should be less than V _{∥ П} • τ _ii , equal to the mean free path of ions in the plasma stream, where

, τ _ii is the time between ion-ion collisions in the plasma; τ _ei is the time between electron-ion collisions; m _e , m _i are the mass of the electron and ion, respectively; T _e , T _i - temperature of electrons and ions, respectively. During the transit time τ _{with a} separation plasma of length h _c, ions with a component V _{⊥ П} should be separated, therefore, the internal volume of the separation zone should be separated by longitudinal channels with a transverse channel size equal to V _⊥ • δ _c , where V _⊥ is the transverse velocity of the ion after separation (V _⊥ << V _⊥П ). This condition is only necessary, but not sufficient. A sufficient condition is the absence of a positive potential on the walls of the channels. To eliminate the departure from the wall of ions with velocity components

it is necessary that the mobilities of electrons and ions in the transverse direction are approximately the same. Therefore, a longitudinal magnetic field is applied to the separation and condensation zone, which is necessary to increase productivity and uniformity

the thickness of the coatings on the other hand, and its value is determined from the inequality r _i > V _⊥ • τ _c >> r _e , where V _⊥ • τ _c = b is the transverse size of the channel; r _e , r _i - Larmor radius of the electron, ion, respectively; V _⊥P is the transverse velocity of the ion before separation;

- time of flight by the plasma of the separation zone.

In this case, the electrons are magnetized in the channels (r _e << V _⊥ • τ _c ) and their transverse mobility decreases (ω _e • τ _ei ) ² times and becomes comparable and even less than the mobility of the ions (ω _e is the Larmor frequency of electron rotation, τ _ei is the time between electron-ion collisions).

The magnitude of the magnetic field is such that the ions are not magnetized in the channels (r _i > V _⊥ • τ _c ) and their velocity component V _⊥ does not change. In this case, only that part of the ions leaves for the channel walls, which during the time τ _c, at a speed from V _⊥П to V _⊥, travels the distance between the channel walls; b = V _⊥ • τ _c and that part of the ions that does not have time to travel the distance b during the time τ _c remains in the plasma. Thus, ions leave the separation zone at a maximum angle to the axis equal to α ~ b / h _c radians. In addition, the topology of the magnetic field also plays an important role. Secondly, the parallelism of the lines of force of the magnetic field by the walls of the channels is required, and secondly, the perpendicularity of the lines of force of the condensation surface is required. The introduction of longitudinal channels into the separation zone determines the inhomogeneity of the plasma in the region of the projection of the channel walls on the condensation surface.

. В данном случае, указанная область заполняется плазмой за счет того, что поперечная компонента скорости иона после сепарации имеет конечную величину V_⊥≠ 0, но V_⊥<< V_∥П. Для практической реализации способа необходимы минимальные потери плазмы (ионной компоненты) в анодной области дугового разряда и в зоне транспортирования при гарантированном отсутствии загрязнения потока плазмы химическими примесями, наличие минимальных (оптимальных) энергозатрат и заданного уровня неоднородности толщины покрытий. На всех этапах получения потока плазмы потери плазмообразующего вещества определяются рядом факторов: топологией магнитного поля

, его величиной

а также величинами потенциала в анодной области (ΔU_а) и в области транспортирования (U_см.т), а также параметрами профиля поверхности, ограничивающей область транспортирования.To eliminate the inhomogeneity of the plasma, the condensation surface is installed at a distance from the separation zone equal to

. In this case, said region is filled with the plasma due to the fact that the transverse component of the velocity of the ion after separation has a finite value V _⊥ ≠ 0, but V _⊥ << V _∥P. For the practical implementation of the method, minimal losses of plasma (ion component) in the anode region of the arc discharge and in the transportation zone are required, with guaranteed absence of contamination of the plasma stream with chemical impurities, the presence of minimal (optimal) energy consumption and a given level of coating thickness inhomogeneity. At all stages of obtaining a plasma flow, the losses of the plasma-forming substance are determined by a number of factors: the topology of the magnetic field

its size

as well as the values of the potential in the anode region (ΔU _a ) and in the transportation region (U _cm ), as well as the parameters of the surface profile, limiting the transportation region.

в области фокусировки (оптической оси) относительно поверхности анода (S_а) и положением заслонки в момент прохождения лазерного излучения через диафрагму. Возможны два варианта ориентации плоскости поляризации относительно плоскости падения: взаимно перпендикулярны или параллельны. Наиболее оптимален второй вариант, так как большая часть излучения поглощается рабочим веществом. Аналогично возможны два варианта ориентации нормали

к торцевой рабочей поверхности в области фокусировки (оптической оси): нормаль

не пересекается с поверхностью анода, т.е. направлена к выходному торцу анода, и нормаль

пересекает поверхность анода. В первом варианте поток лазерной плазмы расширяется в направлении нормали к торцевой рабочей поверхности и истекает, рекомбинируя к выходному торцу анода за пределы генератора без возбуждения пробоя в электродном промежутке. Ситуация обратная, если нормаль

пересекает поверхность анода S_а, поэтому область диаграммы направленности лазерной плазмы является зоной формирования пробоя межэлектродного промежутка с минимальными потерями плазмы и энергозатратами. Однако подобная оптимизация поджига зависит от наличия магнитного поля и его топологии на этом этапе, так как требования к магнитному полю на этапах поджига и генерирования не согласуются. На этапе поджига необходимо, чтобы магнитный поток, пронизывающий торцевую рабочую поверхность в области оптической оси полностью замыкался на анод, а на этапе генерирования - частично. Следовательно, наиболее оптимальным вариантом является отсутствие при поджиге стороннего магнитного поля.In addition to the optical constants of the working substance (consumable cathode material), the energy consumption at the laser ignition stage is additionally determined by a number of factors: the ignition time (τ _l ), the orientation of the plane of polarization of the laser radiation relative to the plane of incidence, and the direction of the normal vector to the end working surface

in the focusing region (optical axis) relative to the surface of the anode (S _a ) and the position of the shutter at the moment of laser radiation passing through the diaphragm. There are two possible orientations of the plane of polarization relative to the plane of incidence: mutually perpendicular or parallel. The second option is most optimal, since most of the radiation is absorbed by the working substance. Similarly, two normal orientation options are possible.

to the end working surface in the focus area (optical axis): normal

does not intersect with the surface of the anode, i.e. directed to the output end of the anode, and the normal

crosses the surface of the anode. In the first embodiment, the laser plasma flow expands in the direction normal to the end working surface and expires by recombining to the output end of the anode outside the generator without causing breakdown in the electrode gap. The situation is reversed if normal

intersects the surface of the anode S _a , therefore, the area of the laser plasma radiation pattern is the zone of formation of the breakdown of the interelectrode gap with minimal plasma loss and energy consumption. However, such optimization of the ignition depends on the presence of the magnetic field and its topology at this stage, since the requirements for the magnetic field at the stages of ignition and generation are not consistent. At the ignition stage, it is necessary that the magnetic flux penetrating the end working surface in the region of the optical axis is completely closed to the anode, and partially at the generation stage. Therefore, the most optimal option is the absence of an external magnetic field during ignition.

магнитного поля распределяют между областью транспортирования

(смешения

) в слое l_ленгм и областью конденсации Ф_конд согласно выражению

за счет замыкания магнитных силовых линий на корпус плазмовода в слое l_ленгм и регулирования его величины, путем изменения диаметра диафрагмы (D_т.см.- l_ленгм) в области торцов корпусов плазмоводов и камеры смешения. Оптимальному распределению силовых линий магнитного поля на выходном торце плазмовода (камеры смешения) способствует расположение соленоида с тыльной стороны подложки. Последнее кроме повышения производительности и снижения энергозатрат за счет оптимизации величин R_FП и R_КСП обеспечивает необходимые условия роста однородности покрытий путем повышения однородности

магнитного поля в области конденсации. Производительность устройства (кроме топологий магнитного поля

его величины

а также величин ΔU_а, U_см.т) лимитирует еще ряд конструктивных факторов: соотношение t_п профиля на внутренней поверхности и величину слоя Ленгмюра (l_ленгм.), а также соотношение параметров емкостного накопителя С основных электродов и емкости С_п цепи смещения. Возможны два предельных режима течения потока плазмы вдоль профилированной поверхности корпуса, находящегося под потенциалом смещения (+U_см.т). Первый режим наблюдается, если шаг профиля t_п больше, чем величина слоя Ленгмюра в плазме: t_п>l_ленгм (фиг. 7). Второй режим наблюдается, если t_п < l_ленгм (фиг. 8). При условии, когда r_i>D_т, ионы могут взаимодействовать с непрофилированным электрическим полем потенциала смещения +U_см. (2-й режим) и только с профилированным электрическим полем потенциала смещения (1-й режим). Длина свободного пробега ионов λ_ii относительно ион-ионных столкновений больше, чем шаг профиля (λ_ii> t_п). Следовательно, при взаимодействии иона с электрическим полем профилированной поверхности соблюдаются законы геометрической оптики. Так как λ_ii> D_Т даже при двух и трех последовательных столкновений с барьером U_см.т, так как λ_ii> D_Т.Reducing the losses of the plasma-forming substance at the generation stage reduces to optimizing the value of the jump in the anode potential ΔU _{and the} arc discharge in a magnetic field at a level at which the losses of the plasma (ion component) are minimal. The last requirement optimizes both energy consumption at the generation stage and the level of chemical impurities by limiting the duration of the discharge, based on the condition: τ _G <τ _T.AN. where τ _G is the discharge duration, τ _T.AN is the formation time of the thermal anode instability, as well as the decrease in the role of regeneration of the transparency of the laser input window, therefore a negative and zero value ΔU _a (ΔU _a ≤ 0) is clearly unacceptable in the method, since the level plasma loss in the anode region is maximum. The application of a longitudinal magnetic field equidistant to the surface of the anode radically changes the nature of the current transfer in the anode region of the discharge. Depending on the magnitude of the magnetic field, as well as the magnitude of the interelectrode distance and the discharge current, two limiting options can be distinguished, due to the fact that in the solid-state plasma generator in the region of the end working surface there is an expansion region in which the concentration of the working substance varies from the concentration of the solid n = 10 ²³ cm ^-3 to a concentration of n ~ 5 • 10 ¹¹ cm ^-3 in the region of the output end of the anode. Two zones can be distinguished in the expansion area: the first for which ω _e τ _ei <1 with the length from the end working surface to the surface where the condition ω _e τ _ei = 1 is satisfied, and the second for which the condition ω _e τ _ei > 1 with the length from the plane where ω _e τ _ei = 1 to the output end of the anode. The current-receiving surface of the anode can be spatially located in the first or second zone. In the first zone, the magnetic field does not significantly affect the plasma, and the nature of its current transport is mainly determined by chaotic motion. Therefore, in this mode, there is no possibility of targeted regulation of both the sign and the value of the potential jump, i.e. the situation is similar to the absence of a magnetic field. In the second case, when the current-receiving surface of the anode is located outside the surface, where ω _e τ _ei = 1, the situation changes. The value (+ ΔU _a ) provides an effective potential barrier for the return (reflection) of the ion component to the plasma, but its value should be optimal and can be estimated by two criteria: by productivity and energy consumption ΔU _a • e≃ (2-3) кT _i where T _i is the ion temperature and the energy of formation of radiation defects W _def. in the substrate ΔU _a <W _def. In addition, the excess value ΔU _a sharply reduces the value of _{T. T. AN} To reduce the value of ΔU _a to the optimal value, it is necessary to change the current value I ₀ in the range between its extreme values: the smallest - I _{Фа = 0} and the largest - I _{н = 0} , which can be realized by purposeful change in the topology of the magnetic field by closing part of the magnetic field lines of the magnetic field of the solenoid to the anode region (anode). In this case, the number of additional electrons on the anode gone (Δn _e) and causing the current to rise to a value proportional to the magnitude of the magnetic flux F _A, closed at the anode region Δn _e = Const.F _A. Thus, the condition ω _e τ _ei = 1 (the magnetic field) defines a minimum radial dimensions of the anode and the conditions for optimizing the values of ΔU _and the ratio of magnetic fluxes (F _K, F _A, F _T) determines the longitudinal dimensions of the anode, which collectively optimize overall dimensions of the anode for maximum productivity, minimum energy consumption and introduced chemical impurities by limiting the duration of the discharge at the generation stage and reducing the role of regeneration of the transparency of the laser input window at the ignition stage at ΔU _a > 0. The optimization of the value of Ф _А can be carried out in two ways: by changing the gap l _АП between the anode solenoid and the screw solenoid. To reduce plasma losses at the transportation stage, it is necessary that the magnetic field lines penetrating the generation region and not closed to the anode region are directed to the transportation region equidistant to the surface (i.e., without shorting to the side surface) that bounds this region. Therefore the rest of the longitudinal flow

the magnetic field is distributed between the transport area

(mixing

) in the layer l _langm and the condensation region _{f cond} according to the expression

due to the closure of magnetic field lines to the plasma duct body in the l _langm layer and adjusting its value by changing the diameter of the diaphragm (D _t.sm. - l _langm ) in the region of the ends of the plasma duct housings and the mixing chamber. The optimal distribution of magnetic field lines at the output end of the plasma duct (mixing chamber) is facilitated by the location of the solenoid on the back of the substrate. The latter, in addition to increasing productivity and reducing energy consumption by optimizing the values of R _FP and R _KSP, provides the necessary conditions for the growth of coating uniformity by increasing uniformity

magnetic field in the field of condensation. Device performance (except for magnetic field topologies

its magnitude

as well _as values ΔU, U _sm.t) limits a number of design factors: the ratio of t _n profile on the inner surface layer and the amount of Langmuir (l _lengm), and the ratio of the capacitive storage parameters From the main electrodes and the capacitance C _n bias _circuit.. Two limiting regimes of the plasma flow along the profiled surface of the body under the bias potential (+ U _{cm t} ) are _possible . The first mode is observed if the profile step t _{p is} greater than the magnitude of the Langmuir layer in the plasma: t _p > l _Langm (Fig. 7). The second mode is observed if t _p <l _Langm (Fig. 8). Provided that r _i > D _t , the ions can interact with the unshaped electric field of the bias potential + U _see (2nd mode) and only with the profiled electric field of the bias potential (1st mode). The mean free path of ions λ _ii with respect to ion-ion collisions is greater than the profile step (λ _ii > t _p ). Consequently, the laws of geometric optics are observed when the ion interacts with the electric field of the profiled surface. Since λ _ii > D _T even in two and three successive collisions with a barrier U _{see t} , since λ _ii > D _T.

на расстоянии, меньшем чем 3t_п области профилированного потенциала (фиг. 2, 3), и, во-вторых, уход ионов на стенку, т.к. уменьшается величина потенциального барьера, отражающего ионы.Profile parameters (height h _p , step t _p , angle at the apex α _P ) are chosen so as to ensure the locking of the droplet phase when the latter collides with the threaded profile. Thus, in the first plasma flow regime, the profiled electric field will change the direction of the ion pulse to “lock” it on the threaded profile and, therefore, reduce productivity, since ions after interaction with the profile will be reflected towards the end working surface of the cathode. In the second mode (t _p <l _Langm ), the electric field is _unshaped at a distance l _E from the surface profile equal to l _E ≃ 3t _P (Fig. 7). Thus, in order to prevent plasma ions from “feeling” the electric field profile, it is necessary that, at a distance from the profiled surface, l _Е ≃ 3t _{П, the} potential of the electric displacement field is equal to U _{cm (at lе = 3tп)} = Т _i . In this case, the ions interact with a non-profiled electric displacement field in the absence of locking of the ion component, which leads to an increase in the productivity of the device in the presence of locking of the droplet phase on the threaded profile of the body. However, the optimal ratio between the step t _{p of the} profile and the size of the layer l _Langm are not sufficient conditions for increasing the productivity of the plasma generator. The latter is due to the fact that the “magnetized” electronic component (r _e << D _t , ω _e τ _ei > 1) goes into the plasma duct body during the plasma flow, partially discharging the capacitance C _{n of} the bias circuit and also decreasing the value of the optimal bias potential. A decrease in U _{cm t} causes, firstly, the location of the equipotential

at a distance less than 3t _{p of the} region of the profiled potential (Fig. 2, 3), and, secondly, the departure of ions to the wall, because the potential barrier reflecting ions decreases.

, где R_КП-сопротивление участка корпус - плазма, Oм; L_г- индуктивность цепи разряда, Гн. В противном случае, если

, емкость C_п успевает разрядиться за время транспортирования до неоптимальной величины смещения (U_см.опт≃ +20B). Неравенство

кроме емкости С_п определяется еще величиной R_кп, которая должна иметь максимально возможное значение. Последнее оптимизируется распределением магнитного поля Ф_Т(см) между областью транспортирования

(смешения

) в слое Ленгмюра и областью конденсации Ф_конд.
Данный способ иллюстрируется следующими примерами.Thus, the potential value U _cm during the plasma transportation time should be in the region of the optimal value, i.e. should be almost constant. The latter is observed if

where R _{KP is the} resistance of the body section - plasma, Ohm; L _g - inductance of the discharge circuit, GN. Otherwise, if

, the capacitance C _p manages to be discharged during transportation to a non-optimal amount of displacement (U _{see opt} ≃ + 20B). Inequality

in addition to the capacitance C _p is also determined by the value of R _Kp , which should have the highest possible value. The latter is optimized by the distribution of the magnetic field Ф _Т (cm) between the transport region

(mixing

) in the Langmuir layer and the condensation region F _cond.
This method is illustrated by the following examples.

где

- концентрации химических элементов Y, Ba, Cu в многокомпонентном покрытии; K_Y, K_Ba, K_Cu - коэффициенты эрозии плазмообразующих веществ Y, Ba, Cu, [г/Кл]; t_Y, t_Ba, t_Cu - коэффициенты транспортирования потоков плазмы Y, Ba, Cu;

и U_Y, U_Ba, U_Cu - величины емкостей (Ф) и напряжения накопителей (В) соответственно. Концентрация O_7-x кислорода газообразной компоненты задавалась импульсным напуском реактивного газа. При этом генерируемая последовательность импульсов плазмы твердофазных элементов Y, Ba, Cu задавалась исходя из порядка чередования подрешеток формируемой кристаллической структуры (1, 2, 3) элементарной ячейки сверхпроводника (см. фиг. 23, 24). А число импульсов плазмы каждой подрешетки n_Y, n_Ba, n_Cu определялось из равенства

где ρ_Y(Ba)(Cu) - плотность материала мишени (Y, Ba, Cu соответственно) кг/м³; S - площадь поперечного сечения зоны смешения, м²; d_{Y,(Ba) (Cu)} - размер соответствующей подрешетки в направлении роста, м; K_Y(Ba)(Cu) - коэффициент эрозии в вакуумной дуге вещества подрешетки (Y, Ba, Cu) соответственно, г/К;

- величина емкости накопителя вакуумной дуги Y, Ba, Cu; U_{Y,(Ba) (Cu)} - величина изменения напряжения накопителя вакуумной дуги на Y, Ba, Cu соответственно. При этом напуск (газообразного элемента) кислорода осуществлялся импульсно исходя из его положения в кристаллической структуре и вводился в область генерирования потоков плазмы твердофазных веществ Ba, Cu путем регулирования величины напряжения

- на импульсном натекателе. Причем генерирование плазмы Y, Ba и Cu осуществлялось за времена τ_г, определяемые из условия τ_Г< τ_ТАН.Г, где τ_ТАН.Г - время развития тепловых анодных неустойчивостей на этапе генерирования. Напуск газообразной компоненты O₂ осуществлялся в течение времени, равного τ_Г. Затем осуществлялось транспортирование потоков Y, Ba, Cu, содержащих ионную, нейтральную и капельную фазу вдоль профилированной поверхности к области смешения. Транспортирование реализовывалось в радиальном электрическом поле Е_т потенциала смещения U_см.т = 20 В и скрещенном с ним продольном магнитном поле Н_т криволинейной плазмооптической системы при выполнении условий плазмооптики: r_i>D_т, r_e<<D_т, ω_e•τ_ei> 1, где r_e, r_i - ларморовские радиусы ионов Y, Ba, Cu и электрона соответственно, D_т - поперечный размер зоны транспортирования, ω_eτ_ei - параметр Холла электронов, ω_e - циклотронная частота электронов, τ_ei - время электрон-ионных столкновений. Плазмооптическая система была выполнена в виде трехзаходных винтовых цилиндров диаметром D_т. Профиль внутренней поверхности корпуса плазмоводов был выполнен треугольным, причем высота профиля h_п = 2 мм и его шаг t_п = 1,5 мм выбраны из неравенств; h_п > d_к, l_ленгм>t_п>d_к, а угол α_П при вершине треугольного профиля был выбран равным α_П= 45^°. Сепарация ионной компоненты потоков плазмы Y, Ba и Cu от капельной фазы в винтовых плазмоводах зоны транспортирования осуществлялась в области слоя Ленгмюра (l_ленгм) потенциала смещения (U_см.т) путем многократных
столкновений капельной фазы с профилированной поверхностью при выполнении условия P•N_к<1 в каждом плазмоводе, где Р - вероятность пролета капли в области транспортирования (Р = 10^-4), N_к - число капель, генерируемых за импульс в вакуумных дугах на Y, Ba, Cu.EXAMPLE N 1. The coating was carried out Y ₁ Ba ₂ Cu ₃ O _7-x in vacuum on an ultrahigh-vacuum installation of pulse-plasma coating. The process chamber and the three-channel device were pumped out by a cryogenic pump to a pressure of 10 ^-7 Pa. Then, three spatially localized vacuum arcs were excited by breaking the interelectrode gaps in three plasma generators with consumable cathodes from Y, Ba, and Cu. The vacuum arc was excited by defocused laser radiation with a duration of τ _l = 10 ns under the condition τ _l << τ _g , and the resulting laser plasma flow was directed to the anode region of the discharge, then the focus area of the laser measurement was blocked by forming a potential jump in the anode region of the discharge ( ΔU _a ) or with a butterfly valve. In this case, the excitation of vacuum arcs in barium and copper vapors was carried out synchronously, and the excitation of a vacuum arc in yttrium vapors was carried out asynchronously based on the arrangement of the sublattices of the formed crystalline unit cell of the layered substance of the coating Y ₁ Ba ₂ Cu ₃ O _7-x . The number of such cycles (m) of vacuum excitation sequences was determined from the equality d _i • m≤l, where d _i is the unit cell size in the direction of coating growth, l is the coating thickness. Then, plasma flows of Y, Ba, and Cu were generated, and the chemical composition of the multicomponent coating of the solid-phase components of Y ₁ Ba ₂ Cu ₃ O _7-x was set from the condition

Where

- concentration of chemical elements Y, Ba, Cu in a multicomponent coating; K _Y , K _Ba , K _Cu - erosion coefficients of plasma-forming substances Y, Ba, Cu, [g / C]; t _Y , t _Ba , t _Cu - transport coefficients of plasma flows Y, Ba, Cu;

and U _Y , U _Ba , U _Cu are the capacitance values (Ф) and storage voltage (V), respectively. The concentration of O _7-x oxygen of the gaseous component was set by pulsed inlet of the reactive gas. In this case, the generated pulse sequence of the plasma of the solid-phase elements Y, Ba, Cu was set based on the alternation order of the sublattices of the formed crystal structure (1, 2, 3) of the unit cell of the superconductor (see Fig. 23, 24). And the number of plasma pulses of each sublattice n _Y , n _Ba , n _{Cu was} determined from the equality

where ρ _{Y (Ba) (Cu)} is the density of the target material (Y, Ba, Cu, respectively) kg / m ³ ; S is the cross-sectional area of the mixing zone, m ² ; d _{Y, (Ba) (Cu)} is the size of the corresponding sublattice in the growth direction, m; K _{Y (Ba) (Cu)} - erosion coefficient in the vacuum arc of the sublattice substance (Y, Ba, Cu), respectively, g / K;

- the value of the storage capacity of the vacuum arc Y, Ba, Cu; U _{Y, (Ba) (Cu)} is the magnitude of the change in the voltage of the vacuum arc accumulator by Y, Ba, Cu, respectively. In this case, the oxygen (gaseous element) was puffed in based on its position in the crystal structure and was introduced into the region of the generation of plasma flows of solid-phase substances Ba, Cu by adjusting the voltage

- on impulse leakage. Moreover, the generation of Y, Ba, and Cu plasma was carried out for times τ _g determined from the condition τ _G <τ _TAN.G , where τ _TAN.G is the time of development of thermal anode instabilities at the generation stage. The inlet of the gaseous component O _{2 was} carried out for a time equal to τ _G. Then, Y, Ba, Cu flows containing the ionic, neutral, and droplet phases were transported along the profiled surface to the mixing region. Transportation was carried out in a radial electric field E _{t of} displacement potential U _{cm t} = 20 V and a longitudinal magnetic field N _{t of a} curvilinear plasma-optical system crossed with it under the conditions of plasma optics: r _i > D _t , r _e << D _t , ω _e • τ _ei > 1, where r _e , r _i are the Larmor radii of Y, Ba, Cu and electron ions, respectively, D _t is the transverse size of the transport zone, ω _e τ _ei is the electron Hall parameter, ω _e is the electron cyclotron frequency, τ _ei - time of electron-ion collisions. The plasma-optic system was made in the form of three-way screw cylinders with a diameter of D _t . The profile of the inner surface of the plasma duct housing was triangular, with the profile height h _p = 2 mm and its pitch t _p = 1.5 mm selected from inequalities; h _p > d _k , l _Langm > t _p > d _k , and the angle α _P at the apex of the triangular profile was chosen equal to α _P = 45 ^° . Separation of the ionic component of the plasma flows of Y, Ba, and Cu from the droplet phase in the screw plasma ducts of the transport zone was carried out in the region of the Langmuir layer (l _Langm ) of the bias potential (U _cm ) by multiple
collisions of the droplet phase with the profiled surface under the condition P • N _k <1 in each plasma duct, where P is the probability of droplet passage in the transport region (P = 10 ^-4 ), N _k is the number of droplets generated per pulse in vacuum arcs on Y , Ba, Cu.

причем продольное

реализовывалось путем совмещения магнитных полей

областей транспортирования при выполнении условий r_i>D_см, λ_ii> D_СМ, r_e << D_см, ω_eτ_ei>> 1, где r_i, r_e - ларморовские радиусы ионов и электрона соответственно, D_см - поперечный размер области смешения, ω_eτ_ei - параметр Холла электронов, ω_e - циклотронная частота электрона, τ_ei - частота электрон-ионных столкновений. Перемешивание ионов различной химической природы осуществлялось за счет многократного столкновения ионов с потенциальным барьером U_см при условии λ_ii> D_СМ, где λ_ii - длина ион-ионных столкновений (λ_ii= 10²СМ) , D_см = 15 см. Химический состав соединения устанавливался на расстоянии, равном L_П∥= V_∥СМ•τ_СМ.ПЛ, где τ_СМ.ПЛ - время радиальной релаксации химического состава в общем потоке плазмы. При этом смешение плазм Ba и Cu осуществлялось за время τ_СМ, определяемое из условия τ_СМ< τ_ТАН.СМ, где τ_ТАН.СМ - время развития тепловых анодных неустойчивостей на этапе смешения. Процесс конденсации осуществлялся в зоне конденсации смешиваемых потоков. При этом подложку устанавливали на расстоянии L_П∥= V_∥СМ•τ_СМ, где τ_СМ - время радиальной релаксации химического состава в общем потоке плазмы. Величина потенциала смещения подложки U_см.п = 5 Т_e выбиралась из условия eU_см.т+eU_см.п<W_деф., а неоднородность магнитного поля

выбиралась из соотношения

,
где Δl - разброс толщины покрытия, l - толщина покрытия. Последнее осуществлялось наложением с тыльной стороны подложки магнитного поля (3-х соленоидов диаметром D_см см и длиной, равной 20 см). При реализации способа получения покрытий осуществлялись измерения параметров покрытий в зависимости от возможных изменений параметров способа. Так, лазерный поджиг в способе не удалось реализовать при τ_Л< τ_Г (τ_Л= 10 Н•С, τ_Г= 500 МК•С) и максимально возможной энергии накачки (50 Дж) лазера ЛТИ - 205, когда нормаль к торцевой рабочей поверхности катодов (Y, Ba, Cu) в зоне фокусировки лазерного излучения не пересекалась с поверхностью анода. Устранение указанного недостатка (ориентации нормали) позволили реализовать поджиг при энергии накачки 20 Дж. Исследовалась возможность применения лазера в режиме свободной генерации (τ_Л≃ τ_Г). В данном случае происходила привязка катодных пятен к области фокусировки лазерного излучения на катоде и оплавление этой зоны, т. е. возникновение большого объема жидкой фазы и, следовательно, возгонка легкоплавких компонент сплавов. Кроме того, резко возросли энергозатраты. Поджиг отсутствовал, если первоначально на зону разряда накладывалось магнитное поле, т. к. последнее изменяет ориентацию оптимальной диаграммы истечения лазерной плазмы. Следовательно, магнитное поле должно накладываться после этапа поджига. Химический состав и энергозатраты в способе зависят от распределения магнитного поля в анодной области разряда. Так, при отсутствии магнитного потока на анодную область Ф_А=0 (магнитная изоляция анода) результаты измерений показали резкий рост потенциала на электродах: U_АК≃ U_К+ ΔU_А≃ 100 B и образование анодных пятен на поверхности анода, т.е. рост уровня привнесенных химических примесей и рост энергозатрат из-за резкого снижения величины τ_ТАН. Изменение величины Ф_А осуществлялось регулированием расстояния между соленоидом анода и соленоидом плазмовода l_АП. При величине l_АП = 5 см потенциал на электродах равен U_АК=U_К+ΔU_А=50 В при отсутствии анодных пятен. Однако микроскопические исследования показали локальные нарушения состава покрытия в области конденсации капельной фазы. Для достижения цели потоки плазмы транспортировались в скрещенных

полях винтовых плазмоводов при различных параметрах способа. Так, при реализации условий l_ленгм<t_п, d_к<T_п (d_к - диаметр макрочастицы, t_п - шаг профиля) происходит снижение производительности при наличии капельной фазы. Капельная фаза присутствует и при реализации условий l_ленгм >t_п, d_к >t_п; аналогичные результаты имеют место и при d_к > h_п, d_к > l_ленгм, что обусловлено отсутствием совокупности условий запирания капельной фазы на профиле винтового корпуса плазмовода. Поэтому результаты измерений оптического рассеяния и анализ поверхности покрытия под микроскопом показали значительную величину рассеяния, равную 0,3, а также дефектность покрытия. Запирание капельной фазы удалось реализовать на этапе транспортирования потоков плазмы в винтовых плазмоводах при выполнении условий P•N_к<1, d_к < t_п< l_ленгм. Последнее подтвердили измерения оптического рассеяния и анализ поверхности покрытия под микроскопом. Так величина рассеяния оказалась равной 0,07%. Однако запирание капельной фазы является необходимым условием реализации способа, но недостаточным. Поэтому осуществлялось определение влияния режима синхронизации на химический состав покрытия. Для этого в первой серии экспериментов покрытие Y₁Ba₂Cu₃O_7-x формировалось путем синхронного возбуждения трех пространственно разделенных вакуумных дуг в среде О₂ при давлении ~10^-1 Па на мишенях из Ba, Cu, Y и последующей сепарации потоков плазмы от капельной фазы их смешения и конденсации. Во второй серии экспериментов, в отличие от первой, задавалась последовательность импульсов плазмы и кислорода, исходя из порядка чередования подрешеток формируемой кристаллической структуры (1, 2, 3) элементарной ячейки сверхпроводника (см. фиг. 23, 24), а число импульсов плазмы каждой подрешетки n_A(n_B)(n_C) определялось из равенства
q_Y,Ba,Cu = C_Y,Ba,Cu•U_Y,Ba,Cu;

где ρ_Y,Ba,Cu - плотность наносимого материала в покрытии,
S_т - площадь поперечного сечения области смешения (Y, Ba, Cu) м², d_Y,Ba,Cu - толщина подрешетки Y, Ba, Cu в элементарной ячейке, м, q_Y,Ba,Cu - заряд, протекающий в цепи разряда.In this case, the transport of plasma Y, Ba, Cu was carried out for a time τ _T , determined from the condition τ _T <τ _TAN.T , where τ _TAN.T is the development time of thermal anode instabilities at the stage of transportation. After separation of the plasma flows of Y, Ba, Cu from the droplet phase, plasma-optical mixing of the plasma flows of Ba, Cu in the crossed

and longitudinal

realized by combining magnetic fields

transportation regions under the conditions r _i > D _cm , λ _ii > D _CM , r _e << D _cm , ω _e τ _ei >> 1, where r _i , r _e are the Larmor radii of ions and electrons, respectively, D _cm is transverse the size of the mixing region, ω _e τ _ei is the Hall parameter of electrons, ω _e is the cyclotron frequency of the electron, τ _ei is the frequency of electron-ion collisions. The mixing of ions of different chemical nature was carried out due to multiple collisions of ions with a potential barrier of U _cm under the condition λ _ii > D _CM , where λ _ii is the length of ion-ion collisions (λ _ii = 10 ² CM), D _cm = 15 cm. Chemical composition of the compound was established at a distance equal to L _П∥ = V _∥СМ • τ _SM.PL , where τ _SM.PL is the time of radial relaxation of the chemical composition in the total plasma stream. In this case, the mixing of Ba and Cu plasmas was carried out during the time τ _SM , determined from the condition τ _SM <τ _TAN.CM , where τ _TAN.CM is the development time of thermal anode instabilities at the mixing stage. The condensation process was carried out in the condensation zone of the mixed flows. In this case, the substrate was installed at a distance L _П∥ = V _∥СМ • τ _СМ , where τ _СМ is the radial relaxation time of the chemical composition in the total plasma flow. The value of the substrate displacement potential U _cm.n. = 5 T _{e was} chosen from the condition eU _cm.t + eU _cmn.n <W _def. , and the inhomogeneity of the magnetic field

was selected from the relation

,
where Δl is the variation in coating thickness, l is the coating thickness. The latter was carried out by applying a magnetic field (3 solenoids with a diameter of D _cm cm and a length of 20 cm) from the back side of the substrate. When implementing the method for producing coatings, measurements of the parameters of the coatings were carried out depending on possible changes in the parameters of the method. Thus, the laser ignition in the method could not be realized at τ _L <τ _G (τ _L = 10 N • C, τ _G = 500 MK • C) and the maximum possible pump energy (50 J) of the LTI laser - 205, when the normal to the end the working surface of the cathodes (Y, Ba, Cu) in the focus area of the laser radiation did not intersect with the surface of the anode. The elimination of this drawback (normal orientation) made it possible to realize ignition at a pump energy of 20 J. The possibility of using a laser in the free-running mode (τ _L ≃ τ _G ) was studied. In this case, the cathode spots were attached to the focus area of the laser radiation at the cathode and this zone was melted, i.e., the appearance of a large volume of the liquid phase and, therefore, the sublimation of low-melting alloy components. In addition, energy costs rose sharply. Ignition was absent if initially a magnetic field was superimposed on the discharge zone, since the latter changes the orientation of the optimal laser plasma flow diagram. Therefore, the magnetic field must be superimposed after the ignition step. The chemical composition and energy consumption in the method depend on the distribution of the magnetic field in the anode region of the discharge. So, in the absence of magnetic flux to the anode region Ф _А = 0 (magnetic insulation of the anode), the measurement results showed a sharp increase in potential on the electrodes: U _AK ≃ U _K + ΔU _A ≃ 100 B and the formation of anode spots on the surface of the anode, i.e. an increase in the level of introduced chemical impurities and an increase in energy consumption due to a sharp decrease in τ _TAN . The change in the value of Ф _{А was} carried out by controlling the distance between the anode solenoid and the plasma duct solenoid l _АП . When the value of l _AP = 5 cm, the potential at the electrodes is equal to U _AK = U _K + ΔU _A = 50 V in the absence of anode spots. However, microscopic studies showed local violations of the coating composition in the field of condensation of the droplet phase. To achieve the goal, plasma flows were transported in crossed

fields of screw plasma ducts for various method parameters. So, when conditions l _langm <t _p , d _k <T _p (d _k is the diameter of the macroparticle, t _p is the profile step) are realized, productivity decreases in the presence of a droplet phase. The droplet phase is also present when conditions l _langm > t _p , d _k > t _p ; Similar results also occur when d _k > h _p , d _k > l _Langm , which is due to the lack of a set of conditions for locking the droplet phase on the profile of the screw housing of the plasma duct. Therefore, the results of optical scattering measurements and the analysis of the coating surface under a microscope showed a significant scattering value of 0.3, as well as a defective coating. It was possible to realize the locking of the droplet phase at the stage of transporting plasma flows in screw plasma ducts under the conditions P • N _k <1, d _k <t _p <l _Langm . The latter was confirmed by measuring optical scattering and analyzing the surface of the coating under a microscope. So the scattering value turned out to be equal to 0.07%. However, locking the droplet phase is a necessary condition for the implementation of the method, but insufficient. Therefore, the influence of the synchronization mode on the chemical composition of the coating was determined. For this, in the first series of experiments, the coating Y ₁ Ba ₂ Cu ₃ O _{7-x was} formed by synchronously exciting three spatially separated vacuum arcs in an O ₂ medium at a pressure of ~ 10 ^-1 Pa on targets from Ba, Cu, Y and subsequent separation of plasma flows from the drop phase of their mixing and condensation. In the second series of experiments, unlike the first, the sequence of plasma and oxygen pulses was set based on the alternation order of the sublattices of the formed crystal structure (1, 2, 3) of the unit cell of the superconductor (see Fig. 23, 24), and the number of plasma pulses of each sublattices n _A (n _B ) (n _C ) was determined from the equality
q _{Y, Ba, Cu} = C _{Y, Ba, Cu} • U _{Y, Ba, Cu} ;

where ρ _{Y, Ba, Cu} is the density of the applied material in the coating,
S _t is the cross-sectional area of the mixing region (Y, Ba, Cu) m ² , d _{Y, Ba, Cu} is the thickness of the Y, Ba, Cu sublattice in the unit cell, m, q _{Y, Ba, Cu} is the charge flowing in the circuit discharge.

The calculated and experimental data showed that n _Y = 1 imp., N _Ba = 2 imp. , n _Cu = 9 imp. The main distinctive feature of the coatings obtained in the second series is that they acquired superconducting properties even at 450 ^o C in oxygen (4 h) and had almost the same sharp transition T _S ---> T _C. Thus, the combination of synchronous and asynchronous excitation of vacuum arcs and O ₂ flows allows you to potentially set the required chemical composition of the compound.

However, setting the synchronization mode for Y, Ba, Cu, and O ₂ is only a necessary condition for obtaining a multicomponent coating. The mixing mode itself is important. According to the criterion of the relationship between λ _ii and D _cm , where λ _ii is the length of ion-ion collisions ((λ _ii = 10 ² CM) = 10 ² cm) D _cm is the transverse size of the mixing zone, two modes are possible: λ _ii <D _CM and λ _ii > D _See In the first case, mixing is difficult due to collisions of ions with each other and occurs due to the large number of collisions n _st equal to n _ST • λ _ii ≃ D _CM . The situation is reversed if λ _ii > D _CM . In this case, the mutual collisions of ions do not limit mixing and are realized due to collisions of ions with a potential barrier of U _cm . However, the position of the magnetic field can violate the mixing conditions. Two variants were studied, r _i <D _cm and r _i ,> D _cm , where r _i are the Larmor radii of ions, D _cm is the transverse size of the mixing region. When r _i <D _cm there is a sharp deterioration in the mixing conditions, because the ionic component is magnetized and the actual mixing of ions of various chemical nature is due to ion-ion collisions and subsequent ion jumps to a distance r _i . In addition, secondly, there is no separation of ions by energy and, thirdly, there is no separation of ions by transverse momenta with a decrease in the productivity of the method due to the formation of a magnetic plug. The disadvantages are eliminated if r _i > D _cm , while the conditions for the magnitude of the magnetic field coincide with the conditions for the magnitude of the magnetic field in the transportation zone. The influence of the topology of the magnetic field is realized through productivity and energy consumption.

, корпус камеры смешения

формируется соответственно скачок потенциала на аноде (ΔU_а), а также величины потенциалов смещения на корпусах плазмоводов U_см.т и камеры смешения U_см, что в свою очередь определяет постоянную времени формирования тепловой анодной неустойчивости при генерировании, транспортировании и смешении. Возможны два варианта:

.The composition of the superconductor includes the initial solid-phase components Y, Ba, Cu and the gaseous component O ₂ . The composition of solid-phase components at the generation stage is determined by the magnitude of changes in the operating voltage and capacity of the vacuum arc storage devices, as well as by the synchronization mode. The gaseous component was introduced, as indicated, pulsed and synchronously according to its position in the unit cell. Three variants of spatial injection of O _{2 were} investigated: into the transport and generation condensation region, while the O ₂ flux was set by the magnitude of the voltage amplitude applied to the pulse leakage. The most optimal from the point of view of coating parameters was injection into the generation region (vacuum arcs of Ba and Cu). The predetermined chemical composition of the coating by setting the capacitance C _n and voltage U across the capacitances in each of the excitation, generation, transportation channels, as well as the synchronization mode can be violated due to the anode processes at the indicated stages. For a given value of the magnetic flux to the anode (Ф _А ), the plasma duct body

mixing chamber body

correspondingly, a potential jump is formed on the anode (ΔU _a ), as well as bias potentials on the plasma duct housings U _cm and mixing chamber U _cm , which in turn determines the time constant for the formation of thermal anode instability during generation, transportation, and mixing. Two options are possible:

.

магнитного поля в объеме плазмы на диаметре подложки D_см-l_ленгм, где l_ленгм - величина слоя Ленгмюра, которая определяется условиями запирания капельной фазы. Величина

в области конденсации (подложки) определяется в свою очередь положением соленоидов областей генерирования, транспортирования, смешения, а также наличием соленоида с тыльной стороны подложки. В связи с этим исследовались два варианта:

В первом варианте неоднородность

толщины покрытия составила величину, равную ~ 10%, неприемлемую для практики. Во втором варианте (с тыльной стороны подложки располагаются три соленоида диаметром D_см и длиной 2 D_см) неоднородность

толщины покрытия составила величину, равную ~3%, практически приемлемую для технологического оборудования. Указанное пространственное положение атомов в подрешетке может быть нарушено за счет избыточной энергии конденсируемых ионов плазмы. Исследовались два варианта: первый - eU_см.т+eU_см.г•Z>W_деф и второй - eU_см.т.+eU_см.г•Z < W_деф Сравнение результатов по плотности поверхностных состояний показало, что она в первом случае составляет величину, равную 10¹² -10¹³ 1/см²•эв, неприемлемую для практического использования тонкопленочной структуры. Во-втором случае уровень плотности поверхностных состояний равен ~10¹¹ 1/см²•эв, а тонкопленочная структура приемлема для практического использования. Однако кроме нарушений на границе подложка - покрытие возможны нарушения исходного пространственного положения атомов за счет высокоэнергетичных ионов лазерной плазмы и плазмы вакуумной дуги. Так, энергия ионов лазерной плазмы при плотности энергии при поджиге 10⁹ Вт/см может достигать несколько кэВ.In the latter case, the anode, the plasma duct body, and the mixing chamber are sources of introduced chemical impurities. So, at τ _Г = 10 ^-2 s> τ _TAN.G chemical analysis showed the presence of anode materials, plasma duct bodies and mixing chambers (stainless steel) in the coating thickness (Fe and Cr). The disadvantage was eliminated with a duration of τ _G <1 μs. In this case, the attachment (anode spots) to the anode, plasma duct housings, and mixing chambers did not have time to form, and the chemical composition of the coating was determined by the C _n and U parameters of the vacuum arcs and the mode of their synchronization. The chemical composition of the coating, in addition to synchronizing the mass values of the components (Y, Ba, Cu) through the parameters C _n and U of the vacuum arcs, the presence of the droplet phase, is also due to the homogeneity of the distribution of components in the coating, which in turn is determined by the inhomogeneity

the magnetic field in the plasma volume on the diameter of the substrate D _cm -l _Langm , where l _Langm is the Langmuir layer, which is determined by the conditions for locking the droplet phase. Value

in the field of condensation (substrate) is determined in turn by the position of the solenoids of the areas of generation, transportation, mixing, as well as the presence of a solenoid on the back of the substrate. In this regard, two options were investigated:

In the first embodiment, heterogeneity

coating thickness was a value of ~ 10%, unacceptable for practice. In the second variant (three solenoids with a diameter of D _cm and a length of 2 D _cm are located on the back side of the substrate), the inhomogeneity

the coating thickness was ~ 3%, which is practically acceptable for technological equipment. The indicated spatial position of atoms in the sublattice can be violated due to the excess energy of the condensed plasma ions. Two options were investigated: the first - eU _{cm t} + eU _{cm g} • Z> W _def and the second - eU _{cm t} + eU _{cm g} • Z <W _def A comparison of the results on the density of surface states showed that it is in the first case amounts to 10 ¹² -10 ¹³ 1 / cm ² • eV, unacceptable for the practical use of thin-film structure. In the second case, the level of density of surface states is ~ 10 ¹¹ 1 / cm ² • eV, and the thin-film structure is acceptable for practical use. However, in addition to disturbances at the substrate – coating interface, violations of the initial spatial position of atoms due to high-energy ions of the laser plasma and plasma of the vacuum arc are possible. Thus, the energy of laser plasma ions at an energy density of 10 ⁹ W / cm when fired can reach several keV.

и электрического

полей на этапе транспортирования и смешения определяли из условия τ_ЕН≤ τ_Г+τ_ПРОЛ , где τ_ПРОЛ - время пролета плазмой области транспортирования и смешения. При реализации способа получения покрытий осуществлялось измерение параметров покрытий в зависимости от возможных изменений параметров способа.EXAMPLE N 2. The same as in example N 1, while the generation of plasma from a plasma-forming substance of complex composition (Ba-Cu alloy) is carried out for a time τ _G , the value of which is chosen less than the time τ _{II of the} life of cathode spots of the second kind: τ _D <τ _II ≃ 1mks and the duration T _EN application of a magnetic

and electric

of the fields at the stage of transportation and mixing was determined from the condition τ _EH ≤ τ _G + τ _PROL , where τ _PROL is the time of flight of the transport and mixing areas by the plasma. When implementing the method for producing coatings, measurements of the parameters of the coatings were carried out depending on possible changes in the parameters of the method.

в плазмоводах и камере смешения создаются токами вакуумных дуг, т. е. при τ_ЕН= τ_Г. Химический анализ покрытия Ba-Cu показал, что его состав нарушен (недостаток Cu) на величину ~1%. Причем по мере увеличения длительности разряда τ_Г величина нарушения состава увеличивалась, т. е. когда τ_ЕН= τ_Г> τ_II. В этом случае нарушение состава обусловлено как фракционной возгонкой в катодной области разряда и последующего продольного разделения состава в зонах транспортирования и смешения. Нарушение состава было предотвращено при длительности наложения магнитного поля Н_т H_см), равной τ_ЕН= τ_Г+τ_ПРОЛ, где τ_Г выбиралась из условия τ_Г< τ_II≃ 1мкс.Usually magnetic fields

in the plasma ducts and the mixing chamber are created by currents of vacuum arcs, i.e., at τ _ЕН = τ _Г. Chemical analysis of the Ba-Cu coating showed that its composition was violated (Cu deficiency) by ~ 1%. Moreover, with an increase in the duration of the discharge τ _{G, the} value of the violation of the composition increased, i.e., when τ _ЕН = τ _Г > τ _II . In this case, the composition violation is caused by both fractional sublimation in the cathode region of the discharge and subsequent longitudinal separation of the composition in the transportation and mixing zones. Violation of the composition was prevented when the duration of the magnetic field N _t H _cm ) was equal to τ _EN = τ _G + τ _PROL , where τ _{G was} chosen from the condition τ _G <τ _II ≃ 1 μs.

, V_∥П и V_⊥ - величина продольной и поперечной скорости иона в потоке плазмы после сепарации по импульсам соответственно, l₀ - поперечный размер стенок между каналами. Величина магнитного поля определялась из неравенства
r_i> V_⊥•τ_c>> r_e ,
где V_⊥•τ_c= b = (0,5-1) см - поперечный размер канала, r_e = 10^-2 см, ларморовский радиус электрона, r_i = 10² см - ларморовский радиус иона,

время пролета плазмой зоны сепарации (каналов). При реализации способа получения покрытий осуществлялось измерение параметров покрытий в зависимости от возможных изменений параметров способа. Так, результаты взрывной литографии показали, что нарушение вышеизложенных условий обусловливает наличие подпыла на вертикальные стенки профиля и отсутствие формирования рисунка.EXAMPLE N 3. The same as in example N 2, 1, but the substrate was profiled, while after mixing the plasma flows the ions are separated by transverse momenta by transporting in longitudinal channels, the length of the channels and the distance L _to the condensation surface was determined from expression
V _∥ • τ _ii <h _С + L _К ,
Where

, V _{∥ П} and V _⊥ are the longitudinal and transverse velocity of the ion in the plasma stream after separation by pulses, respectively, l ₀ is the transverse size of the walls between the channels. The magnitude of the magnetic field was determined from the inequality
r _i > V _⊥ • τ _c >> r _e ,
where V _⊥ • τ _c = b = (0.5-1) cm is the transverse size of the channel, r _e = 10 ^-2 cm, the Larmor radius of the electron, r _i = 10 ² cm is the Larmor radius of the ion,

plasma transit time of the separation zone (channels). When implementing the method for producing coatings, measurements of the parameters of the coatings were carried out depending on possible changes in the parameters of the method. So, the results of explosive lithography showed that violation of the above conditions causes the presence of dust on the vertical walls of the profile and the absence of the formation of a pattern.

магнитного поля распределяют между областью транспортирования

(смешения

) в слое Ленгмюра (l_ленгм) и областью конденсации (Ф_к) согласно выражению

путем замыкания магнитных силовых линий на диафрагму высотой l_ленгм и регулирования ее высоты, или на корпус в области его торцов, при этом величину емкости С цепи смещения корпуса плазмовода выбирали из условия

где R_КП - величина сопротивления участка "корпус плазмовода - плазма", Ом; L_г-, C_н - величина индуктивности, емкости цепи разряда соответственно Гн, Ф; а величину емкости С_см цепи смещения корпуса камеры смешения выбирают из условиях

где R_ксп - величина сопротивления участка "корпус камеры смешения - плазма".EXAMPLE N 4. The same as in example N 3, in this case, when the plasma flow is generated, a positive potential jump ΔU _{a is formed} in the anode region of the discharge due to the closure of a part of the magnetic flux Ф _к that penetrates the cathode region of the vacuum arc equal to Ф _And , by adjusting the distance l _AP between the solenoid of the anode region of the discharge and the solenoid of the transportation region, the rest of the longitudinal
flow

the magnetic field is distributed between the transport area

(mixing

) in the Langmuir layer (l _Langm ) and the condensation region (Ф _к ) according to the expression

by closing the magnetic field lines to the diaphragm of height l _Langm and adjusting its height, or to the housing in the region of its ends, the capacitance C of the bias circuit of the plasma duct housing being selected from the condition

where R _KP - the value of the resistance of the section "plasma duct body - plasma", Ohm; L _g -, C _n - value of inductance, capacitance of the discharge circuit, respectively, H, F; and the value of the capacitance C _cm the bias circuit of the housing of the mixing chamber is selected from the conditions

where R _KSP - the resistance value of the plot "mixing chamber body - plasma".

. Соотношение между

и Ф_конд определяет производительность способа через

, т.к. эта величина определяется величиной слоя Ленгмюра (пропорциональна площади слоя) и задается одним из условий транспортирования плазмы вдоль профилированной поверхности: l_ленгм>t_п>d_к. При этом нарушение этого неравенства снижает производительность способа. Оптимальное соотношение между потоками на подложку Ф_к и магнитный поток на формирование слоя Ленгмюра на этапе транспортирования (смешения) реализуется при условии

или, когда D_т >> l_ленгм

Однако, как показал эксперимент, оптимизация топологии магнитного поля есть только необходимое условие повышения производительности способа. Так как плазма генерируется, транспортируется и смешивается в скрещенных

полях, достаточным условием является оптимизация условий реализации Е через соответствующие потенциалы ΔU_а, U_см.т, U_см. Так величина ΔU_а определяется, как ранее описано, величиной Ф_А. Величины U_см.т и U_см. определяются во время транспортирования и смешения соотношением постоянных времени участка корпус-плазма, равной R_кп•C_п и контура, равной

. Так, если

меньше, чем R_кп•C_п (R_ксп•C_см), то емкость С_п (см) успевает разрядиться за счет электронного тока на корпус и в слой Ленгмюра и, как следствие, резкое падение U_см.т, U_см и производительности. Ситуация обратна, если

или

В этом случае U_см.т и U_см, за время прохождения плазмой зоны транспортирования и смешения, изменяются незначительно, а следовательно, и производительность.When implementing the method for producing coatings, the parameter of coatings was measured depending on possible changes in the parameters of the method. Productivity and energy consumption in the method depend on the distribution of the magnetic field in the zones of generation of transportation, mixing and condensation, forming a single magnetic system, and together with ΔU _a , U _t , U _{cm a} single system of crossed fields. Thus, the distribution of the magnetic field in the anode region of the discharge, when there is no magnetic flux to the anode region Ф _А = 0 (magnetic insulation of the anode), the measurement results showed a sharp increase in potential on the electrodes: ΔU _AK = U _K + ΔU ≃ 100 B, the formation of anode spots on the surface of the anode, i.e. an increase in the level of introduced chemical impurities and energy consumption due to a sharp decrease in τ _TAN.G. The change in the value of Ф _{А was} carried out by controlling the distance between the anode solenoid and the plasma duct solenoid l _АП . When the value of l _AP = 5 cm, the potential at the electrodes is equal to U _AK = U _K + ΔU ≃ 50 B in the absence of anode spots. Then optimization was carried out in the distribution of magnetic flux.

. Relationship between

and _{f cond} determines the performance of the method through

because this value is determined by the size of the Langmuir layer (proportional to the area of the layer) and is determined by one of the conditions for transporting the plasma along the profiled surface: l _Langm > t _p > d _k . Moreover, violation of this inequality reduces the performance of the method. The optimal ratio between the fluxes to the substrate Ф _к and the magnetic flux for the formation of the Langmuir layer at the stage of transportation (mixing) is realized under the condition

or, when D _t >> l _Lengm

However, as the experiment showed, optimization of the topology of the magnetic field is only a necessary condition for increasing the productivity of the method. Since plasma is generated, transported and mixed in crossed

fields, a sufficient condition is to optimize the conditions for the implementation of E through the corresponding potentials ΔU _a , U _{cm t} , U _see So the value ΔU _a is determined, as previously described, by the quantity Ф _А. The values of U _{cm t} and U _{cm are} determined during transportation and mixing by the ratio of the time constants of the section of the casing-plasma equal to R _kn • C _p and the circuit equal to

. So if

less than R _kp • C _p (R _ksp • C _cm ), then the capacitance C _p (cm) manages to discharge due to the electronic current to the body and to the Langmuir layer and, as a result, a sharp drop in U _cm , U _cm and performance. The situation is reversed if

or

In this case, U _{cm t} and U _cm , during the passage of the plasma through the transport and mixing zones, change insignificantly, and therefore, productivity.

.It should be noted that the values of R _KP and R _KSP determined by the magnitude of the magnetic flux, closed in the Langmuir layers

.

 Thus, the considered conditions for improving the quality of coatings only in combination of features ensure the achievement of the goal.

 A device for implementing the method of producing multicomponent coatings for achieving the goal must satisfy in all its structural elements (ignition unit, plasma generator, plasma duct, mixing chamber) a number of technical conditions: preservation of the chemical uniformity of the stream (s) and its purposeful change with minimal energy consumption and material losses coverings.

 The chemical composition of the coating in the device is due to the possibility of its violation in the ignition unit, plasma generator, plasma duct and mixing chamber.

So, in the ignition unit, chemical composition violations are caused by:
1. the introduction of a foreign substance, in particular by erosion of the structural elements of the ignition unit, as well as its principle of operation.

2. generating a droplet phase,
3. instability of ignition,
4. absence (synchronization block) of the synchronization mode of the ignition blocks and the jet gas inlet system.

 There are known designs of ignition units, the plasma in which is excited by injection of a portion of gas, high voltage, mechanical movement of the electrode, electron beam, and laser radiation. The use of a portion of gas in the injection device does not satisfy the first requirement. The use of high voltage causes a violation of the first requirement due to erosion of the dielectric, as well as electrodes of a chemical nature different from the consumable cathode, in addition, the metallization of the dielectric surface and the burning of the electrodes of the ignition unit causes the instability of the ignition process. The use of an electrode mechanical displacement device due to the inertia of the system as a whole does not allow the synchronous ignition to be performed within the discharge duration (<1 ms) in plasma generators, in addition, the periodic ignition of the ignition electrode and its mechanical weakness for a number of substances (for example, semiconductors) instability of the ignition process.

To the greatest extent, the requirements (1-5) are satisfied by electron beam and laser ignition. However, they differ significantly in the focusing conditions of electron and laser fluxes to a density of ~ 10 ⁸ W / cm ² (the condition of the solid – plasma transition) on the end working surface of the cathode. So the energy density in the case of electrons is ε = j • U, where j is the current density of the emission from the cathode, U is the accelerating voltage, i.e. To realize jU ~ 10 ⁸ W / cm ² , high-performance electron guns are required. The presence of high voltages (50 kV) and a thermal cathode in the presence of dense flows of metal plasma, as well as flows of reactive technological gases (N ₂ , O ₂ ) causes, firstly, the occurrence of spurious electrical discharges and, therefore, erosion of the structural elements of the ignition unit and, secondly, the destruction of the thermal cathode and, as a consequence, the instability of j • U and the actual ignition itself. Another reason for the instability of the ignition (coating composition) is due to the effect of dagger penetration of the end working surface of the cathode in the focusing zone.

 In the case of laser firing, the depth of the hole is equal to the skin layer (for metals several hundred angstroms), and in the case of the electron beam is equal to the mean free path of electrons at an energy of 50 kV, those are several microns, which leads to a sharp decrease in the yield of erosion products ( plasma) from the focusing area to the interelectrode gap of the plasma generator. For laser firing, the indicated drawbacks are either absent (focusing), and the mass transfer of plasma (or gas) to the thermal cathode is replaced in this case by mass transfer to the optical input, or these drawbacks are not so pronounced (dagger effect). Thus, to realize the goal of the proposal, laser firing (block) in the device is most optimal. The synchronization mode in the device is carried out by a synchronization unit, the control N-channel output of which is connected to each synchronization input of the ignition unit and the jet gas inlet system.

In plasma generators, violations of the chemical composition of the coating can be predetermined, on the one hand, by the type of electric discharge (glow or vacuum-arc), on the other hand, due to
- the ability to adjust the output of the quantity (mass) of plasma depending on electrical parameters, for example, capacitance (C _n ) and operating voltage (U).

- the presence of a droplet phase in the plasma stream,
- cross mass transfer between consumable cathodes of the generators, for example, in plasma, in the droplet phase,
- energy parameters of plasma flow ions,
- anode processes,
- discharge duration τ _G.

 According to the criterion of the type of discharge in a plasma generator, the vacuum-arc discharge is most optimal, since unlike other types of electric discharges, it lacks a working gas (for example, Ar), which sharply increases the chemical purity of the coating. In addition, the mass of the substance (m) of the consumable cathode of the generator is proportional to the capacitance (C) and the change in operating voltage (U). However, the chemical composition of the coating can be locally disturbed within the droplet (fractions - tens of microns) due to heterogeneity (longitudinal, i.e. in thickness, and radial, i.e. along the surface) of different chemical nature, as well as due to cross mass transfer between the cathodes of plasma generators in plasma and droplet phase. Due to this, films or particles (droplets) of foreign substance are present on the end working surfaces of the cathodes, which makes the chemical composition of the plasma from C and U ambiguous in each of the generators and, therefore, in the coating.

The energy aspect of the violation of the chemical composition of the coating is due to the presence in the plasma flows (especially laser) of high-energy ions (up to units of keV), capable of incorporating into the underlying sublattices of a layered unit cell of a multicomponent coating due to its high thermalization. Therefore, energy separation of the plasma flow is necessary. In plasma ducts (mixing chamber), a violation of the chemical composition of the coating may be due to
- the presence of the droplet phase in the plasma stream behind the output end of the plasma duct (mixing chamber),
- cross-mass transfer between plasma ducts in plasma and droplet phase through the mixing chamber,
- anode processes on the inner surface of the plasma duct body and mixing chamber.

 The plasma formation process in the device is accompanied by several mechanisms of the formation of the droplet phase (i.e., laser plasma and vacuum arc plasma are multiphase).

создается током вакуумной дуги. Таким образом, торцевую рабочую поверхность расходуемой мишени и зону расширения можно рассматривать как источник N капель за импульс.The region of macroparticle formation is located on the end working surface of the sacrificial cathode and the expansion of the working substance from the density of the solid (~ 10 ²² cm ^-3 ) to the plasma density (5 • 10 ¹¹ cm ^-3 ), i.e. at a distance of several centimeters from the end working surface. Therefore, the entrainment of particles by the plasma flow in this case is localized in space and is in the region of expansion of the substance. In pulsed solid-state plasma generators, a longitudinal separation of the plasma and the droplet phase is observed due to the difference in the velocities and final duration τ _{G of the} combustion of the vacuum arc (τ _G <10 ⁻³ s).
Therefore, during the burning pulse of the vacuum arc, the droplets do not fly outside the anode and propagate in the plasma duct after the discharge in the generator in the absence of plasma and even a magnetic field, when the magnetic field

created by the vacuum arc current. Thus, the end face of the sacrificial target and the expansion zone can be considered as a source of N drops per pulse.

In a plasma duct, the nature of the propagation of macrodrops is determined by two factors: the phase state of the droplets and the nature of the interaction with the inner wall of the plasma duct body. There are two options for changing the number of drops N _k after the expansion zone: due to crushing and their coagulation.

The droplet phase at the time of formation is in different phase states (liquid and solid). The first option is most dangerous for crushing. However, the time before the first collision with the inner surface of the plasma duct body, macrodrops cool with evaporation by thermal radiation. Moreover, drops with a diameter d _k equal to or less than two values of α ^-1 , as well as a surface layer equal to α ^-1 in drops of a larger diameter, cool (solidify) almost instantly, since this process is not limited by the thermal conductivity of the particulate material, where α is particulate matter absorption coefficient in the wavelength region of the maximum of the Planck function. The droplet phase with a diameter d _to <300 μm during the flight time τ _{1, K} before the first collision should undergo a phase transition: liquid ---> solid. Therefore, for its crushing at the moment of collision, it must undergo a reverse phase transition: solid body ---> liquid. The crushing conditions of the liquid phase are realized at a certain value of the ratio of the kinetic energy of a drop to its energy surface.

, где S_к - сечение взаимодействия (площадь) капель, n_к - концентрация капель. Более того, процесс коагуляции эффективен в случае жидких микрокапель, но они застывают и, более того, взаимно не сталкиваются

.Comparison of the characteristic energy of destruction of the surface of the droplet with the magnitude of the surface tension shows that they differ by 4–5 times. Under the conditions of solid-phase substances, the energy conditions for crushing are more stringent: the kinetic energy of the droplet should be sufficient for the phase transition of the initially solid, as well as droplets cooled during the passage (solid -> liquid) and subsequent crushing proper, taking into account the first remark. The mechanism (coagulation) in the plasma duct will be possible only if the mean free path λ _KK relative to the drop-drop collision is less than the characteristic size of the plasma duct (its length L _g ), i.e. when

where S _k is the interaction cross section (area) of the droplets, n _k is the droplet concentration. Moreover, the coagulation process is effective in the case of liquid microdrops, but they freeze and, moreover, do not collide

.

Thus, the number of drops N _k is constant for the input end of the plasma duct, but depends on a number of controlled parameters (for example, gas content, discharge current) and can be determined from experiment.

(и равного величине угла между направлением распространения макрочастицы и нормалью к поверхности профиля, расположенной в радиальной плоскости корпуса) можно выделить три группы макрочастиц. Для первой группы угол влета изменяется в пределах α_п> γ > 0, где α_п - угол при вершине треугольного профиля. Указанные макрочастицы после соударения с поверхностью выступа отражаются от последнего и соударяются с поверхностью соседнего выступа, отражаются от него и под углом 2α_п-γ₁ распространяются в направлении выходного торца, т.е. они являются незапертыми после столкновения с треугольным профилем (см. фиг. 4). Для второй группы угол влета изменяется в диапазоне α ≤ γ₂< 2α_п, и макрочастицы после столкновения с выступами профиля обратно отражаются в направлении влета, т.е. являются запертыми. Для третьей группы угол влета изменяется в диапазоне

, и макрочастицы "запираются" в пределах соседних выступов. Максимальную величину выступов h_{п max}, если

во втором варианте профиля выбирают такой, чтобы в результате столкновений при входе макрочастиц третьей группы внутрь треугольного профиля они потеряли первоначальный импульс. Величину h_{п max} можно уменьшить в два раза, если учесть, что реальный профиль имеет вид, изображенный на фиг. 5, поэтому макрочастица, отразившись от области К, испытывает еще n_ст столкновений при выходе из треугольного профиля. Таким образом, для запирания макрочастиц на треугольном профиле (фиг. 5) необходимо устранить первую группу макрочастиц. Рассмотрим условия их запирания. Последнее осуществляется тем, что, во-первых, угол влета макрочастиц увеличивается от величины, заключенной в диапазоне 0 < γ < α_п после первого столкновения с треугольным профилем до величины, заключенной в диапазоне γ = α_п÷ 2α_п и при втором столкновении макрочастиц с треугольным профилем их новый угол влета γ = α_п÷ 2α_п соответствует уже второй группе макрочастиц. Во-вторых, тем, что макрочастицы отражаются по закону tgφ = K_вtgψ , где φ и ψ - углы падения и отражения соответственно, K_в - коэффициент восстановления (К_в~0,5). При этом началу запирания соответствует случай, когда после первого столкновения с профилем частицы в области r_NТ начинают распространяться радиально, а не к выходному торцу. Причем по мере уменьшения α_п у них появляется компонента скорости к входному торцу плазмовода. Этому условию соответствует верхняя величина α_п, равная

,
где D_т - поперечный размер зоны транспортирования,
r_к - величина радиуса катода (мишени),
F - величина отрезка прямой, проходящей через торцевую рабочую поверхность в точке D_т/2+r_к и точкой касания к поверхности плазмовода в области цилиндра r_NТ. По мере дальнейшего уменьшения α_п капельная фаза начинает испытывать уже два столкновения в области между соседними выступами профилями, т. е. на одно больше чем при отсутствии профиля. Этому условию соответствует верхняя величина α_п, равная

.The first necessary condition for locking the droplet phase is the presence of a triangular profile of the inner surface, provided that the height and step of the profile are larger than the size of the droplet phase (otherwise the surface is "smooth"). In this case, according to the criterion of the angle of entry, varying within

(and equal to the angle between the direction of propagation of the particles and the normal to the surface of the profile located in the radial plane of the body), three groups of particles can be distinguished. For the first group, the angle of entry varies within α _p >γ> 0, where α _p is the angle at the apex of the triangular profile. These particles after collision with the surface of the protrusion are reflected from the latter and collide with the surface of the adjacent protrusion, are reflected from it and propagate in the direction of the output end face at an angle 2α _p −γ ₁ , i.e. they are unlocked after a collision with a triangular profile (see Fig. 4). For the second group, the angle of entry varies in the range α ≤ γ ₂ <2α _p , and the particles after collision with the protrusions of the profile are reflected back in the direction of entry, i.e. are locked. For the third group, the angle of entry varies in the range

, and the particles are "locked" within adjacent protrusions. The maximum size of the protrusions h _{p max} if

in the second version of the profile, one is chosen so that as a result of collisions at the entrance of the particles of the third group into the triangular profile, they lose their initial impulse. The value of h _{p max} can be reduced by half if we take into account that the real profile has the form depicted in FIG. 5, therefore, the particulate, reflected from region K, experiences another n _st collisions upon exiting the triangular profile. Thus, for locking particles on a triangular profile (Fig. 5), it is necessary to eliminate the first group of particles. Consider the conditions for their locking. The latter is due to the fact that, firstly, the angle of entry of the particles increases from a value enclosed in the range 0 <γ <α _p after the first collision with a triangular profile to a value enclosed in the range γ = α _p ÷ 2α _p and during the second collision of the particles with a triangular profile, their new entry angle γ = α _p ÷ 2 α _p corresponds to the second group of particles. Secondly, the fact that the particles are reflected according to the law tgφ = K _in tgψ, where φ and ψ are the angles of incidence and reflection, respectively, K _in is the recovery coefficient (K _in ~ 0.5). In this case, the start of locking corresponds to the case when, after the first collision with the profile, the particles in the region r _NT begin to propagate radially, and not to the output end. Moreover, as α _p decreases, they have a velocity component to the input end of the plasma duct. This condition corresponds to the upper value of α _p equal to

,
where D _t - the transverse size of the transportation zone,
r _to - the radius of the cathode (target),
F is the value of the straight line segment passing through the end working surface at the point D _t / 2 + r _k and the point of tangency to the surface of the plasma duct in the region of the cylinder r _NT . With further reduction α _n droplet phase already starts to experience two collisions in the region between adjacent protrusions profiles r. E. More than one in the absence of the profile. This condition corresponds to the upper value of α _p equal to

.

With a further decrease in α _{p, the} number of collisions between adjacent protrusions (t _p ) increases and they can be considered as regions of the runoff of the droplet phase.

, для второго: α_п< γ₂< 5α_п.A second variant of profiling the inner surface of the housing is possible, FIG. 6. In this case, the angle of entry for the third group is in the range

, for the second: α _p <γ ₂ <5α _p .

, при новом угле влета частиц первой группы, равном γ = α_п-2α_п. Эффективность очистки можно повысить путем увеличения угла β_Т до 360^o. В данном случае можно выделить область II (фиг. 3), в которой угол β_II изменяется в пределах 180^°< β_II< 360^°, a β_т,о= 180^°. В этой области каждая макрочастица, вылетевшая с торцевой рабочей поверхности, испытывает минимум два столкновения, а вершины треугольного профиля области 1 являются источником

макрочастиц для области II. Число "не запертых" на вершинах профиля макрочастиц после области II равно

. Дальнейшее увеличение угла плазмовода до 540^o позволяет выделить в нем область III, в которой угол β_III изменяется в пределах 360^o < β_III< 540^°, а β_т,о= 180^°. После этой области каждая макрочастица испытывает минимум три столкновения. Необходимое и достаточное условие полного запирания макрочастиц в плазмоводе можно записать в виде неравенства P•N_к <1, где N_к - число макрочастиц, генерируемых в импульсе плазмы, P - геометрическая вероятность запирания, равная

. В этом случае все макрочастицы из общего числа N_к "запираются" в плазмоводе. Эффективность очистки от капельной фазы можно
существенно повысить, если внутри внешнего корпуса расположить дополнительный корпус, внешняя и внутренняя поверхность которого профилированы аналогично внешнему. В данном случае (фиг. 4) уменьшается величина β_т,o и она становится меньше 180. Однако уменьшение β_т,o по всему сечению плазмовода возможно в том случае, даже при наличии дополнительного корпуса для центральной области плазмовода (фиг. 1) β_т,o = 180^°, и не зависит от числа дополнительных корпусов. Увеличение числа дополнительных корпусов, если малый диаметр 2r_NТ проекции сечения основного корпуса больше нуля 2r_NТ > 0, будет повышаться эффективность очистки от капельной фазы, однако их средние диаметры должны быть такими, чтобы β_тo для всех зон между корпусами должен быть приблизительно одинаков. В противном случае геометрическая вероятность пролета будет различаться по сечению плазмовода. Характер взаимодействия капельной фазы с внутренней поверхностью камеры смешения аналогичен взаимодействию с поверхностью корпуса (корпусов) плазмовода, поэтому требования к ее профилю аналогичны. Однако наличие камеры смешения увеличивает азимутальный угол β_т между входным торцом плазмовода и выходным торцом камеры смешения до величины β_т,c т.е. практически появляется новая зона (зоны) столкновения с угловым размером одной зоны столкновения, равным β_тo, поэтому условия запирания капельной фазы становятся более жесткими.For the first group, the angle of entry varies within 0 <γ ₁ <α _p . In this case, the first group extends in the direction of the output end after a collision with one protrusion, and not two, as in the case of the first version of the profile, which is a drawback of the second option before the first version of the profile. The locking conditions of the first group are carried out similarly to the locking conditions in the first version of the profile. According to the criterion of the first collision in the plasma duct, we can distinguish region 1 (Fig. 3), in which the angle β ₁ varies in the range 0 <β ₁ ≤ π, and β _{t, o} = 180 ^° . In this region, each particulate emitted from the end working surface experiences at least one collision with the wall, which is not sufficient for locking the droplet phase, since there is a first group of particulates at γ <α _p and because the real triangular profile differs from the theoretical one in region of the peak (Fig. 5). Thus, on a triangular profile, the particles 32 that fall between the protrusions of the profile are “locked” if γ ≥ α _p and are not “locked” if γ <α _p , and also the particles 33 that are not on the top of the real protrusion are not “locked” , i.e. for the last particles the walls are “smooth”. The number of “non-locked” particles at the top of the particle profile after the first collision region is

, at a new angle of entry of particles of the first group equal to γ = α _p -2α _p . The cleaning efficiency can be increased by increasing the angle β _T to 360 ^o . In this case, we can distinguish region II (Fig. 3), in which the angle β _II varies within 180 ^° <β _II <360 ^° , and β _{t, o} = 180 ^° . In this region, each particulate emitted from the end working surface experiences at least two collisions, and the vertices of the triangular profile of region 1 are a source

particulate for region II. The number of “non-locked” particles at the vertices of the profile of particles after region II is

. A further increase in the angle of the plasma duct to 540 ^o allows you to select region III in it, in which the angle β _III varies within 360 ^o <β _III <540 ^° , and β _{t, o} = 180 ^° . After this area, each particle experiences a minimum of three collisions. A necessary and sufficient condition for the complete locking of the particles in the plasma duct can be written in the form of the inequality P • N _k <1, where N _k is the number of particles generated in the plasma pulse, P is the geometric probability of locking equal to

. In this case, all of the particulates from a total of N _{to the} "locked" in the plasma guide. The efficiency of cleaning from the droplet phase can
significantly increase if an additional case is located inside the external case, the external and internal surface of which is profiled similarly to the external. In this case (Fig. 4), the value of β _{t, o} decreases and it becomes less than 180. However, a decrease in β _{t, o} over the entire cross section of the plasma duct is possible even if there is an additional housing for the central region of the plasma duct (Fig. 1) β _{t, o} = 180 ^° , and does not depend on the number of additional buildings. Increasing the number of additional shells when small diameter 2r _NT projection section of the main body is greater than zero 2r _NT> 0, will increase the cleaning efficiency of the droplet phase, however, their average diameter should be such that the β _verily for all zones between the housings should be about the same. Otherwise, the geometric probability of span will vary across the cross section of the plasma duct. The nature of the interaction of the droplet phase with the inner surface of the mixing chamber is similar to the interaction with the surface of the plasma duct body (s), therefore, the requirements for its profile are similar. However, the presence of the mixing chamber increases the azimuthal angle β _t between the input end of the plasma duct and the output end of the mixing chamber to β _{t, c} i.e. practically there is a new zone (s) of collision with a collision angular size area equal to _verily β, so dropping the phase locking conditions become more stringent.

Thus, the nature of the propagation of macroparticles in plasma-optical systems (and mixing chambers) is determined by the nature of the interaction of solid particles with the surface of the housing (housings), therefore, all geometric dimensions of the device nodes must be determined from the necessary and sufficient conditions for locking the droplet phase on the triangular profile of the inner surface of the housing (housings) ) and which are physically reduced, firstly, to the extinction of kinetic energy in inelastic collisions, a microparticle ---> the profile surface K _{at the} nth hour energy (where K _in is the recovery coefficient, K _in ~ 0.5) per one collision and, secondly, a change in the direction of the initial pulse of microparticles during collisions in the direction to the input end of the plasma duct or inside a triangular profile. In this case, the limiting geometric parameters of the device are the characteristics of the triangular profile (step, height, angle at the apex and the dullness of the vertex), as well as the azimuthal angle β _t (and β _{t, s} ) of the reversal of the ends of the plasma duct (or plasma ducts and mixing chambers). Moreover, these parameters are reproducible technical characteristics laid down at the stage of its manufacture (plasma ducts and mixing chambers).

полей камеры смешения (магнитного поля

соленоидов камеры смешения и радиального электрического поля

корпуса камеры смешения), где они смешиваются и в виде многокомпонентного однофазного потока плазмы поступают в область конденсации через выходной торец камеры смешения.As transportation in the plasma ducts, the plasma flows are cleaned from the droplet phase on the profiled (threaded) inner surface and enter the crossed region

fields of the mixing chamber (magnetic field

mixing chamber solenoids and radial electric field

mixing chamber bodies), where they are mixed and in the form of a multicomponent single-phase plasma flow enter the condensation region through the output end of the mixing chamber.

,
4) минимальным потерям плазмы на этапе смешения,
5) запиранию капельной фазы (остатка от этапа транспортирования),
6) отсутствие перекрытого массопереноса между каналами устройства.The mixing chamber must meet in the aggregate a number of requirements:
1) the conditions of uniform radial mixing,
2) uniformity of coatings in thickness,
3) minimum energy costs for creating crossed fields

,
4) minimal plasma losses at the mixing stage,
5) locking the droplet phase (residue from the transportation stage),
6) the absence of an overlapped mass transfer between the channels of the device.

The first requirement is limited by the value of r _i (more precisely, by the ratio of r _i and D _c (r _i > D _c , r _i <D _c and λ _ii and D _c (λ _ii > D _c , λ _ii <D _c ), as well as _NC and l _c1 , where D _с is the diameter of the mixing chamber γ _NC is the value of the angle at the apex of the mixing chamber (practically the angle of fusion of the initial) plasma flows; l _c1 is the length of the mixing region (output segment of the mixing chamber, see description of the device manufacturing method) The effect of r _i and λ _ii on mixing was considered previously.

.Depending on the value of γ _NC in the mixing chamber, there are two limiting options: a) frontal collision of plasma flows when γ _NC = π and b) γ _NC = 0.
For practice, neither the first a) nor the second b) cases are acceptable: a) due to large plasma losses and low uniformity of coatings in thickness and composition, and case b) due to technical difficulties in implementation, but in this embodiment plasma losses are minimal, high uniformity in thickness and composition of the coating. Therefore, γ _NC should be optimized in the range 0 <γ _NC <π. The best option is when γ _NC = 180 ^° -2α _NT .
Consider the effect of l _c1 . In the mixing chamber, in addition to the angular size γ _NC, there are two linear ones that must be optimized. The first linear dimension is the distance from the output end of the mixing chamber to the spatial information region of the plasma flows and which in essence is the mixing length l _c1 (or the length of the output segment, see the description of the manufacturing method of the device). The second linear dimension from the spatial domain of the plasma flows to its input ends (or the length of the input segment l _s2 , see the description of the manufacturing method of the device). The value of l _{c2 is} determined from the condition of radial docking of the mixing chamber with a system of multi-pass screw plasma ducts, and since the radial dimensions of the latter are optimized, this implies the value of l _c2 equal to

.

(см. фиг. 33). Величина однородности покрытий по толщине после камеры смешения определяется величиной однородности магнитного поля

в зоне конденсации и которая в свою очередь определяется рядом факторов:
а) величиной γ_N,C(α_N,T) и б) величиной l_с1, а также длиной соленоида с тыльной стороны подложки.The value of l _c1 in the case of mixing conditions can be optimized based on the condition of at least one collision of the ion component with the energy barrier U _see the mixing chamber, which randomizes the initial (before mixing) radial distribution of elements (ions). Consequently,

(see Fig. 33). The uniformity of coatings in thickness after the mixing chamber is determined by the uniformity of the magnetic field

in the condensation zone and which in turn is determined by a number of factors:
a) the value of γ _{N, C} (α _{N, T} ) and b) the value of l _s1 , as well as the length of the solenoid on the back of the substrate.

The indicated values have already been considered from the mixing condition. It should be noted that a change in these values to satisfy the mixing conditions and uniformity of coatings are not contradictory: i.e. as l _c1 increases and γ _{N, C} decreases _{, the} characteristics improve.

It should be noted that the nature of the current transfer to the profiled surface of the mixing chamber is similar to the current transfer to the profiled surface of the plasma duct body, therefore, the requirements are similar. In this case, when both cases are galvanically connected, the case of parallel connection of the resistances R _ksp R _{kp is realized} .

и топологии суммарного магнитного поля

в плазмоводах и камере смешения.The plasma transport coefficient (productivity), as well as the uniformity of the coatings over the thickness in the device for obtaining multicomponent plasma flows through their synthesis, depends on the values

and topology of the total magnetic field

in plasma ducts and mixing chamber.

The absence of mutual influence of the combined plasma ducts will be observed only if the magnetic flux of the scattering fields in each of the plasma ducts is zero, i.e. ΔФ _NT = 0. Otherwise, partial closures of the magnetic field lines to the plasma duct body will be observed, which leads to a decrease in the transport coefficient (productivity), since in each of the mixed flows part of the plasma of the working substances is lost (goes into the wall): Δn ~ ΔФ _{N, T.} Similarly for the mixing chamber. The mutual influence of magnetic fields of combined plasma ducts can be of two types: by changing the uniformity of the magnetic field or by changing the field topology. In combined curvilinear systems there are no options with ΔФ _{N, Т} = 0, but in any spiral duct plasma duct this value is less than in the case of toroidal plasma ducts. Therefore, from the point of view of reducing the plasma loss of the working substances and the uniformity of the coating thickness in the device, a system of screw plasma ducts is preferable compared to a system of toroidal plasma ducts with the equivalent droplet phase separation effect in both cases. In addition, the azimuthal angle between the ends, and therefore the degree of separation of the droplet phase, is unlimited in them.

In practice, the dimensions of the device (plasma ducts, mixing chambers) are a limiting factor, in particular, this is due to the span and centrifugal effects of separation of the plasma components, as well as plasma losses, because the magnetic flux and the flux of electrons (ions) are proportional to the area of the inner surface of the plasma duct body, therefore, it should be minimal (i.e., the minimum volume). The reduction in the size of the device from screw plasma ducts is reduced to a decrease in radial and longitudinal dimensions. The radial dimensions will be smallest when a tight arrangement of screw plasma ducts is provided in the plane of the cross section, see FIG. 10. The minimum distance between them is determined from the value of the technological gap h _T.

. Рассмотрим прямоугольный треугольник Δ СОМ (фиг. 11):

Получаем величину

Продольные габариты устройства будут наименьшими, когда обеспечена плотная компоновка винтовых плазмоводов в плоскости продольного сечения, проходящей через центральную ось О конструкции, т.е. когда в одном шаге системы многозаходных винтовых цилиндров укладывается N плазмоводов с внешним диаметром D_Т и зазором 2h_Т. Последнее условие запишется в виде

, где L_э - величина дуги эллипса АВ (см. фиг. 11).Conditions smallest radial dimensions is formed as follows: in the plane A-A (Figure 10.) At r _NT radius circumferentially evenly and with a gap 2h _T relative to each other are distributed N ellipse with the minor axis O _T equal to the outer radius of the winding duct body (FIG. 10) forming an angle with the axis O

. Consider a right triangle Δ COM (Fig. 11):

We get the value

The longitudinal dimensions of the device will be the smallest when a tight arrangement of screw plasma ducts is provided in the plane of the longitudinal section passing through the central axis O of the structure, i.e. when N plasma ducts with an external diameter D _T and a gap of 2h _T are stacked in one step of a multi-start screw cylinder system. The last condition is written as

, where L _e - the magnitude of the arc of the ellipse AB (see Fig. 11).

,
а величину H в равенство

(см. фиг. 12), получаем уравнение для нахождения величины α_NT

,
где E(...) эллептический интеграл второго рода.Expressing the parameters of the ellipse a, b and ε through specific expressions and substituting the obtained value of L _e in the expression

,
and the value of H in the equality

(see Fig. 12), we obtain the equation for finding α _NT

,
where E (...) is an elliptic integral of the second kind.

Deviation of α _NT and r _{NT values} to a smaller side from that indicated in the equalities leads to the impossibility of real assembly of the device, since in this case the surfaces of the plasma ducts begin to intersect. Deviation quantities r and α _{NT NT} upwards from the said leads to an unjustified increase of the device size and, consequently, the plasma losses on the plasma duct redundant surfaces and increase the role of transit effects on the composition of the coatings. Recent comments apply to the mixing chamber.

h_тm - величина технологического зазора между системами многозаходных винтовых плазмоводов. В области подложки вследствие наличия системы N многозаходных винтовых плазмоводов или m подобных систем, во-первых, неравномерно распределение силовых линий магнитного поля и, во-вторых, из-за кручения вектор

не перпендикулярен поверхности подложки, что в совокупности ухудшает равномерность покрытия и возможность конденсации на профилированные подложки без "подпыла" на вертикальные стенки. Указанные недостатки можно частично снять путем расположения с тыльной стороны подложки N соленоидов или полностью путем симметричного расположения относительно плоскости подложки (или двух подложек, сложенных вместе (устройства и просто магнитной системы) или двух аналогичных устройств, причем направления кручения винтовых плазмоводов в обоих устройствах противоположны, а соленоиды плазмоводов включены согласно (см. фиг. 15). В противном случае, когда направления кручения плазмоводов одинаковы, проблема усугубляется, так как суммируемые векторы магнитных полей соленоидов плазмоводов в области подложки имеют одинаковое направление, а в случае, когда направления противоположны, то суммарный вектор

области конденсации перпендикулярен поверхности пoдложки. Последнее справедливо, если соленоиды плазмоводов включены согласно, так как в противном случае в области подложки реализуется топология магнитного поля антипробкотрона, что устраняет перпендикулярность суммарного магнитного поля

поверхности подложки, и снижает, кроме того, производительность. Таким образом симметричное расположение с обратным кручением при согласном включении обеспечивает повышение производительности в 2 раза при реализации требований, изложенных в цели изобретения.Three variants of the device for producing coatings are possible according to the criterion of the number of mixing chambers and the number of systems of multi-pass screw plasma ducts: firstly, one mixing chamber and one or more systems of multi-pass screw plasma ducts, and secondly, several mixing chambers (i.e., the output end face of one connected to one of the input ends of the other mixing chamber) and one system of multi-pass screw plasma ducts and; thirdly, several mixing chambers and several systems of multi-pass screw plasma ducts. In all cases, the optimal value of the angle γ _{N, m} is determined from the condition γ _{N, m} ≤ 180 ^° -2α _NT , where N is the number of plasma ducts in the system of equidistant multi-start screw plasma ducts, m is the number of systems of equidistant multi-start screw plasma ducts in the device. The radius of the base Z _CNm in this case is determined from the expression

h _tm - the size of the technological gap between systems of multiple screw plasma ducts. In the substrate region, due to the presence of a system of N multi-helical plasma ducts or m similar systems, firstly, the distribution of magnetic field lines is uneven and, secondly, because of torsion, the vector

not perpendicular to the surface of the substrate, which together worsens the uniformity of the coating and the possibility of condensation on the profiled substrates without "dusting" on the vertical walls. These disadvantages can be partially removed by arranging N solenoids on the back of the substrate or completely by symmetrically positioning relative to the plane of the substrate (or two substrates stacked together (a device and simply a magnetic system) or two similar devices, the torsion directions of the screw plasma ducts in both devices being opposite, and the plasma duct solenoids are turned on according to (see Fig. 15). Otherwise, when the torsion directions of the plasma ducts are the same, the problem is aggravated, since s of the magnetic field of solenoids plasma ducts in the region of the substrate in the same direction, and in the case when the directions are opposite, the total vector

the condensation area is perpendicular to the surface of the substrate. The latter is true if the plasma duct solenoids are switched on according to, since otherwise the topology of the magnetic field of the antiprobotron is realized in the region of the substrate, which eliminates the perpendicularity of the total magnetic field

surface of the substrate, and reduces, in addition, performance. Thus, a symmetric arrangement with reverse torsion with consonant inclusion provides a 2-fold increase in productivity when implementing the requirements set forth in the purpose of the invention.

The reliability and level of impurities in the plasma of solids with a laser ignition unit can vary due to mass transfer from the end working surface of the cathode to the input window (or transparent screen) both in the ignition mode and in the main discharge combustion mode, when mass transfer is maximal and occurs subsequent regeneration of their transparency by laser radiation. The latter leads to an increase in the roughness of the working surface of the input window (or transparent screen), as well as introduced chemical impurities towards the end working surface of the sacrificial cathode. The specified mass transfer is carried out by a plasma stream, which contains a droplet phase, ionized and neutral components. To eliminate the deposition of the droplet phase onto the input window (or transparent screen), they must be installed at a distance from the end working surface of the cathode, during which the droplet phase cools down (there is a phase transition of the liquid into a solid) and is elastically reflected from their surface: i.e. case of a telephoto lens. In order to reduce the flow of ionized and neutral plasma components to the surface of the input window (or a transparent screen), the distance from the end working surface to the focusing system should be greater than the focal length of the latter. Such a technical solution allows, firstly, to use the diaphragm by placing it on the optical axis of the ignition unit between the end working surface and the radiation input window at a distance equal to the focal length of the focusing system, and, secondly, to reduce the growth of the crater depth in the laser focusing zone radiation on the end working surface, and thirdly, to coordinate the diameter D _d in the diaphragm with the size of the Langmuir layer (D _d <l _Langm ). In this case, the laser radiation is partially defocused, which leads to a decrease in the crater depth in the ignition mode and its growth is commensurate with its decrease due to erosion of the end working surface in the main discharge combustion mode. The use of a diaphragm reduces the plasma flow to the radiation input window.

Almost the minimum value of D _d is chosen so that, firstly, the laser radiation in focus does not cause erosion of the material of the diaphragm in the ignition and combustion mode of the main discharge (due to the excitation of the anode spots on the diaphragm), which reduces the level of chemical purity of the plasma stream in comparison with initial level of purity of the substance of the sacrificial cathode. And, secondly, the size of the hole in the diaphragm D _{d is} limited from below to the diameter d _to drops: D _d > d _k . From above, the diameter D _{d of the} diaphragm is limited by the size of the Langmuir layer: D _d <l _Langm.
In this case, the positive potential jump at the anode (+ ΔU _a ) captures the region of the diaphragm hole, and the ion component is reflected by the potential barrier ΔU _a from the hole into the plasma. An ionized component (ions) can pass into the aperture of the diaphragm, whose energy is greater than the value of the potential jump + ΔU _a . Otherwise (when D _d > l _{Langm the} potential barrier (+ ΔU _a ) near the optical axis of the diaphragm is absent and ions pass through the hole in the diaphragm unhindered. In addition, the plasma flow through the diaphragm can be reduced if the thickness of the diaphragm S is taken to be S = (3-4) D _d . In this case, a diaphragm cavity of size D _d xS is a hollow anode, so the plasma in this case penetrates the cavity several calibers of the latter. Optimizing the size of the diaphragm reduces the flow of plasma ions through the hole to the input window cheniya, which increases the reliability of the plasma generator and the level of the plasma stream purity size D _d of the diaphragm has a physical upper limit -. l _lengm and bottom -. d _a However l _lengm determines the condition of "locking" the ionic component of the plasma, and the value d _k - - input laser radiation The optimal value of D _d located between these extreme values.

 The distance d between the end working surface of the sacrificial target and the focusing lens is limited to a minimum value of f when the diaphragm is in the region of the end working surface and maximum equal to 2f when the diameters (and density) of the initial laser radiation (to the lens) and focused ( after the lens) on the sacrificial cathode.

The reliability of the plasma generator can be increased by reducing the level of introduced impurities, if, firstly, the laser beam between the diaphragm and the input window is galvanically isolated from the anode and grounded or applies a negative potential, and secondly, a longitudinally movable beam is placed in front of the input window a transparent screen fixed in the feed mechanism and moved by a mechanism for controlling the length of the corrugated tube. In the first case, the reflecting potential barrier (+ ΔU _a ) is eliminated in the beam line and ions are deposited on the surface of the beam line, in the second case, as part of the transparent screen is dusted due to the residual mass transfer, a clean transparent surface of the transparent screen is introduced into the laser beam synchronously with the target being consumed .

 To eliminate mass transfer and the effect of regeneration on the chemical composition in the area between the diaphragm and the input window, a normally open shutter is installed in the form of a lever, which is driven by an electromagnetic system that includes an armature connected to the smaller lever of the shutter, as well as a magnetic circuit and two coils located in the magnetic circuit . One of the coils is connected in series to the discharge circuit of the pump lamps of the emitter, and the second - according to the first and sequentially to the discharge circuit of the plasma generator. Such switching on the coils allows you to pre-set (accelerate) the shutter with the armature before initiating a discharge in the generator, which allows, firstly, to avoid the passage of the droplet phase and the neutral component at the leading edge of the discharge current in the plasma generator due to the inertia (acceleration) of the shutter and the armature and, secondly, block the aperture opening at maximum speed, since the shutter already has a non-zero speed by the time the discharge current (mass transfer) appears.

. Начальное положение (нормально открытое) конца большого рычага-заслонки в области диафрагмы устанавливают путем регулирования, чтобы к моменту прекращения тока разряда лампы накачки край большего рычага-заслонки совпадал с кромкой отверстия в диафрагме, что позволяет беспрепятственно пройти через диафрагму лазерному излучению за время ~10^-9 с. За это время положение заслонки остается практически неизменным. Затем происходит перекрытие заслонкой отверстия в диафрагме вплоть до момента прекращения разряда (т.е. массопереноса). После чего осуществляется возвращение пружиной заслонки в нормально открытое состояние до момента следующих импульсов тока накачки в лампе и тока разряда в генераторе.In this case, the overlap time of the hole in the diaphragm decreases in proportion to the initial speed and the product of the ratio of the lengths of the levers of the damper and the cross-sectional areas of the laser beam and the hole in the diaphragm. In addition, for a given value of the aperture in the diaphragm, the use of the shutter-lever allows decreasing the stroke (δ) of the armature (δ ~ 0.2 - 0.3 mm), which in turn increases the strength (speed) of the armature, since in this region δ magnitude of force is proportional

. The initial position (normally open) of the end of the large damper lever in the diaphragm region is established by adjusting so that by the time the discharge current of the pump lamp ceases, the edge of the larger damper arm coincides with the edge of the hole in the diaphragm, which allows laser radiation to pass through the diaphragm unhindered for a time of ~ 10 ^-9 p. During this time, the position of the flap remains virtually unchanged. Then, the shutter closes the opening in the diaphragm until the discharge ceases (i.e., mass transfer). After that, the damper spring returns to the normally open state until the next impulses of the pump current in the lamp and the discharge current in the generator.

 The essence of the proposed method lies in the fact that the body of the plasma duct in the form of a screw cylinder and the body of the mixing chamber with profiled inner surfaces is approximated by a combination of welded segments that are elements of a straight hollow cylinder (pipe) with a profiled inner surface.

, при этом большие оси эллипса в сечениях развернуты на угол, равный γ_к, в плоскости, перпендикулярной оси трубы (фиг. 27 и фиг. 28). Второй вариант реализуется в том случае, если плоскости торцов сегмента образуют угол, равный

(угол между большими осями эллипсов сечений), т. е. как в первом случае, но в отличие от него малые оси эллипсов в сечениях развернуты на угол, равный γ_К в диаметральной плоскости трубы. Таким образом, эллипсы сечений в обоих вариантах имеют равные большие оси, но разные малые. Во втором варианте малая ось в

раз больше, что обусловливает и большую длину периметра эллипса, поэтому этот вариант менее предпочтителен, так как суммарный вакуумный шов в таком винтовом корпусе больше,
чем в первом варианте, что в свою очередь обусловливает рост трудоемкости изготовления и вероятности образования негерметичности.There are two options for manufacturing a screw plasma duct housing from straight pipe segments. The first option is implemented if the planes of the end faces of the segment form an angle equal to

while the large axes of the ellipse in sections are rotated at an angle equal to γ _k in a plane perpendicular to the axis of the pipe (Fig. 27 and Fig. 28). The second option is realized if the planes of the end faces of the segment form an angle equal to

(the angle between the large axes of the ellipses of the sections), i.e., as in the first case, but in contrast to it, the small axes of the ellipses in the sections are rotated by an angle equal to γ _K in the diametrical plane of the pipe. Thus, the ellipses of the cross sections in both cases have equal major axes, but different small ones. In the second embodiment, the minor axis in

times more, which leads to a greater length of the perimeter of the ellipse, therefore this option is less preferable, since the total vacuum seam in such a screw housing is greater,
than in the first option, which in turn leads to an increase in the complexity of manufacturing and the likelihood of leakage.

(см. фиг. 12), причем

Таким образом,

.We define the parameters of the segment — the distance l _k and the angle γ _K — between the normals of the consecutive K sections of the straight pipe for the approximation of a screw cylinder by a set of straight segments (see Fig. 29). To determine l _to calculate the length of the helix along the inner surface of the screw cylinder. The total length of the helix is

(see Fig. 12), and

Thus,

.

и по внутреннему (когда

).In this case, two calculation options are possible - according to the outer diameter of the multi-helical screw cylinder system (in this case

and inward (when

)

, где L_в - длина винтовой линии.Calculation and manufacture of a screw cylinder according to the internal diameter is preferable in the sense that in this region the distance between the two sections is minimal. Therefore, knowing the minimum possible distance l _sv between two welds, we can determine the maximum number of approximation segments

where L _in - the length of the helix.

где

, причем для винтового цилиндра на один шаг φ_Т= π.
Определим угол поворота γ_К между двумя последовательными сечениями сегмента (см. фиг. 29).Using the expansion of a series of integrands, we obtain the following approximate expression for l _to :

Where

, and for a screw cylinder one step φ _T = π.
We determine the angle of rotation γ _K between two consecutive sections of the segment (see Fig. 29).

Подобная последовательность операций позволяет изготовить при φ_т = π винтовой корпус на один шаг винтового цилиндра. Для реализации винтового корпуса на m шагов необходимо изготовить число сегментов, равное к•m.Thus, γ _K is equal to

Such a sequence of operations makes it possible to fabricate at φ _t = π a screw housing for one step of a screw cylinder. To implement a screw housing with m steps, it is necessary to produce the number of segments equal to • m.

Из условия стыковки выходного с N входными сегментами, расположенными вдоль образующих ДМ и ДЕ конуса с углом при вершине γ_NC (см. фиг. 33), следуют требования к геометрии торцов выходного и N входных сегментов. У выходного сегмента торец со стороны входных сегментов получают сечением трубы N плоскостями, каждая из которых наклонена к оси трубы под углом ξ₁ (фиг. 33) и повернута относительно соседних по азимуту на угол

вокруг оси выходного сегмента (фиг. 35). Величина ξ₁ определяется из условия

(фиг. 33). Соблюдение соотношения

обеспечивает равномерное распределение по азимуту N входных сегментов относительно оси LO (фиг. 33) выходного сегмента, а также стыковку соседних входных сегментов между собой по периметру ДО (фиг. 35). Входные сегменты сопрягаются с выходным путем стыковки их торцов, поэтому каждый из входных сегментов должен иметь часть торца ДС и АВ (фиг. 33), образованного сечением плоскостью, размещенной под углом ξ₁ (фиг. 35) относительно трубы (сегмента), а также для сопряжения входных торцов между собой, оставшаяся часть ДО их торцов должна быть образована вторым сечением, двумя скрещенными под углом

плоскостями, при этом линия пересечения скрещенных плоскостей должна иметь угол ξ₃ с осью трубы (сегмента) и совпадать с осью выходного сегмента, при этом величина ξ₃ определяется из соотношения

(фиг. 33, фиг. 32). Длину входных сегментов l_с2 выбирают из условия равенства радиальных размеров камеры смешения с системой многозаходных винтовых плазмоводов (фиг. 33, фиг. 32), т.е.A method of manufacturing a mixing chamber is similar to the method of manufacturing screw plasma ducts, i.e., from straight segments of a profiled pipe with an inner diameter D _c . In the particular case of D _t -D _with . The mixing chamber (Fig. 30) is made of (2N + 1) segments: one output segment 34, N - input segments 35 and N - transition segments 36. The length of the output segment l _{c1 is} selected from the condition of at least one collision of the ion component with an energy barrier U _cm wall (Fig. 33):

From the condition of coupling the output with N input segments located along the generators of the DM and DE cones with an angle at the vertex γ _NC (see Fig. 33), the geometry requirements for the ends of the output and N input segments follow. At the output segment, the end face from the input segments is obtained by N pipe section by planes, each of which is inclined to the pipe axis at an angle ξ ₁ (Fig. 33) and is rotated relative to the azimuthally adjacent ones

around the axis of the output segment (Fig. 35). The quantity ξ ₁ is determined from the condition

(Fig. 33). Compliance ratio

provides a uniform distribution in azimuth of N input segments relative to the LO axis (Fig. 33) of the output segment, as well as the joining of adjacent input segments among themselves along the perimeter of the DO (Fig. 35). The input segments are mated to the output by docking their ends, so each of the input segments must have a part of the end face of the DS and AB (Fig. 33), formed by a section by a plane placed at an angle ξ ₁ (Fig. 35) relative to the pipe (segment), and for the input ends to be interconnected, the remaining part BEFORE their ends must be formed by a second section, two crossed at an angle

planes, while the line of intersection of the crossed planes must have an angle ξ ₃ with the axis of the pipe (segment) and coincide with the axis of the output segment, while the value ξ ₃ is determined from the relation

(Fig. 33, Fig. 32). The length of the input segments l _{c2 is} chosen from the condition that the radial dimensions of the mixing chamber are equal to the system of multi-pass screw plasma ducts (Fig. 33, Fig. 32), i.e.

Геометрия камеры смешения существенно отличается от геометрии винтовых плазмоводов, поэтому необходимо осуществить переход от винтовой геометрии плазмоводов с углом захода α_NT к прямолинейной по круглым (разъемным) соединениям.

The geometry of the mixing chamber differs significantly from the geometry of screw plasma ducts, so it is necessary to make the transition from the screw geometry of plasma ducts with an angle of entry α _NT to straight-line along round (detachable) connections.

а также поворот плоскости сечения на угол

т.е. ξ₄ = ξ₁
(фиг. 34) в направлении кручения (левое, правое) системы N многозаходных винтовых плазмоводов в плоскости, перпендикулярной плоскостей осей выходного и входного сегментов и параллельной оси выходного сегмента. Аналогично рассмотренному проводится сечение второго торца входных сегментов на расстоянии l_с2, что обеспечивает стыковку между входными и переходными сегментами. Введение переходных сегментов обеспечивает угловую стыковку по круглому (т.е. разъемному) сечению системы N многозаходных винтовых плазмоводов с камерой смешения. Необходимо отметить, что при реализации угла

в переходных сегментах может произойти нарушение радиальной стыковки, что обусловлено величиной длины l_с3 переходного сечения. Причем при величине

неизбежно произойдет взаимное пересечение винтовых цилиндров. Поэтому в h_т необходимо это учесть путем добавки в h_г величины l_C3/cosα_NT.
Таким образом, требование к длине переходного сегмента можно сформулировать следующим образом: l_cв< l_C3< r_NTcosα_NT, где l_св - минимально возможное расстояние между двумя сварными швами.This transition is possible using a transition segment connecting each input segment of the mixing chamber with a screw plasma duct. Thus, one end of the transition segment connected to the plasma duct is made by a section perpendicular to the axis of the pipe. The second end, connecting with the input segment, must ensure the rotation of the axis of the segment (normal to the end) in two planes (Fig. 33, Fig. 31, Fig. 32) by an angle ξ ₁ (Fig. 33) in the plane of the axes of the output and input segments ( i.e., when the axes of the output and transition segments are parallel in this plane), and the parallelism of the axes ensures the equality of the radial dimensions of the mixing chamber and the system of multi-pass screw plasma ducts, which is realized (see Fig. 33) under the condition

as well as rotation of the section plane by an angle

those. ξ ₄ = ξ ₁
(Fig. 34) in the torsion direction (left, right) of the system N of multi-pass screw plasma ducts in a plane perpendicular to the planes of the axes of the output and input segments and parallel to the axis of the output segment. Similarly to the considered section of the second end of the input segments is carried out at a distance l _s2 , which provides a connection between the input and transition segments. The introduction of transition segments provides an angular docking over a circular (i.e., detachable) section of the system N of multi-pass screw plasma ducts with a mixing chamber. It should be noted that when implementing the angle

in the transition segments, a radial docking violation may occur, which is due to the length l _{c3 of the} transition section. Moreover, at

inevitably there will be a mutual intersection of the screw cylinders. Therefore, in h _t, it is necessary to take this into account by adding l _C3 / cosα _NT to h _g .
Thus, the requirement to the length of the transition segment may be formulated as follows: l _{CB <l C3 <r NT cosα NT} , where _{communication} l - minimum possible distance between the two weldings.

 The execution of the segments of the mixing chamber in this way, in compliance with the above ratios, ensures an optimal transition from the screw design of the plasma-optical system to the mixing chamber with the tightness conditions fulfilled and without violating the profile parameters.

и азимутальным углом

,
с частью периметров торцов N выходных сегментов, образованных сечением плоскостью под углом

.After manufacturing the N input, N transition and one output segments, they are assembled by moving in space by combining the perimeters of the ends of the end segment end, made at an angle

and azimuth angle

,
with a part of the perimeters of the ends of the N output segments formed by a plane section at an angle

.

и

.In this case, the adjacent output segments of the remaining part of the DO end are docked (see Fig. 33 and Fig. 35), made by cross-section with crossed planes at angles

and

.

с торцами N переходных сегментов, выполненных под углами

После сборки осуществляют сварку камеры смешения по внешней поверхности совмещенных торцов. В результате получают конструкцию камеры смешения, которую в свою очередь стыкуют торцами N переходных сегментов с выходными торцами N винтовых плазмоводов, образующих в совокупности с генераторами плазмы устройство для получения покрытий.Then the transition segments are assembled with the system of output and input segments by combining the free ends of the N input segments made at angles

with ends N of transition segments made at angles

After assembly, the mixing chamber is welded on the outer surface of the combined ends. The result is a design of the mixing chamber, which, in turn, is joined by the ends of N transition segments with the output ends of N screw plasma ducts, which together with the plasma generators form a coating device.

 Thus, the considered manufacturing conditions for a screw plasma duct housing, multi-pass spiral plasma duct systems and mixing chambers together ensure the achievement of the set goal - improving the quality of coatings.

Claims (22)

1. A method of producing coatings in vacuum, including the excitation of a vacuum arc of duration τ _g in the vapor of the material of the sacrificial cathode by breaking the interelectrode gap, applying a magnetic field

in the field of generating and transporting a plasma flow, generating a plasma flow with a droplet phase in the cathode spots of the arc, transporting the plasma flow to the condensation region in a radial electric field

bias potential U _{cm t} and in a crossed longitudinal magnetic field

plasmooptical system under the following conditions:
r _i > D _t , r _e <<D _t , ω _e τ _ei > 1,
where r _i , r _e are the Larmor radii of the ion and electron, respectively;
D _t - the transverse size of the transportation zone;
ω _e τ _ei is the Hall parameter of electrons;
ω _e is the cyclotron frequency of the electron;
τ _ei is the time of electron-ion collisions,
separation of the plasma flow from the droplet phase, applying a bias potential to the condensation surface, injecting reactive gas of a component of the coating material, mixing and condensing plasma flows, characterized in that, in order to improve the quality of coatings and expand technological capabilities, the coating composition is determined by the spatial separation of vacuum arcs and regulation of their synchronization mode at the stage of excitation, as well as regulation of the magnitude of the charge of electricity flowing at the stage of generation, and the excitation of vacuum smart arcs are carried out cyclically in time synchronously or sequentially by injection of a laser plasma, while additional ions are separated by transverse momenta, transportation is carried out in spatially separated regions, and their mixing is performed in the field of spatial alignment of the magnetic fields of the transport regions in crossed

- fields after separation of the droplet phase due to multiple collisions of ions with a potential barrier of U _cm under the following conditions:
r _i > D _cm , r _e <D _cm , ω _e τ _ei > 1, λ _ii > D _cm ,
where D _cm is the transverse size of the mixing region;
λ _ii - the length of ion-ion collisions,
the composition of the multicomponent coating A, B, C ... is set from the condition:
N _A / N _B / N _C = K _A t _A • C _{A A} • U _A / K _B t _B S _B U _B / K _C t _{C S B} U _C ,
where N _A , N _B , N _C ... are the concentrations of chemical elements A, B, C ... in a multicomponent coating;
K _A , K _B , K _C. .. - erosion coefficients of plasma-forming substances A, B, C, kg / C;
t _A , t _B , t _C ... - transport coefficients of plasma flows A, B, C, ... C _ON , C _HB and C _NS ...;
U _A , U _B , U _C - values of capacitance (Ф) and voltage changes (V) of capacitive storage, respectively,
moreover, the pulse sequence is set based on the arrangement of the sublattices (layers) of the formed crystalline unit cell of the layered coating material (for example, a superconductor sandwich structure), the number of plasma pulses n _A , n _B ... in the sequence for each sublattice is determined from the expression:

where ρ _{A (B) (C)} is the density of the applied material (chemical element) in the coating, kg / m ³ ;
S _t is the cross-sectional area A (B) (C) of the transportation (mixing) region, mg;
d _{A (B) (C)} is the thickness of the sublattice A (B) (C) in the unit cell, m;
q _{A (B) (C)} - charge flowing in the discharge circuit, C;
and the number of cycles m of the sequence of plasma pulses is determined from the equality d _i • m = l, where d _i is the unit cell size in the direction of coating growth, m; l is the coating thickness, m; during transportation and mixing, the ionic component of the plasma is separated from the droplet phase in the region of the Langmuir layer l _{Langm of} the displacement potential U _cm by repeated collisions with the profiled surface under the condition: P x N _k <1, where P is the probability of droplet passage in the transport region and mixing, N _к - the number of droplets generated per pulse, and the generation, transportation and mixing of plasma flows is carried out for the time τ _g , τ _t and τ _cm , determined from the conditions: τ _g <τ _{tan g} and τ _t <τ _{tan. t} , τ _cm <τ _tan.cm , where τ _tan.g , τ _tan.t , τ _tan.sm - development time of thermal anode instabilities, processes of generation, transportation and mixing of plasma flows, respectively, the magnitude of the electric field E _{t of} the displacement potential U _{cm t is} chosen from the expression:
eU _{see t} + ezU _{see p} <W _def. ,
where W _def. - energy defect formation on the surface of the condensation,
e is the electron charge,
Z is the ionization multiplicity of the ion,
and the magnitude of the bias potential U _cm at condensation is chosen from the expression: eU _{cm t} + ezU _{cm b} <W _def. wherein the condensation of the plasma stream is carried out in a magnetic field H _k , the heterogeneity

which is set from the relation

,
where ΔН _to - the spread of the magnetic field,
Δl is the variation in coating thickness,
l is the thickness of the coating,
by applying a magnetic field to the condensation region and on the back side of the condensation region, and the vacuum arc is excited by defocused laser radiation of duration τ _l under the condition τ _l ≪ τ _g , and the resulting laser plasma flow is directed to the anode region of the discharge and overlap the focus area of laser radiation . 2. The method of producing coatings according to claim 1, characterized in that the generation of plasma from a plasma-forming substance of complex composition is carried out for a time τ _g , the value of which is chosen less than the time τ _{II of the} life of cathode spots of the second kind: τ _g <τ _II , transportation is carried out at the condition L _t <λ _ii is satisfied, and the duration τ _EH of the magnetic

and electric

fields at the stages of transportation and mixing is determined from the condition τ _EH = τ _g + τ _prol , where τ _prol is the time of flight by the plasma of the transport and mixing area. 3. The method of producing coatings according to claims 1 and 2, characterized in that the plasma stream is additionally transported in longitudinal channels, the volume of which is limited by conducting non-magnetic walls, on which the Earth's potential is superimposed, while the length h _{c of the} channels and the distance L _to the condensation surface determined from the condition:
V _∥п • ε _ii <h _c + L _к ,
Where

V _∥п ,
V _{⊥ p} - the longitudinal velocity of the ion in the plasma stream after separation by momentum,
l _c - the transverse size of the walls between the channels,
at the same time, a magnetic field is applied to the separation and condensation zone

parallel to the walls of the channels and perpendicular to the surface of the condensation, and the magnitude of the magnetic field strength H _s is determined from the condition:
r _i > V _⊥ • τ _c ≫r _e ,
where V _⊥ τ _c is the transverse dimension of the channel,

- the time of flight of the plasma separation zone, while the separation of ions by momentum is carried out after mixing the plasma flows. 4. The method of producing coatings according to claims 1 to 3, characterized in that the condensation zone of several longitudinally directed plasma flows is located at a distance selected from the expression L _p∥ = V _∥pl • τ _p , where τ _p is the time of radial density relaxation in total plasma flow. 5. The method of producing coatings according to claims 1 and 2, characterized in that the condensation zone of several radially directed plasma flows is located at a distance equal to L _{k, r} ≥l _Langm from the border of their mutual overlap. 6. The method of producing coatings according to claims 4 and 5, characterized in that the instability of the excitation

when forming longitudinally or radially directed plasma flows, they are set from the condition:

where Nv is the number of excitation pulses, ΔNv = Nv-Nr is the number of failures of excitation of the vacuum arc, Nr is the number of excitation pulses of the vacuum arc. 7. The method of producing coatings according to claims 1 and 2, characterized in that, in order to increase the productivity of the method and reduce energy consumption, a positive potential jump ΔUa is generated in the anode region of the discharge when the plasma flow is generated due to the closure of a part of the magnetic flux F _to this region penetrating the cathode region of the vacuum arc by adjusting the winding density in the cathode region of the discharge or by adjusting the distance between the solenoid of the anode region of the discharge and the solenoid of the transport region, with the rest longitudinal flow Part F _{t (cm)} F = _{SCI (llengm)} + F _cond magnetic field region is partitioned between conveying F _{m (llengm)} (mixture F _{cm (llengm))} in layer l _lengm domain and condensing _cond F according to the expression:

by locking the magnetic field lines to the plasma duct body in the l _langm layer and adjusting its magnitude.  8. The method of producing coatings according to claims 1 and 2, characterized in that the sublattice of complex chemical composition is condensed from several spatially separated vacuum arcs synchronously excited over time, the number of which is determined from the number of chemical elements entering the sublattice, while the reaction gas is poured into the generation region pulsed and synchronously with the excitation of vacuum arcs. 9. A device for producing coatings in vacuum, containing a plasma generator with an ignition unit and a movable consumable cathode located inside the hollow anode and a solenoid covering the anode, while the electrodes of the plasma generator are connected by the discharge circuit to the power supply in the form of a capacitance C _n , as well as plasma duct adjacent the output end of the generator, designed as a hollow body and a coil disposed on the housing, connected to the capacitance C _f bias circuit duct housing, the anode and the plasma duct solenoid form a magnetic mouth system The properties are included in accordance with, as well as a reactive gas inlet system, a mixing chamber and a substrate, characterized in that, in order to improve the quality of multicomponent coatings and expand technological capabilities, it is equipped with additional plasma generators, each of which is connected to the input end of the plasma duct, and the output the end face of each of the plasma ducts is connected to one of the input ends of the mixing chamber, and the output end of the mixing chamber is located in the region of the substrate, and the plasma ducts form the equi system istantnyh multistart screw cylinder diameter D _t, the diameter of the ring formed by the displacement D _m of equidistant helical lines on the cylinder, r _NT radius with an angle of approach α _NT, and maximum distance between the duct h _t determined from the relations:

where N is the number of plasma ducts in the system of equidistant multi-start screw cylinders, E [. ..] is an elliptic integral of the second kind, while the input ends of at least one mixing chamber are a continuation of the plasma duct cylinders, the body of which is formed by combining one end of the direct output cylinder and N input straight or screw hollow cylinders so that the axes of the input cylinders pass along the generatrices of the cone facing the apex to the outlet end of the mixing chamber with the angle α _NC at the apex and the base radius r _NC , determined from the relations:

where h _t is the gap between the plasma ducts, each of the input cylinders being surrounded by a solenoid, and the output cylinder being surrounded by N coaxial solenoids, forming a magnetic system of the mixing chamber, while each of the windings of the solenoid of the input cylinder is connected according to one of the windings of N solenoids of the output cylinder and in the magnetic system of the device, while the profile of the inner surface of the plasma duct body is triangular, with the profile height h _p and its pitch t _p selected from the inequalities: h _p > d _k , l _Langm > h _p > d _k , and the angle α _p at the apex tre golnogo profile chosen from the condition:

where D _t is the transverse size of the transportation zone;
r _to - the radius of the cathode (target);
F is the value of the straight line segment passing through the end working surface at the point D _t / 2 + r _k and the point of tangency to the surface of the plasma duct in the region of the cylinder r _NT ,
moreover, the azimuthal angle β _T between the ends of the plasma duct is selected from the condition:

,
where d _p - the bluntness of the vertices of the triangular profile,
β _then is the maximum angle formed by the vectors drawn at the point of intersection of the outer circle with the tangent to the inner circle of the projection of the screw plasma duct onto a plane perpendicular to the axis of the plasma duct (angular size of one collision zone),
wherein the ignition units of the plasma generators are equipped with a synchronization unit, the control n - channel output of which is connected to each synchronization input of the ignition unit and the reactive gas inlet system.  10. The device for producing coatings according to claim 9, characterized in that the ion separator for transverse impulses is located in the area of the output end of the mixing chamber and consists of N solenoids located on the hollow body, the internal volume of which is divided by the conducting walls into a system of longitudinal channels, each of the separator solenoids is included according to and in series in the magnetic system of the device. 11. The device for producing coatings according to claim 9, characterized in that each input end face of the additional mixing chamber is a continuation of the mixing chamber of the system of equidistant multi-start screw cylinders, while the radius of the base r _{C, N, m of the} cone is selected from the ratio

,
where h _ТМ is the distance between systems of equidistant multi-start screw cylinders, each of which contains N screw cylinders;
m is the number of equidistant multi-helical screw cylinder systems.  12. The device for producing coatings according to claim 9, characterized in that the inner surface of the mixing chamber is profiled similarly to the body of the plasma duct.  13. The device for producing coatings according to claim 9, characterized in that the plasma duct or mixing chamber is provided with at least one additional body made in the form of an equidistant hollow body, the outer and inner surface of which is profiled similarly to the main body.  14. The device for producing coatings according to claims 9 to 11, characterized in that solenoids are located on the rear side of the substrate mutually coaxially with the solenoids of the output cylinder of the mixing chamber, the windings of which are connected in accordance with and sequentially in the magnetic system of the device.  15. The device for producing coatings according to claims 9 to 11, characterized in that it contains at least two similar devices, the output ends of which are located symmetrically to the plane of the substrate, and their magnetic systems are turned on according to the torsion direction of the multi-thread screw plasma ducts opposite.  16. The device for producing coatings according to claim 9, characterized in that it contains several devices, the input ends of which are located on a circle and directed either to the center or from the center of the circle, while the magnetic systems of devices located symmetrically with respect to the center are included according to. 17. The device for producing coatings according to claim 9, characterized in that the ignition unit is made in the form of a laser emitter located on the same optical axis, oriented to the end working surface of the sacrificial cathode, a laser radiation input window and a focusing system in the form of a lens placed between the window input laser radiation and a laser emitter, as well as a diaphragm with a hole located on the same optical axis between the end working surface of the consumable cathode and the input window at a distance from the lens equal to the focus CB distance f of the lens focusing system, wherein the focal length and the magnitude of the distance d from the end working surface of the sacrificial cathode to the lens are related by the inequality f <d <2f, and the diameter D _d orifice aperture is selected from the condition l _lengm> D _d> d _k, wherein d _to - droplet diameter, l _Langm - the magnitude of the Langmuir layer in the field of radiation input, while the normal

to the end working surface of the sacrificial cathode in the region of the optical axis of the ignition unit crosses the inner surface of the anode.  18. The device for producing coatings according to 17, characterized in that the ignition unit is equipped with a transparent shutter movable across the optical axis located in front of the laser radiation input window.  19. The device for producing coatings according to 17, characterized in that the ignition unit is equipped with an electromagnetic system, a pulsed current source and a damper in the form of a lever, one end of which is located in the region of the aperture opening, and the second is located in the region of the laser input window and connected with the armature of an electromagnetic system containing a magnetic circuit and at least one coil connected to a pulsed current source, the input of which is connected to the output of the synchronization unit.  20. The device for producing coatings according to claim 19, characterized in that the electromagnetic system contains two solenoids included in accordance with one of the solenoids included in the pump circuit of the laser emitter of the ignition unit of the plasma generator, and the second solenoid in the discharge circuit of the same plasma generator . 21. An apparatus for producing coatings according to claim 9, characterized in that, in order to increase productivity, the value of the gap l between the ends of _an anode and the plasma duct solenoids chosen from the condition 0 <l _an <2D _and where D _a - diameter of the anode, with this diaphragm with an inner diameter equal to D _{t (cm)} - 2l _Langm and galvanically connected with the housings are placed on the inner surface in the region of the ends of the housings of the plasma ducts and the mixing chamber, and the bias circuit capacitance of the plasma duct is selected from the condition

,
where R _kn - the value of the resistance of the section "body" of the plasma duct - plasma, Ohm;
L _g - the magnitude of the inductance of the discharge circuit, GN;
and the value of the capacitance C _cm - the bias circuit of the housing of the mixing chamber is selected from the condition:

where R _ksp is the resistance value of the section "mixing chamber body - plasma". 22. A method of manufacturing a device for producing coatings in vacuum, containing a screw plasma duct with an external D _in and internal 2r _NT diameters of the cylindrical screw housing of the plasma duct with an azimuth angle of rotation β _T between the ends and the contact angle α _NT of the screw cylinder with a profiled inner surface, including profiling the inner diameter of a straight pipe

cutting a pipe into K segments by successive K sections of a profiled pipe by planes oriented at an angle

relative to the normals to the cylindrical surface, placed at a distance l _to , the assembly of segments by combining their ends due to their rotation through one segment by an angle φ ≥ π, a turn in the diametrical plane of the pipe by an angle

and additional longitudinal movement of each Kth segment in the direction of torsion along the axis of the screw cylinder to a height

and subsequent welding on the outer surface of the ends of the segments of the screw cylinder, and the distance l _to and the angle γ _K between the normals of consecutive K sections of the profiled pipe are selected from the expressions:

23. A method of manufacturing a device for producing coatings according to item 22, wherein the mixing chamber for N plasma flows is made of (2N + 1) segments, N input, N transition and one output segment, while the end of the output segment from the input segments are obtained by a section of a straight profiled pipe N by planes, each of which is inclined to the pipe axis at an angle

and rotated in azimuth relative to adjacent planes at an angle

around the axis of the output segment, the second end of the output segment is formed by a section of a plane perpendicular to the axis of the pipe at a distance l _c1 , determined from the condition:

where L _in - the length of the screw plasma duct;
h _c - the length of the ion separator for transverse momenta,
moreover, part of the ends N of the input segments connecting in the mixing chamber with the output segment is formed by a section of a plane placed at an angle

relative to the axis in the plane of the axes of the output and input segments, and the remaining part of these ends N input segments connecting in the mixing chamber to each other in adjacent segments, is formed by a second cross section of two crossed at an angle

planes, while the intersection lines of the crossed planes form an angle with the axis of the input segments

and coincide with the axis of the output segment, the second end face N of the input segments is formed by a pipe section at a distance

from the first plane at an angle

in the plane of the axes of the output and input segments, as well as at an angle

in a plane perpendicular to the plane of the axes of the output and input segments and parallel to the axis of the output segment in the torsion direction of the multi-helical plasma ducts, and the ends of the transition segments from the input segments are formed by a plane section at an angle

in the plane of the axes of the output and input segments, as well as at an angle

in the plane perpendicular to the plane of the axes of the output and input segments and parallel to the axis of the output segment in the torsion direction of the multi-thread screw plasma ducts, and the second end face N of the transition segments is formed by a section of a plane perpendicular to the pipe axis at a distance l _c3 equal to

and the subsequent assembly of the segments by moving them in space by combining the perimeters of the parts of the end face of the output segment, made at an angle

and azimuth angle

with a part of the perimeters of the ends of the N input segments formed by a plane section at an angle ξ ₁ and ξ ₄ , as well as the combination of the ends of the N input segments made at angles ξ ₁ and ξ ₄ with the ends of the N transition segments made at angles ξ ₁ and ξ _{4 of the} subsequent by welding along the outer surface of the combined ends, as well as by joining the ends of the N transition segments of the mixing chamber with the output ends of the N screw plasma ducts.

Images