Magnetically filtered vacuum-arc plasma deposition systems

This article is a brief historical review of R&D carried out by the KIPT scientists in the field of magnetic filtering of vacuum-arc plasma flows to be applied in thin film deposition technology.

Saved in:
Bibliographic Details
Published in:Вопросы атомной науки и техники
Date:2002
Main Author: Aksenov, I.I.
Format: Article
Language:English
Published: Національний науковий центр «Харківський фізико-технічний інститут» НАН України 2002
Subjects:
Online Access:https://nasplib.isofts.kiev.ua/handle/123456789/79281
Tags: Add Tag
No Tags, Be the first to tag this record!
Journal Title:Digital Library of Periodicals of National Academy of Sciences of Ukraine
Cite this:Magnetically filtered vacuum-arc plasma deposition systems / I.I. Aksenov // Вопросы атомной науки и техники. — 2002. — № 5. — С. 139-141. — Бібліогр.: 26 назв. — англ.

Institution

Digital Library of Periodicals of National Academy of Sciences of Ukraine
id nasplib_isofts_kiev_ua-123456789-79281
record_format dspace
spelling Aksenov, I.I.
2015-03-30T09:23:13Z
2015-03-30T09:23:13Z
2002
Magnetically filtered vacuum-arc plasma deposition systems / I.I. Aksenov // Вопросы атомной науки и техники. — 2002. — № 5. — С. 139-141. — Бібліогр.: 26 назв. — англ.
1562-6016
PACS: 52.77.-j
https://nasplib.isofts.kiev.ua/handle/123456789/79281
This article is a brief historical review of R&D carried out by the KIPT scientists in the field of magnetic filtering of vacuum-arc plasma flows to be applied in thin film deposition technology.
en
Національний науковий центр «Харківський фізико-технічний інститут» НАН України
Вопросы атомной науки и техники
Low temperature plasma and plasma technologies
Magnetically filtered vacuum-arc plasma deposition systems
Article
published earlier
institution Digital Library of Periodicals of National Academy of Sciences of Ukraine
collection DSpace DC
title Magnetically filtered vacuum-arc plasma deposition systems
spellingShingle Magnetically filtered vacuum-arc plasma deposition systems
Aksenov, I.I.
Low temperature plasma and plasma technologies
title_short Magnetically filtered vacuum-arc plasma deposition systems
title_full Magnetically filtered vacuum-arc plasma deposition systems
title_fullStr Magnetically filtered vacuum-arc plasma deposition systems
title_full_unstemmed Magnetically filtered vacuum-arc plasma deposition systems
title_sort magnetically filtered vacuum-arc plasma deposition systems
author Aksenov, I.I.
author_facet Aksenov, I.I.
topic Low temperature plasma and plasma technologies
topic_facet Low temperature plasma and plasma technologies
publishDate 2002
language English
container_title Вопросы атомной науки и техники
publisher Національний науковий центр «Харківський фізико-технічний інститут» НАН України
format Article
description This article is a brief historical review of R&D carried out by the KIPT scientists in the field of magnetic filtering of vacuum-arc plasma flows to be applied in thin film deposition technology.
issn 1562-6016
url https://nasplib.isofts.kiev.ua/handle/123456789/79281
citation_txt Magnetically filtered vacuum-arc plasma deposition systems / I.I. Aksenov // Вопросы атомной науки и техники. — 2002. — № 5. — С. 139-141. — Бібліогр.: 26 назв. — англ.
work_keys_str_mv AT aksenovii magneticallyfilteredvacuumarcplasmadepositionsystems
first_indexed 2025-11-26T00:09:48Z
last_indexed 2025-11-26T00:09:48Z
_version_ 1850591675218395136
fulltext MAGNETICALLY FILTERED VACUUM-ARC PLASMA DEPOSITION SYSTEMS I.I. Aksenov National Science Center “Kharkov Institute of Physics and Technology”, Akademicheskaya Str. 1, Kharkov 61108, Ukraine; phone: (0572) 356452, e-mail: iaksenov@kipt.kharkov.ua This article is a brief historical review of R&D carried out by the KIPT scientists in the field of magnetic filtering of vacuum-arc plasma flows to be applied in thin film deposition technology. PACS: 52.77.-j In spite of that fact, that vacuum-arc coating deposi- tion method is known from 1870’s [1,2] the process of its industrialization started one century later, namely since 1970’s. At that time the group of Ukrainian researchers began its regular investigation in d.c. vacuum arc dis- charge: in Kharkov Institute of Physics and Technology (KIPT) there was developed vacuum-arc method of wear resistant coating deposition based on nitrides of such met- als as Ti, Mo, Zr. There were created also the first indus- trial installations [3]. For a comparatively short time the method have gained wide recognition in tool production. However the presence of macroparticles (MPs) of cathode material in vacuum-arc plasmas retards a wider applica- tion of the method in such fields as optics, microelectron- ics, precision mechanics, medicine. The most effective approach to reduce the MPs con- centration in erosion plasma flows is based on spatial sep- aration of trajectories of MPs and ion species. This can be most easily realized in the plasma source with magnetic focusing of the plasma flow [4]. However, the MPs prob- lem is most effectively solved with use of magnetic plas- ma filters. The first magnetic filter was developed at the KIPT in the mid 1970’s. In 1976 the application was reg- istered for the invention of a plasma filter with a curvilin- ear plasma duct bent as a quarter of torus, as well as with S- and Ω-shaped plasma duct [5] (Fig.1). Having used that filter the Kharkov group obtained the results that stimulated the wide-scale researches in the field of vacu- um-arc synthesis of amorphous carbon (or DLC) films and other high-quality films in many countries of the world. In general terms, the mechanism of filtering plasma in magnetic filters can be described as follows. Between the substrate and the cathode there is a certain barrier in- stalled (Fig.2) [6-8]. Baffles or the walls of the bent tube (plasma duct) can serve as the barrier mentioned. In their motion in straight lines the MPs are confronted by this barrier and fail to arrive at the substrate, while the ion and electron components of the plasma flow, owing to the magnetic field of particular configuration, go round the barrier and reach the substrate. Due to rebound of MPs from the walls many of them reach the filter exit and arrive at the substrate. So, the effi- ciency of filtering is the higher, the longer is the plasma guiding channel, the narrower is it, the greater is the angle of the bend. However, in this case the losses of the plasma flow transported increase, the productivity of the system drops, its complexity and its cost is raised. All attempts to simplify the systems enter in contra- diction with requirements, the fulfillment of which is nec- essary for an efficient filtering of plasma. Considerable efforts were undertaken in many research centers to solve this inconsistent problem. In the KIPT there were devel- oped some versions of magnetic plasma filter over and above those mentioned. The most of them were not pub- lished due to confidentiality of their application in defense industries. So, e. g., the filter with curved plasma duct of so called opened architecture described in ref. [9] had been developed in the mid 1980’s as the unite of the high power set-up intended for high rate coating deposition by the method of electron beam and cathodic vacuum-arc evaporation of metals. But the short communication about this device has been published by its authors just in 2001 [10]. The filters with S- and Ω-shaped plasma duct [5] and the filtering plasma source comprising a cathode with cylindrical lateral working surface (Fig.3) [11] were de- veloped much earlier than their analogues described in ref. [12] and [13] respectively. Fig.1. Curvilinear-filter. Toroidal 900 (a), S-shaped (b) and Ω-shaped plasma ducts Problems of Atomic Science and Technology. 2002. № 5. Series: Plasma Physics (8). P. 139-141 139 (a) (b) (c) mailto:iaksenov@kipt.kharkov.ua An attempt to solve the problem of large square sur- faces treatment by macroparticles-free plasmas was un- dertaken by Sablev et al. [14,15]. But they were failed us- ing the systems in industrial application due to their com- plexity and low degree of plasma filtering. Fig. 2. Rectilinear plasma filters with magnetic “island” (a) and passive flat screen (b) Fig. 3. Filtered plasma source with “dome” type of the magnetic system (prototype) The existing methods of defining the plasma contami- nation with MPs are based on counting the number of MPs in coatings. So, the optimization of filters is rather complicated challenge. The problem is simplified by means of computation of MPs motion in a plasma duct of filters. We solved the problem in two-dimensional ap- proximation for axisymmetric and plane-symmetric sys- tems [16]. The method is rather useful for the compara- tive estimation of filtering properties of the system of dif- ferent geometry. Besides, using this method it’s quite eas- ily to find out that the use of long and narrow plasma guiding ducts which have a lot of knees, is unadvisable approach if MPs are in solid state. The success can be achieved choosing appropriate systems of baffles (ribs) and their arrangement inside the duct of simple shape [17]. Another important characteristic of the filtering sys- tem is its capacity. Analysis of plasma motion along a toroidal magnetic field shows that for a successful plasma travel the field intensity must be above 1 T. At this field (i) it is impossible to provide a stable burning of the d.c. arc discharge in the system and (ii) the plasma injection into a field of such strength is practically impossible too. Thus, it was reasonable to consider plasma motion in con- dition of partly magnetized plasma [18]. In this case plas- ma has a very high resistivity across the magnetic field and very high conductivity along the field. Electrons move along the central lines of magnetic field, crossing the cathode. And ions follow the electrons to conserve plasma quazineutrality. This model is based on the Moro- zov’s plasma optics principles [19]. Later it was investi- gated in details by Boercker et al. [20]. They showed that this model, which is known as the “flux-tube model” is quite useful for qualitative estimation of plasma motion in curved duct, but it is much less advisable to be used for quantitative calculation. There were developed more per- fect theoretical models [21-23]. These models describe mechanism of plasma losses caused by plasma diffusion across magnetic fields. They predict displacements of plasma flow due to gradient and centrifugal drift. But these models are developed for ideal toroidal fields, so they are quite useless in cases of distorted systems. Here the drift shift of plasma flow may be directed toward the side, which is not predicted by the theory. On the con- trary, the very first model developed [18] explains these phenomena quite simply: the plasma moves along the flux of magnetic lines crossing the cathode active surface where the plasma flow arises. So, one can direct the plas- ma in any run choosing an appropriate geometry of mag- netic lines. Thus, we have rather effective means, which are very helpful and useful for optimizing and designing magnetic plasma filters. Fig. 4. Filtered plasma source with linear-to-radial transformation of plasma flows. Schematic (a), magnetic field configurations (b, c) (a) (b) (c) The development of high-efficiency plasma filter is the key task for opening the way to a wide-scale applica- tion of the vacuum-arc method for high technology prac- tice. There are created a great number of original design structures for to solve this problem [24]. However, none of them so far has an appropriate efficiency. The plasma throughput is not more then 25…50%. From this view- point the recent developments of KIPT seam to be promising. One of them is the system with radial flow of filtered plasma [25] (Fig. 4). It comprises two similar vac- uum-arc magnetic-focusing plasma sources. They ar- ranged coaxially, facing each other. Axial flows of plas- ma here are transformed into radial flow moving through the annular gap between the anodes of the sources. The gap is wide compared to the length of the plasma path to- ward the exit. So, the diffusion to the walls across the magnetic field is not intensive. The gradient and centrifu- gal drifts in the annular gap are closed, so the plasma drift losses are negligible. As a result the total losses are very small. The transmittance is very high: plasma throughput is about 90%. The system coefficient is about 8,5%. None of other system has such a high characteristic. Besides, the system with radial flow has a very large area of fil- tered plasma flow cross section. It is very useful for rect- angular variant of the system (Fig. 5). This version is sup- posed to be used for deposition of coatings on flat sub- strates of large area and also on web materials. Fig. 5. Source of double filtered plasma flows Another variant of plasma filtering system, which is developed in KIPT recently, is the filter with an L-shaped duct [26]. Its transportation efficiency is about 65%, i. e. less than for previous “radial” variant. It is likely due to the presence of gradient and centrifugal losses because of the system is asymmetric. But 65% is much higher as compared to convenient systems. The L-shaped filter is used presently in industrial pro- duction of storage system elements for deposition of ul- tra-thin protecting DLC films. This is a promising step for vacuum-arc method of coating deposition in high tech- nologies application. REFERENCES 1) D. M. Mattox, The History of Vacuum Coating Technology, Albuquerque, 2002, p.p. 22, 23. 2) R. L. Boxman. Proc. of the XIXth Int. Symp. on Discharge and Electrical Insulation in Vac. (IS- DEIV), Xi’an, China, Sept., 2000, p.1. 3) I. I. Aksenov, A. A. Andreev. Problems of Atom- ic Sci. and Tech. Series: Plasma Physics, 3 (3) 4 (4) (1999) 242. 4) I. I. Aksenov (Axenov) et al. US Patent No. 4,551,221, Nov. 5, 1985 (Rus.). 5) I. I. Aksenov, V. A. Belous, V. G. Padalka, USSR AS No. 605425, Oct. 26, 1976 (Rus.). 6) I. I. Aksenov (Axenov) et al., Canadian Pat. No. 1176599. 7) I. I. Aksenov et al., USSR AS No. 1708133, 08.06.1990 (Rus.). 8) I. I. Aksenov, A. I. Timoshenko, V. M. Khoroshikh. USSR AS No. 1584727, 27.03.1989 (Rus.). 9) M. Kühn, P. Maja and F. Richter, Diamond and Related Materials, 2 (1993) 1350-1354. 10) V. A. Belous, V. I. Safonov, G. N. Kartmazov, Proc. of the 4th Int. Symp. “Vacuum Technolo- gies and Equipment”, Kharkov, Apr. 23-27, 2001, p. 312. 11) I. I. Aksenov, V. G. Bren, USSR AS, No. 913744, Nov. 16, 1981. 12) S. Anders et al., Proc. of the XVIIth ISDEIV, Berkeley, CA, 1996, p.p. 904-908. 13) D. M. Sanders, S. Falabela, US Patent No. 5, 282, 944 Feb. 1, 1994. 14) V. A. Osipov et al. Pribory i Tekhnika Oeksperi- menta, 6 (1978) 173(Rus.). 15) G. V. Kliuchko et al. US Pat. No. 4492845, Jan. 8, 1985. 16) I. I. Aksenov, D. Yu. Zaleskij, V. E. Strel’nitskij, 1st Int. Congr. on Radiation Physics, High Current Electronics and Modification of Materials, Sept. 2000, Tomsk, Russia. Proc. Vol. 3, p.p. 130-138. 17) I. I. Aksenov, V. E. Strel’nitskij. Kharkov Sci. Assambly ISDF-5, Kharkov, Ukraine, 2002, Proc. p. 39-64. 18) I. I. Aksenov et al. Plasma Physics and Con- trolled Fusion. 28, 5 (1986) 761. 19) A. I. Morozov. Dokl. Akad. Nauk SSSR, 163 (1965) 1363 (Rus.). 20) D. B. Boercker et al., J. Appl. Phys., 69 (1) (1991) 115. 21) B. Altercop et al., J. Appl. Phys., 79 (9) (1996) 6791. 22) Xu Shi et al., IEEE Trans. on Pl. Sci. 24 (6) (1996) 1309. 23) A. Anders, S. Anders, I. Brown. SSST, 4 (1995) 1-12. 24) A. Anders, Surf. and Coat. Techn., 120-121 (1999) 319-330. 25) I. I. Aksenov. Proc. 4th Int. Symp. Vac. Tech. and Equip., Kharkov, Ukraine, 2001, p. 139-145. 26) I. I. Aksenov et al. Surface and Coatings Technol- ogy, 2002 (to be published). 141 I.I. Aksenov Akademicheskaya Str. 1, Kharkov 61108, Ukraine; phone: (0572) 356452, e-mail: iaksenov@kipt.kharkov.ua References