992 ieee transactions on plasma science, …staff.ustc.edu.cn/~fuzp/course/paper/review of...

992 IEEE TRANSACTIONS ON PLASMA SCIENCE, VOL. 35, NO. 4, AUGUST 2007

Review of Cathodic Arc Deposition forPreparing Droplet-Free Thin Films

Hirofumi Takikawa, Member, IEEE, and Hideto Tanoue

Abstract—Cathodic arc plasma is one of the potential ion plat-ing physical vapor deposition methods to prepare protective coat-ings on cutting tools, metal mold, etc. In particular, TiN, CrN, andTiAlN films are coated on industrial cutting tools using cathodicarc plasma. However, the cathode spot of the vacuum arc generatesmacrodroplets as coproducts of cathodic arc plasma containinghigh-energy ions. These macrodroplets may pose problems withsmooth-surface films that are used for advanced high-precision ap-plications. This paper reviews cathode phenomena particularly fora graphite cathode, the techniques for reduction of macrodropletgeneration, and the techniques for macrodroplet removal from theprocessing plasma. The reduction technique includes steered arc,pulsed arcs, etc. The removal technique includes shielded arcs andfiltered arcs. Recent filtered arc deposition systems are referred.

Index Terms—Cathodic arc, droplet suppression, filtered arc,macrodroplet, thin film deposition.

I. INTRODUCTION

ONE INDUSTRIAL application of vacuum arc dischargeis coating technology. When an arc discharge is generated

under medium and higher vacuum, in general, a cathode spotis formed, but no anode spot is formed. The cathode spot isvery active with high temperature and evaporates the cathodematerial. At the cathode spot region, very dense plasma isgenerated, and the evaporated cathode material is ionized andthe ions deposit solid film upon reaching the solid surface. Suchvacuum arc deposition is a major method in physical vapordepositions (PVDs) and has the advantage of higher ion energy,compared to the other PVD methods. The ion source of the arccathode is generally solid. Therefore, no crucible is needed, andthe sources can be freely mounted on the wall of the processchamber. In principle, no gas introduction is necessary. How-ever, vacuum arc deposition is suitable for reactive depositionbecause the ions generated in the vacuum are highly reactivewith the gases due to their high energy.

Manuscript received July 10, 2006; revised March 21, 2007. This work wassupported in part by Nissin Electric Co. Ltd., Itoh Optical Industrial Co. Ltd.,Fukoku Co. Ltd., ShinMeiwa Industries, Ltd., Toyo Tanso Co. Ltd., NipponCarbon Co. Ltd., Ferrotec Corporation, Onward Ceramic Coating Co. Ltd., andKurita Seisakusho Co. Ltd., by the Excellent Research Project of the ResearchCenter for Future Technology, Toyohashi University of Technology, by theResearch Project of the Venture Business Laboratory, Toyohashi University ofTechnology, by the 21st Century COE Program “Intelligent Human Sensing”from the Ministry of Education, Culture, Sports, Science and Technology, by aGrant in Aid from the Japan Society for the Promotion of Science (JSPS), andby a JSPS Core University Program.

The authors are with the Department of Electrical and Electronic Engineer-ing, Toyohashi University of Technology, Toyohashi 441-8580, Japan (e-mail:[email protected]).

Color versions of one or more of the figures in this paper are available onlineat http://ieeexplore.ieee.org.

Digital Object Identifier 10.1109/TPS.2007.897907

Fig. 1. Image of droplet emission from metal cathode spot.

Fig. 2. Graphite cathode spot and image of droplet emission. (a) Graphitecathode spot (dc 50 A, no gas introduction). (b) Image of droplet emission ofgraphite cathode.

Vacuum arc deposition is also called cathodic arc depositionor arc ion plating, and various types of industrial systemsare in widespread use around the world. Its major use is tomake protective coatings on cutting/machining tools and slidingmaterials. Examples of film materials are TiN, TiC, CrN, andTiAlN. Such technology has long been investigated, and therelevant information is available in books and the literature[1]–[5]. Nowadays, the introduction of basic-type vacuumarc deposition apparatus (namely, nonfiltered arc depositionapparatus) has almost reached the saturation point, and the

0093-3813/$25.00 © 2007 IEEE

TAKIKAWA AND TANOUE: REVIEW OF CATHODIC ARC DEPOSITION 993

Fig. 3. SEM micrographs of filings of graphite cathode and droplets (arc current, 50 A; pressure, less than 0.01 Pa: no gas introduction). (a) Raw graphite.(b) Droplet type A: graphite fragment. (c) Droplet type B: molten fragment. (d) Droplet type C: flakes without nanotubes. (e) Droplet type D: with MWCNTs.

future development of high added-value film formation is muchawaited.

The principal problem involved in the vacuum arc depositionis the macrodroplet (hereafter, droplet) that is coemitted fromthe cathode spot. The droplets adhere to the film in prepa-ration and roughen its surface, causing deterioration in thecomposition uniformity and the exfoliation of the film. Thisis why adherence of droplets to the film must be avoided.High-value feature applications include diamondlike carbon(DLC) film (in wide field) [6]–[8], transparent conductive oxidefilms [9]–[11], [43]/large-scale integrated wiring (electrical andelectronic applications), and new composite nitride/oxide films(in mechanical applications, among others) [12], [13]. Theexpectations for hydrogen-free tetrahedral DLC in particularare growing yearly.

Fig. 4. Raman spectrum of graphite-cathode droplet that was collected fromthe source chamber in T-FAD.


Fig. 5. Steered arc. (a) Magnetic field applied to the cathode. (b) Electric field distribution of the vacuum arc and retrograde motion of cathode spot. (c) Samplephotograph of steered Cu cathode spot.

The measures to overcome the droplet problem in cathodicvacuum arc may be roughly classified into two categories,namely: 1) a method to suppress droplet generation and2) a method to prevent the droplets from reaching the substrate.In this paper, the cathode phenomena and droplet emission arereviewed first, and a recent unique result on carbon nanotubesformation at cathode graphite is introduced, which has interest-ing possibilities as a new nanotechnology in material scienceand engineering. Then, the methods to overcome the dropletproblem are reviewed, and recent filtered systems designedparticularly for preparing hydrogen-free tetrahedral amorphouscarbon (ta-C) film and multielement-metallic compound filmsare introduced.

II. DROPLET EMISSION FROM CATHODE SPOT

Droplet emission phenomena from cathode spot in vacuumarc have also been studied for some time [3]–[5], [14]. Fig. 1provides a general image of droplet emission from a metalcathode. The cathode spot has a molten layer interface betweenthe cathode and plasma. Most of the droplets are considered tobe emitted from the lip of the molten layer, and some dropletsare emitted in the perpendicular direction. Droplets emittedfrom the metal cathode are usually in the molten state.

Only a graphite cathode emits visually observable droplets,as shown in Fig. 2(a). Droplets from other metal cathodesare not visually observable since their temperature is not highenough to emit radiation. The droplets themselves emitted

from the graphite cathode are interesting to observe. SEMmicrographs of the droplets that were collected from the sourcechamber in T-shape filtered arc deposition system (T-FAD,described later) are shown in Fig. 3. Some of the dropletsare graphite fragments without a sign of having once melted(Type A). Some of the droplets have a shape that has oncemelted (Type B). Most of the droplets from the graphite cath-ode are flakelike as shown in Fig. 3(d) (Type C). They areconsidered to have originated and exfoliated from the bottomand rim of the cathode spot. On the surface of some droplets(Type D), multiwalled carbon nanotubes (MWCNTs) are ob-served, which can reportedly be generated in the arc spot, andthere is a transformed interface layer on the arc spot surface[15]–[18]. Based on these and previous results, an illustration ofa graphite cathode is shown in Fig. 2(b). The alphabetical lettersin Fig. 2(b) correspond to the respective SEM micrographsin Fig. 3.

Fig. 4 shows the Raman spectra of a graphite-cathode dropletcollected from the source chamber in T-FAD as well as the rawgraphite of the cathode. The Raman spectrum of the dropletshows wider G and G′-peaks and a stronger D-peak, comparedto graphite. This indicates that most droplets are somewhatamorphous, corresponding to their surface morphology.

III. REDUCTION OF DROPLET EMISSION

Several methods that are used to reduce the droplet genera-tion are described in Sections III-A–D.


Fig. 6. Various types of methods to reduce droplet generation. (a) Distributed arc using electron beam. (b) Distributed arc using preionized gas flow.(c) Shunting arc.

A. Steered Arc [19]–[21]

A steered arc is commonly used. As shown in Fig. 5(a), amagnetic field usually with a lateral component is applied onthe cathode surface with permanent magnets or electromagnetcoils. The cathode spot is driven by this magnetic field in theretrograde direction as expected by Fleming’s law, due to theion motion in the potential hump formed in front of the cathodespot, as shown in Fig. 5(b). As an example, the copper (Cu)cathode spot motion is shown in Fig. 5(c), evidencing a circulartrace. The concept of the steered arc is to avoid overheating ofthe cathode spot by keeping it at one position, as well as toregulate the uniform erosion.

B. Current-Controlled Arc [22]–[24]

In order to reduce the droplet generation due to cathodespot overheating, there is a method to reduce the arc cur-rent before the cathode spot becomes hot, i.e., using the dcpulse. The generation of large-size droplets can thus be sup-pressed. However, the total effect of droplet suppression is notenough.

C. Distributed Arc [25]–[27]

This is a method that uses a heated cathode by means ofresistive heating, RF heating, or electron beam heating, asshown in Fig. 6(a). In this case, a larger amount of electronscan be easily emitted, and a larger cathode spot can be formed,and/or the motion speed of the cathode spot can be fast. Thus,

Fig. 7. Shielded arcs. (a) Simple shielded arc. (b) Superconductor shieldedarc under magnetic field.

the droplet generation is suppressed. A similar situation re-portedly occurs when preionized gas is introduced through thecathode [27].


Fig. 8. Various types of filter systems. (a) Rectilinear. (b) Bent. (c) Rectangular. (d) Knee. (e) Torus. (f) S-shape. (g) Off-plane double bend. (h) Dome.(i) Venetian blind. (j) Co-axial (pulse). (k) Electrostatic filter with laser trigger (pulse). (l) Mechanical pulse.

D. Shunting Arc [28]

When a pulsed high current passes through the material, itssurface is heated by the skin effect and then the surface isevaporated. As the current still increases, the main current turnsinto vapor and the arc discharge forms through the vapor. Thisis called a shunting arc, and it can evaporate even silicon.

IV. REMOVAL OF DROPLETS FROM PLASMA

In the method described in the previous section, the dropletelimination performance is not sufficient. A better way to obtainsmoother film is to prepare the film out of the line of sightfrom the cathode. Two methods are available. The shieldedarc method is simpler. The droplets are physically blocked bythe shield plate in front of the substrate, as shown in Fig. 7[29]–[31]. In order to transport the plasma behind the shield, amagnetic field can be applied and a superconductor shield canbe used [32].

An even more effective method is the magnetically filteredarc, which was first reported in 1978 [33], [34]. In most cases,the plasma is transported from the cathode to the substrate, andthe droplets are eliminated by the plasma transportation wall.Many review papers of the filtered arc system and technologyhave been reported [35]–[43]. The filtered arc system is var-

Fig. 9. Photograph of 120◦ filter with high-current arc source.

iously called filtered arc deposition (FAD), filtered cathodicvacuum arc (FCVA), filtered cathodic arc, or filtered vacuumarc. A typical filtered arc system with its different electromag-netic plasma transportation duct or droplet filter configurationsis shown in Fig. 8(a)–(h). Electromagnetic coils transportingplasma in the out of line of sight direction can be positionedin the chamber, instead of placing them outside of the filterduct. The off-plane double bend filter [44], [45] is nicknamedFCVA and is now commercially available. Most FAD unitshave electromagnetic coils outside of the plasma duct andhave baffles inside the duct wall. However, some types havefreestanding coils inside the plasma duct or the chamber.


Fig. 10. Schematic sketches of new FADs. (a) T-FAD. (b) Crank-FAD. (c) X-FAD. (d) Y-FAD.

Other interesting filters have been developed. Examples areshown in Fig. 8(i)–(l). In the Venetian-blind filter, the plasmapasses between the vane lamellae, and the droplets are caught orreflected by the lamellae [46], [47]. A coaxial filter is operatedwith a large current pulse, and the plasma is driven by a self-magnetic field [48], which can be applied to an ICF70 flangeand is also commercially available in Japan. An electrostaticfilter can be used with a pulsed arc having a laser trigger[49]. However, recently only the laser triggered arc is usedwithout the electrostatic filter. Mechanical filters can be usedin a pulse arc [38], which may also be used in pulsed laserdeposition [50].

V. NEW FILTERED ARC SYSTEMS

In this section, recent innovative filtered arc systems aredescribed. Fig. 9 shows the 120◦ bend filter, which has been de-veloped for preparing ta-C among DLCs in order to deposit theprotective top-layer coat on a next-generation hard-disk systemby IBM group [51], [52]. The vacuum arc plasma is operatedin the pulsed high-current mode [53], [54]. In this system,the generation of droplets is suppressed by rapidly choppingthe current, and the generated droplets are trapped with manybaffles placed on the inner wall of the duct during the plasmatransport to the exit of the 120◦ bent duct. Graphite droplets arenot easy to filter by using conventional filter system, since thesolid droplets bounce off the duct wall.

Recently, it has been demonstrated that hydrogen-free ta-Cfilm is very available as a protective coating on cutting tools foraluminum dry machining [55]. Cathodic arc deposition, more-over, is the only promising method to prepare the hydrogen-free ta-C on an industrial scale. For longer tool life, thicker

ta-C film is required. Although the DLC, even hydrogen-freeDLC, can be prepared on flexible rubber substrates by filteredarc [56], [57], it is well known that it is difficult for thickta-C film to adhere to solid materials, due to its high elasticmodulus. Therefore, a bonding layer is sometimes introducedbetween the substrate and the DLC film. In order to preparethick ta-C with a bonding interlayer, an X-shape filtered arcdeposition system (X-FAD) has been designed and developed[58], [59]. This is an integrated system combining T-FAD [60],[61] and crank-FAD. Fig. 10(a)–(c) shows schematic sketchesof T-FAD, crank-FAD, and X-FAD, respectively. T-FAD hasbeen designed particularly for DLC preparation using graphitecathode with the idea that the transportation direction of plasmaand the trajectory of droplets are separate from each other, usingT-shape filter duct. In T-FAD, the carbon plasma generatedbetween the cathode and the anode is transported with a 90◦

bend under a magnetic field at the connection point of the duct,although the droplet travels in a straight direction away fromthe bent plasma. Crank-FAD as a metal arc evaporator has beendesigned in order to connect to T-FAD, providing it with adroplet-filtering function. In crank-FAD, the plasma is slightlyshifted at the crank point, and the metal droplets in moltenform are caught at the wall opposite to the metal cathode. InX-FAD, T-FAD and crank-FAD share a part of an X-shapeplasma transportation duct. A ta-C film that is more than 1 µmthick was prepared with a Cr interlayer [59].

Fig. 10(d) shows another filtered arc system with Y-shapeplasma transporting duct (Y-FAD). The two cathodic arcsources are connected to the Y-duct, which has been developedfor two purposes. One provides high-speed deposition withtwo of the same cathodes. The other is to prepare multimetalcomposite film or multilayer film using two different cathodes.


Fig. 11. Photographs of the X-FAD and Y-FAD prototypes. (a) X-FAD.(b) Y-FAD.

Two plasmas are not easily combined, since their electricalpolarities are the same and they tend to remain separate fromeach other. Therefore, in order to mix the two plasma beams,a magnetic mixer is placed downstream of the Y-duct. Tomake the film by vertical incident ions, two aligner coils arelocated in front of and behind the substrate. More details of thissystem will be reported elsewhere after further experimentalexamination.

Fig. 11 shows photographs of the X-FAD and Y-FADprototypes.

VI. CONCLUSION

Cathodic arc deposition is considered to be a promising drydeposition method. DLC and oxide films that are deposited bythe cathodic arc are drawing interest in different applications invarious fields. However, the droplet problem is always involvedin the method and prevents its widespread use as a sputteringsystem. This paper reviewed the droplet emission problem,particularly in case of graphite cathode, and presented a wayto overcome it in cathodic arc deposition. The most effectiveway in industrial use might be the filtered arc approach. Manybreakthrough systems have been developed. Further develop-ment of new techniques and sophisticated systems of cathodicvacuum arc deposition can be expected to provide new andhighly functional films in the near future.

ACKNOWLEDGMENT

The authors would like to thank H. Hikosaka, a Master coursestudent, for his assistance in the preparation of this paper.

REFERENCES

[1] H. Randhawa, “Cathodic arc plasma deposition technology,” Thin SolidFilms, vol. 167, no. 1/2, pp. 175–185, Dec. 1988.

[2] R. L. Boxman, D. M. Sanders, and P. J. Martin, Eds., Handbook ofVacuum Arc Science and Technology, Fundamentals and Applications.Park Ridge, NJ: Noyes, 1995.

[3] R. L. Boxman and S. Goldsmith, “Macroparticle contamination in ca-thodic arc coatings: Generation, transport and control,” Surf. Coat. Tech-nol., vol. 52, no. 1, pp. 39–50, Mar. 1992.

[4] D. M. Sanders, “Review of ion-based coating processes derived from thecathodic arc,” J. Vac. Sci. Technol. A, Vac. Surf. Films, vol. 7, no. 3,pp. 2339–2345, May 1989.

[5] D. M. Sanders, D. B. Boercker, and S. Falabella, “Coating technologybased on the vacuum arc: A review,” IEEE Trans. Plasma Sci., vol. 18,no. 6, pp. 883–894, Dec. 1990.

[6] A. Grill, “Diamond-like carbon: State of the art,” Diam. Relat. Mater.,vol. 8, no. 2, pp. 428–434, Mar. 1999.

[7] J. Robertson, “Diamond-like amorphous carbon,” Mater. Sci. Eng. R,vol. 37, no. 4, pp. 129–281, May 2002.

[8] A. C. Ferrari and J. Robertson, “Interpretation of Raman spectra of dis-ordered and amorphous carbon,” Phys. Rev. B, Condens. Matter, vol. 61,no. 20, pp. 14 095–14 107, May 2000.

[9] R. Miyano, K. Kimura, K. Izumi, H. Takikawa, and T. Sakakibara, “Prepa-ration of metal nitride and oxide thin films using shielded reactive vacuumarc deposition,” Vacuum, vol. 59, no. 1, pp. 159–167, Oct. 2000.

[10] V. Z. Zhitomirsky, R. L. Boxman, and S. Goldsmith, “Plasma distribu-tion and SnO2 coating deposition using a rectangular filtered vacuumarc plasma source,” Surf. Coat. Technol., vol. 185, no. 1, pp. 1–11,Jul. 2004.

[11] H. Takikawa, K. Kimura, R. Miyano, and T. Sakakibara, “ZnO film for-mation using a steered and shielded reactive vacuum arc deposition,” ThinSolid Films, vol. 377/378, pp. 74–80, 2000.

[12] A. Bendavid, P. J. Martin, and H. Takikawa, “The properties of nanocom-posite aluminium-silicon based thin films deposited by filtered arc depo-sition,” Thin Solid Films, vol. 420/421, pp. 83–88, 2002.

[13] P. J. Martin, A. Bendavid, J. M. Cairney, and M. Hoffman, “Nanocom-posite Ti–Si–N, Zr–Si–N, Ti–Al–Si–N, Ti–Al–V–Si–N thin film coatingsdeposited by vacuum arc deposition,” Surf. Coat. Technol., vol. 200, no. 7,pp. 2228–2235, Dec. 2005.

[14] J. E. Daalder, “Cathode spots and vacuum arcs,” Physica, vol. 104C,no. 1/2, pp. 91–106, 1981.

[15] H. Takikawa, M. Yatsuki, O. Kusano, and T. Sakakibara, “Carbon nan-otubes fabricated at the cathode spot in a vacuum arc,” Trans. Inst. Electr.Eng. Jpn., vol. 119-A, no. 8/9, pp. 1156–1157, 1999.

[16] H. Takikawa, M. Yatsuki, T. Sakakibara, and S. Itoh, “Carbon nanotubesin cathodic vacuum arc discharge,” J. Phys. D, Appl. Phys., vol. 33, no. 7,pp. 826–830, Apr. 2000.

[17] H. Takikawa, Y. Tao, R. Miyano, T. Sakakibara, X. Zhao, and Y. Ando,“Formation and deformation of multiwall carbon nanotubes in arc dis-charge,” Jpn. J. Appl. Phys., vol. 40, no. 5A, pp. 3414–3418, May 2001.

[18] H. Takikawa, M. Kato, Y. Hibi, T. Sakakibara, T. Tahara, and S. Itoh,“Transformation of graphite into multi-walled carbon nanotubes by ACtorch arc,” Phys. B, vol. 323, no. 1–4, pp. 287–289, 2002.

[19] S. Ramalingam, “Controlled vacuum arc material deposition, method andapparatus,” Int. Patent 85/03954, Feb. 27, 1985.

[20] P. D. Swift, “Macroparticles in films deposited by steered cathodic arc,”J. Phys. D, Appl. Phys., vol. 29, no. 7, pp. 2025–2031, Jul. 1996.

[21] R. Miyano, H. Takikawa, T. Sakakibara, and Y. Suzuki, “Retrogrademotion velocity of graphite cathode spot in vacuum arc deposition appara-tus,” Trans. Inst. Electr. Eng. Jpn., vol. A-114, no. 2, pp. 117–122, 1994.[in Japanese].

[22] K. Keutel, H. Fuchs, H. Mecke, and C. Edelmann, “Modified pulse arcdeposition for reducing droplet emission,” in Proc. IEEE 18th ISDEIV,1998, pp. 562–565.

[23] M. Ellrodt and H. Mecke, “Possibilities of influencing vacuum arc deposi-tion by time-dependent arc current,” Surf. Coat. Technol., vol. 185, no. 1,pp. 1–11, 2004.

[24] K. Kimura, H. Takikawa, and T. Sakakibara, “Discharge characteris-tics of vacuum arc with pulse current and efficiency of macroparti-cle suppression,” J. Inst. Appl. Plasma Sci., vol. 7, pp. 3–10, 1999.[in Japanese].


[25] A. I. Vasin, A. M. Dorodnov, and V. A. Petrosov, “Vacuum arc with adistributed discharge on an expandable cathode,” Sov. Tech. Phys. Lett.,vol. 5, no. 12, pp. 634–636, 1979.

[26] V. Belevsky and V. Kopetenko, Pribori i Technika Eksperimenta, vol. 4,pp. 188–190, 1970.

[27] H. Shinno, M. Fukutomi, M. Fujitsuka, and M. Okada, “In situ coating oflow-Z materials by reactive vacuum arc-deposition with a stabilized arccathode,” J. Nucl. Mater., vol. 133/134, pp. 749–753, 1985.

[28] K. Yukimura, “Shunting arc generation of a heated wire in a low pressuregas—Fundamental research of pulsed ion source,” Trans. Inst. Electr. Eng.Jpn., vol. A-118, no. 9, pp. 954–958, 1998. [in Japanese].

[29] H. Brandolf, “Vapor deposition apparatus and method,” U.S.Patent 4 511 593, Apr. 16, 1985.

[30] Y. Taki, T. Kitagawa, and O. Takaki, “Preparation of diamond-like carbonthin films by shielded arc ion plating,” J. Mater. Sci. Lett., vol. 16, no. 7,pp. 553–556, Apr. 1997.

[31] I. I. Aksenov, V. A. Belous, V. V. Vasil’ev, Y. Y. Volkov, andV. E. Strel’nitskij, “A rectilinear plasma filtering system for vacuum-arcdeposition of diamond-like carbon coatings,” Diam. Relat. Mater., vol. 8,no. 2–5, pp. 468–471, Mar. 1999.

[32] H. Takikawa, N. Miyakawa, and T. Sakakibara, “Development of shieldedcathodic arc deposition with a superconductor shield,” Surf. Coat. Tech-nol., vol. 171, no. 1–3, pp. 162–166, 2003.

[33] I. I. Aksenov, V. A. Belous, V. G. Padalka, and V. M. Khoroshikh, “Appa-ratus to rid the plasma of a vacuum arc of macroparticles,” Instrum. Exp.Tech., vol. 21, no. 5, pp. 1416–1418, 1978.

[34] I. I. Aksenov, V. A. Belous, V. G. Padalka, and V. M. Khoroshikh, “Trans-port of plasma streams in a curvilinear plasma-optics system,” J. PlasmaPhys., vol. 4, no. 4, pp. 425–428, Jul./Aug. 1978.

[35] R. L. Boxman, V. Zhitomirsky, B. Alterkop, E. Gidalevich, I. Beilis,M. Keidar, and S. Goldsmith, “Recent progress in filtered vacuum arcdeposition,” Surf. Coat. Technol., vol. 86/87, no. 1–3, pp. 243–253, 1996.

[36] D. A. Karpov, “Cathodic arc sources and macroparticle filtering,” Surf.Coat. Technol., vol. 96, no. 1, pp. 22–33, Nov. 1997.

[37] P. J. Martin, A. Bendavid, and H. Takikawa, “Ionized plasma vapor depo-sition and filtered arc deposition; processes, properties and applications,”J. Vac. Sci. Technol. A, Vac. Surf. Films, vol. 17, no. 4, pp. 2351–2359,Jul. 1999.

[38] A. Anders, “Approaches to rid cathodic arc plasmas of macro andnanoparticles: A review,” Surf. Coat. Technol., vol. 120/121, pp. 319–330,1999.

[39] D. M. Sanders and A. Anders, “Review of cathodic arc deposition technol-ogy at the start of the new millennium,” Surf. Coat. Technol., vol. 133/134,pp. 78–90, 2000.

[40] P. J. Martin and A. Bendavid, “The filtered arc process and materialsdeposition,” Surf. Coat. Technol., vol. 142–144, pp. 7–10, 2001.

[41] A. Anders, “Energetic deposition using filtered cathodic arc plasmas,”Vacuum, vol. 67, no. 3/4, pp. 673–686, Sep. 2002.

[42] I. I. Aksenov, V. E. Strel’nitskij, V. V. Vasilyev, and D. Y. Zaleslij, “Ef-ficiency of magnetic plasma filters,” Surf. Coat. Technol., vol. 163/164,pp. 118–127, 2003.

[43] P. J. Martin and A. Bendavid, “Review of the filtered vacuum arc processand materials deposition,” Thin Solid Films, vol. 394, no. 1/2, pp. 1–5,Aug. 2001.

[44] X. Shi, B. K. Tay, H. S. Tan, E. Liu, J. Shi, L. K. Cheah, and X. Jin,“Transport of vacuum arc plasma through an off-plane double bend filter-ing duct,” Thin Solid Films, vol. 345, no. 1, pp. 1–6, May 1999.

[45] B. K. Tay, Z. W. Zhao, and D. H. C. Chua, “Review of metal oxide filmsdeposited by filtered cathodic vacuum arc technique,” Mater. Sci. Eng. R,vol. 52, no. 1–3, pp. 1–48, May 2006.

[46] A. I. Ryabchikov and I. B. Stepanov, “Investigations of formingmetal-plasma flows filtered from microparticle fraction in vacuumarc evaporation,” Rev. Sci. Instrum., vol. 69, no. 2, pp. 810–812,Feb. 1998.

[47] O. Zimmer, “Vacuum arc deposition by using a Venetian blind particlefilter,” Surf. Coat. Technol., vol. 200, no. 1–4, pp. 440–443, Oct. 2005.

[48] S. Y. Chun, A. Chayaraha, A. Kinomura, N. Tsubouchi, C. Heck,Y. Horino, and H. Fukui, “Macroparticle-free Ti–Al films by newly devel-oped coaxial vacuum arc deposition,” Jpn. J. Appl. Phys., vol. 38, no. 4B,pp. L467–L469, Apr. 1999.

[49] C. F. Meyer and H.-J. Scheibe, “Filtered laser-arc: A new technol-ogy for deposition of particle-free films,” presented at the Int. Conf.Metallurgical Coatings Thin Films (ICMCTF), San Diego, CA, 1999,Paper B4-9.

[50] T. Yoshitake, G. Shiraishi, and K. Nagayama, “Elimination of dropletsusing a vane velocity filter for pulsed laser ablation of FeSi2,” Appl. Surf.Sci., vol. 197/198, pp. 379–383, 2002.

[51] B. Petereit, P. Siemroth, H.-H. Schneider, and H. Hilgers, “High cur-rent filtered arc deposition for ultra thin carbon overcoats on magnetichard disks and read–write heads,” Surf. Coat. Technol., vol. 174/175,pp. 648–650, 2003.

[52] P. Siemroth, J. Berthold, B. Petereit, H.-H. Schneider, and H. Hilgers, “Anew generation of filtered arc sources for ultrathin top coats on magnetichard disks,” Surf. Coat. Technol., vol. 188/189, pp. 684–690, 2004.

[53] T. Schülke, A. Anders, and P. Siemroth, “Macroparticle filtering of high-current vacuum arc plasmas,” IEEE Trans. Plasma Sci., vol. 25, no. 4,pp. 660–664, Aug. 1997.

[54] T. Witke, T. Schülke, B. Schultrich, P. Siemroth, and J. Vetter, “Com-parison of filtered high-current pulsed arc deposition (φ-HCA) withconventional vacuum arc methods,” Surf. Coat. Technol., vol. 126, no. 1,pp. 81–88, Apr. 2000.

[55] H. Fukui, J. Okida, N. Omori, H. Moriguchi, and K. Tsuda, “Cuttingperformance of DLC coated tools in dry machining aluminum alloys,”Surf. Coat. Technol., vol. 187, no. 1, pp. 70–76, Oct. 2004.

[56] N. Miyakawa, S. Minamisawa, H. Takikawa, and T. Sakakibara,“Physical–chemical hybrid deposition of DLC film on rubber by T-shapefiltered-arc-deposition,” Vacuum, vol. 73, no. 3/4, pp. 611–617, 2004.

[57] H. Takikawa, N. Miyakawa, S. Minamisawa, and T. Sakakibara, “Fab-rication of diamond-like carbon film on rubber by T-shape filtered-arc-deposition under the influence of various ambient gases,” Thin SolidFilms, vol. 457, no. 1, pp. 143–150, Jun. 2004.

[58] H. Hikosaka, Y. Iwasaki, H. Takikawa, T. Sakakibara, H. Hasegawa, andN. Tsuji, “Development of X-shape filtered arc deposition apparatus forthick ta-C film coating,” Trans. Inst. Electr. Eng. Jpn., vol. A-126, no. 7,pp. 757–762, 2006. vol. FM-126, [in Japanese].

[59] H. Tanoue, H. Hikosaka, Y. Iwasaki, H. Takikawa, T. Sakakibara, andH. Hasegawa, “Development of X-shape filtered arc deposition (X-FAD)apparatus and DLC/Cr film preparation,” IEEE Trans. Plasma Sci., 2007.to be published.

[60] H. Takikawa, K. Izumi, R. Miyano, and T. Sakakibara, “DLC thin filmspreparation by cathodic arc deposition with a super droplet-free system,”Surf. Coat. Technol., vol. 163/164, pp. 368–373, 2003.

[61] Y. Iwasaki, S. Minamisawa, H. Takikawa, T. Sakakibara, andH. Hasegawa, “Influence of duct bias on deposition rate of DLC film inT-shape filtered arc deposition,” Vacuum, vol. 80, no. 11, pp. 1266–1271,2006.

Hirofumi Takikawa (M’06) was born in Japan in1962. He received the M.S. and Ph.D. degrees in en-gineering from Toyohashi University of Technology,Toyohashi, Japan, in 1986 and 1992, respectively.

He is currently a Professor in the Departmentof Electrical and Electronic Engineering, ToyohashiUniversity of Technology.

Prof. Takikawa is a member of the Institute ofElectrical Engineers of Japan and Japan Society ofApplied Physics.

Hideto Tanoue was born in Japan in 1983. Hereceived the B.S. degree in electrical and electronicengineering from Toyohashi University of Technol-ogy, Toyohashi, Japan, in 2005. He is currentlyworking toward the M.S. degree in the Departmentof Electrical and Electronic Engineering, ToyohashiUniversity of Technology.

Mr. Tanoue is a Student Member of the JapanSociety of Applied Physics.

992 ieee transactions on plasma science, …staff.ustc.edu.cn/~fuzp/course/paper/review of...

Documents