modification of thin film properties by sputtered particles · sputtering process is discussed in...

The novelty of the functions provided byplasma-assisted physical vapor depositionfilms, together with their durability for prac-tical use, is emphasized as the area wherethe thin film process has a significantimpact. Remarkable advances have beenmade in recent years in science and technol-ogy of the thin film process for deposition.Up to the present, thin films have been stud-ied from the viewpoint of the relationshipbetween their structure and properties [1-3].On the other hand, the commercial use ofthin films has been growing at a surprisinglyrapid rate for the last two decades in almostall the industrial fields such as electronicsand optics. These films are practicallyformed by depositing materials onto a sup-porting substrate to build up a thin filmthickness through a complicated thin filmprocess rather than by thinning down bulkmaterials by simple methods such asmechanical polishing and ion milling.

Nowadays, thin film technology has beenclassified by the method of film formationsuch as physical vapor deposition (PVD) orchemical vapor deposition (CVD). In addi-tion, the resulting films are also character-ized by their chemical bondings such asorganic or inorganic films. Therefore, it is

essential to understand the elementarystages of formation in thin film process inwhich low starting materials are modifiedfrom bulk to thin films.

In this review paper, I first describe a newconcept of the thin film process in materialsynthesis on the basis of ion-surface interac-tion during deposition. Some examples ofpractical applications of SiO2 and Ni-Si-Bfilms are given.

Irrespective of thin film preparation meth-ods such as PVD and CVD, the thin filmprocess comprises three elementary stagesincluding decomposition, transport, andnucleation and growth mechanisms [4].Fig. 1 shows the flow chart of the thin filmprocess, where starting materials are succes-sively modified to the resulting films. In thefirst stage, starting materials in the form ofgas, liquid or solid are decomposed into var-ious fragments of neutrals or ions in theform of atoms, molecules, clusters or pow-ders by the external powers of plasma, laser,ion, microwave and thermal energies. Thedecomposed fragments thus formed willtravel through the medium of gas or liquidand approach to the substrate. This phase isreferred to as the transport stage in the thinfilm process. The Monte Carlo simulation

1

豊田中央研究所 R&D レビュー　Vol. 28 No. 3 ( 1993. 9 )

Modification of Thin Film Properties by SputteredParticles

Yasunori Taga

薄薄膜膜ププロロセセスス技技術術

多賀康訓

Review

Keywords Thin film, Sputter deposition, Kinetic energy, Film properties

1. Introduction

2. Thin Film Process

technique is helpful in understanding themechanism of the transport stage. Thechemical reaction between the transportmedium and the decomposed fragments isalso important in the reactive process of thinfilm formation. The decomposed fragmentsland on the substrate for nucleation andgrowth, which result in the formation offunctional films. Thin film properties arestrongly influenced at this final stage,because the energy of the decomposed frag-ments is dissipated in the very shallow sur-face region of the substrate. This dissipationof the energy of the decomposed fragmentsmay enhance the surface migration of adatoms, chemical reaction between landingfragments and adsorbed molecules, andfinally reconstruction into the structures ofthe resulting films.

Consequently, we have possibilities ofobtaining superior properties of materialsin the form of thin films compared to bulkthrough the thin film process shown inFig. 1.

On the other hand, the influence of thekinetic energy of particles in the thin filmprocess on thin film formation ranges fromsimple substrate cleaning for enhancedadhesion to morphological changes and epi-taxial growth. Peculiar features of sputterdeposition technique are generally explainedin terms of the energy of the sputtered parti-cles which are decomposed and emitted as aresult of high energy ion bombardment inplasma. A complex variety of processessimultaneously occurs whenever energeticparticles interact with a substrate or growingfilms. For a comprehensive description ofthese processes, I refer to a number of excel-lent reviews in the literature [2, 3]. Thesputtering process is discussed in view ofthe energy distribution of sputtered particlesand some chemical effects on the substratesurface by high energy ion bombardment.

The importance of ion-surface interactionprocesses in thin film deposition depends on

2


Fig. 1 Flow chart of the thin film process.

3. Ion-Surface Interaction Processes

the particular deposition process employed.For a comprehensive description of theseprocesses, a number of excellent reviewscan be seen in the literature [5-7]. In thepresent discussion, the ion-surface interac-tion processes illustrated in Fig. 2 are con-sidered, where sputtering, ion reflection,penetration and trapping are brieflyreviewed.

One of the principal effects during ionbombardment of a solid surface is the ejec-tion or sputtering of the surface atoms. It iswell understood that the energetic primaryions lose energy in a series of elastic andinelastic collisions with target atoms. Therecoiling target atoms give rise to secondarycollisions in a collision cascade. The energyof primary ions may eventually be trans-ferred to the surface atoms which will beejected if the surface binding energy is over-come. Sputtering results from collisionsequences less than five atomic layersbeneath the surface. These atoms, ejected asa result of sputtering, deposit onto a sub-strate which may be close to the target.

Sigmund [1] reported the most successfulmodel of sputtering theory that I have

Fig. 2 Schematic diagram of the ion-surface

interaction process.

known to date, and Andersen and Bay [8]compiled the most complete set of experi-mental sputtering yields by energetic ionsranging from 0.1 to 100 KeV. However, thesituation is less satisfactory for sputteringcompound targets of multielements sincethere exists no universal sputtering theorybut only limited experimental data. Thisphenomenon is usually referred to preferen-tial sputtering; the degree of preferentialsputtering depends on projectile energy,mass and target atom binding energy.Compositional changes due to ion bombard-ment can be significant where growing filmsare irradiated. Many measurements of pref-erential sputtering have been reported, andCoburn [9] has compiled data on metalalloys and oxides.

Energy distributions of sputtered particlesattain their peak at 1 to 2 eV and extend toabout 100 eV. Film properties are largelyinfluenced by the energies of the depositingatoms. In addition, a small percentage ofthe sputtered particles is ionized, and has asimilar energy distribution to the neutralcomponent except for the peak at a higherenergy of about 5 to 10 eV [10].

Ion-surface interactions result in the dis-placement of surface and bulk atoms causedby the collision cascade process which iscompleted within 10

–13sec. Displacement or

thermal spikes may occur, and surfaces mayundergo crystalline-amorphous transitions.Recoil implantation and cascade mixing cre-ate the surface defects, which can enhancediffusion process and accelerate the forma-tion of compounds such as oxide or nitrides[11].

Thin films deposited by magnetron sput-tering or ion beam sputtering of a target maybe subjected to significant bombardment byprimary particles scattered off the target sur-

3


face onto the growing film. The growingfilm then contains a percentage of primaryparticles as well as being subjected to radia-tion-induced structural changes. On theother hand, the distribution of atoms ejectedas a result of ion bombardment from targetsdoes not simply obey a cosine law [12], buttends to concentrate along the direction ofso-called focused collisions. Hoffman andThornton investigated the relationshipbetween the bombardment of the growingfilm by high energy particles elasticallyreflected from the target surface andmechanical properties of thin films [13, 14],and concluded that the compressive stresshad been caused by an atomic peening effectby the argon atoms reflected from the targetsurface.

It may be said from the above discussionon the ion-surface interaction that the phe-nomena occurring during thin film growthby sputtering are very complicated [15, 16].In order to understand the phenomena, it isnecessary to study the parameters for thinfilm deposition more systematically fromthe following viewpoint: The influence ofkinetic energy of sputtered particles on thinfilm properties. For this purpose, precisemeasurements of energy distribution of sput-tered ions are first established and thenapplied to practical SiO2 and Ni-Si-B films.

In this review, the thin film process ofsputter deposition is described on the basisof the effects of kinetic energy of sputteredparticles and ion bombardment during depo-sition on thin film properties.

In sputter deposition at high gas pressure,film properties are largely influenced by theenergies of the depositing atoms. In con-ventional evaporation, the maximum energy

is only ~0.1 eV, whereas sputtered particleenergy distributions peak at 1 to 2 eV andextend to about 100 eV [17]. A small per-centage of the sputtered particles is ionizedand has a similar energy distribution to theneutral component but peaking at a slightlyhigher energy [10]. However, it is well rec-ognized that the kinetic energy of sputteredions plays an important role in adding chem-ical activity to the growing film, whichresults in changes of the crystallinity, orien-tation, morphology and adhesion to the sub-strate [18].

Measurements of the energy distribution(ED) of sputtered ions were performed withan ultra high vacuum - secondary ion massspectrometry (UHV-SIMS) apparatus.Details were described previously [19]. Theenergy and mass of the secondary ions wereanalyzed with an electrostatic ion energyanalyzer (IEA) and a quadrupole mass ana-lyzer, respectively. The absolute energy ofthe IEA was about 2.0 eV. The absoluteenergy of the IEA was calibrated by Al+

thermal ions. The sample in the SIMS studywas a mirror-polished B-doped Si(111) witha resistivity of 1-2 Ωcm, which was alsoused as a substrate for thin film deposition.The sample was cleaned by Ar ion sputter-ing after being introduced into the chamber.

The obtained EDs were characterized bythe following parameters:

(1) Most probable energy Em; (2) fullwidth at half-maximum (FWHM); (3) tailfactor N; (4) average kinetic energy Ea. Inthe usual cases, the N value is defined by theslope of the distribution in the high energyregion. In the present study, however, theintensity of the secondary ions in the higherenergy region was fairly low. In order toavoid an inaccurate definition of N, the fol-lowing procedure correction was performed.

4


4. Characterization of Sputtered Particles

We made an assumption that each distribu-tion obeyed Thompson's theory [20] with anN value as a fitting parameter. That is,

where E is the energy of sputtered ions, Eb isthe surface binding energy and C is a con-stant. N = 2 gives Thompson's theory. TheN value was defined by fitting this functionto the observed distribution. The N valuerepresents an approximate measure of thedistribution function and the Em valuedirectly associated with the surface bindingenergy as shown by Thompson's theory.

Although many theoretical models havebeen proposed to describe the ionizationphenomena of sputtered particles, a unifiedexplanation cannot be given at present. Thepresent author [21-24] has established themethods of simultaneous determination ofsputtering yield and the ionization probabili-ty for thin film targets with known thicknessand composition. As a result, the sputteringyield of metals under O2

+ bombardment wasinversely proportional to the so-called ener-gy transfer factor in the classical head-oncollision model and a linear relationship wasfound between the ionization potential and amodified degree of ionization which can beexpressed by the ionization potential and thesputtered O2

+ current. The essential features of the transport of

sputtered particles from a target to a sub-strate during sputter deposition were studiedby calculation using the Monte Carlo tech-nique [25, 26]. The study takes into consid-eration the change in momentum as well asthe kinetic energy loss of sputtered particlesin their collisions with ambient gasmolecules, to gain an understanding of theeffects of these factors and of the number ofsputtered particles arriving at a substrate on

the mechanism of growth of a thin film bysputter deposition. Some theoretical predic-tions using the above calculation were madefor several selected conditions of sputterdeposition. Fig. 3 shows a typical example of the

Monte Carlo calculation of the trajectoriesof 50 particles during high and low pressuresputter deposition, for easy understanding ofthe entire transfer process [26].

Fig. 3 Trajectories of 50 silver particles ejected into an argongas ambient from the origin with Thompson's energyand angular distribution at a sputtering voltage of 2 kVand T = 305 K: (a) PAr = 6.6×10-1 Pa., D = 4.5 cm;(b) PAr = 2 Pa., D = 4.5 cm; (c) PAr = 6.6×10-1 Pa., D= 3.0 cm; (d) PAr = 2 Pa., D = 3.0 cm.

5


f(E) = CE / (E + Eb)N+1・・・・・・(1)

5.1 SiO2 Films SiO2 films have been recognized to be

one of the most important materials in elec-tronic devices and sensors. These films areexpected to have high insulative propertieseven after preparation at low temperatureprocess. In this section, the author tried [27,28] to elucidate the effects of growth condi-tions of sputter deposition on the electricalproperties of SiO2 films with a specialemphasis on the energy distribution of sput-tered particles. Fig. 4 shows the energy distributions of

Si+, SiO+, Si2O+ and Al+ thermal ions under

oxygen exposure at 5.3 × 10–5

Pa. Table 1denotes the most probable energy (Em) andthe average energy (Ea) of secondary ionsunder different conditions of primary ions(Ar+, O+) and targets (Si, SiO

2). The values

of Em in the case of O+→ Si were similar tothat of Ar+ → SiO2, while those of Ar+ →Si+O2 (gas) do not resemble the above cases,that is, the Em of SimOn

+ (m, n = 1, 2, ...) isalmost equal to that of Al+ thermal ions. Itis thought that the formation mechanisms ofSimOn

+ in the case of Ar+ → Si+O2 (gas) isdifferent from those of Ar+ → SiO2 and O+

→ Si. It is thought that these molecular ionsare created at the surface really as a result ofthe interaction between the secondary parti-

cles and adsorbed atoms. The mechanism ofthe molecular ion formation mentionedabove can be interesting compared to that ofchemical sputtering [29, 30].

Radio frequency (RF) magnetron sputter-ing technique was used for deposition ofSiOx thin films. Hot-pressed SiO2 and poly-crystalline Si targets of 99.9 % purity werebonded to a water-cooled copper backingplate. After evacuation to 1.3 × 10

–4Pa.,

sputtering gas was introduced to 6.6 × 10–1

Pa. SiO2 films 50 nm thick were thusdeposited at the target-to-substrate distanceof 150 mm.

Arrangements in a deposition chamber for

Fig. 4 Changes in ion energy distributions of Si+, SiO+ andSi2

+ from Si under Ar ion bombardment during O2exposure to 5.3×10–5 Pa.

6


1.1 (11.0) 0.6 (5.2) — 0.7 (7.0) — —4.9 (16.1) — 4.2 (7.3) 4.6 (10.9) 3.0 (5.0) 3.3 (4.5)4.2 (17.7) 2.0 (10.7) 2.8 (8.8) 3.2 (12.9) 1.9 (5.2) 1.6 (3.9)1.1 (14.2) 0.6 (7.4) * (2.4) 0.4 (6.3) * (1.8) * (1.0)

Ar+→ SiO+ → SiAr+→ SiO2

Ar+→ Si+O2

Si+ Si2+ SiO+ Si2+ Si2O+ Si2O2+

Table 1 The most probable energy Em and the average energy Ea of sputtered

ions under various conditions. Ea values are shown in parenthesis.

* indicates the Em value equal to Al+ thermal ions.

5. Applications

thin film preparation were schematicallyshown in Table 2. Modes 1, 2 and 3 werenormal methods, but in Mode 4, the experi-mental arrangement used was speciallydesigned to isolate the target from oxygengas, maintaining it in an inert gas atmo-sphere while introducing oxygen to the sub-strate. As a result, experimental apparatusallowing reactive sputter deposition, inwhich the chemical reaction occurs at thedeposition surface, has been developed.

Results of characterization and electricalproperties of SiOx films are summarized inTable 3.

It is interesting to note the results of themeasurement of breakdown voltage of thefilm thus prepared in Modes 1-4. Thebreakdown voltages Eb of the film by Modes2 and 4 are high, and those by Modes 1 and3 are rather low. In order to elucidate thedifferences in Eb between the films byModes 2 and 4, the results of J-E measure-ments [31] are depicted in Fig. 5, where the

Growth rate (Å/min)

Refractive indexStoichiometry (AES)Stoichiometry (IR)

Mode 196

2.001.43

4.12

Mode 222

2.141.42

9.9

Mode 32041.711.47

1.32

Mode 463

1.901.43

9.26

Density (g/cm3)

Eb (MV/cm)

SiO2 SiO2 SiO1.8 SiO2

SiO2 SiO2 SiO2 SiO2

Table 3 Characterization and properties of SiOxfilm by Modes 1, 2, 3 and 4.

Table 2 Schematic representation of thin film preparation modes under various conditions.

J-E curve of the thermally grown SiO2 filmis shown for comparison. In the figure, theEb values of each sample are shown byarrows. Similar Eb values are obtained forModes 2 and 4, but the changes in J with Eare quite different, that is, J in the film byMode 2 begins to increase at E = 2.5MeV/cm, while J in the film by Mode 4remains unchanged to E = 5.5 MeV/cm, andthen begins to increase gradually. The J-Echaracteristic of the film by Mode 4 is simi-lar to that of thermally grown SiO2 films. Itcan be explained that the current-transportmechanism in the present cases such as highfield and low temperature may be a result ofthe field ionization of the electrons trappedinto a conduction band, which is usuallyreferred to as the Poole-Frenkel effect [31].The trapped electrons in the SiO2 films pre-pared by Mode 2 may be induced by highenergy bombardment of sputtered particles.On the contrary, the Ea values of the sput-tered particles in Mode 4 are low in compar-ison with Mode 2 and the correspondingtrapped electrons in the SiO2 films thus pre-pared will be diminished, resulting in the

Fig. 5 J-E characteristics of Al/SiO2/Si system

for Modes 2 and 4.

7


same J-E characteristic of the thermallygrown SiO2 films (in this case, no defect orvacancy was introduced). The differences inJ-E plots between the films by Modes 2 and4 can be reasonably explained in terms ofthe differences in Em or Ea of sputtered par-ticles.

Comparison between the data shown inTables 1 and 3 reveals that the differences inthe current-voltage characteristics of SiO2

films prepared under various sputtering con-ditions can be reasonably explained in termsof the changes in the most probable energyof the sputtered particles.

5.2 Ni-Si-B Films The Amorphous Ni-Si-B (NSB) ternary

alloy film has been well known as a materialwith a high tensile strength (2.6 GPa.) and alow temperature coefficient of resistance(TCR, 1x10

–6/˚C) originated from the short

mean-free-path (MFP) of electrons in thefilms. It is also well recognized that the val-ues of MFP are quite sensitive to themicrostructure of the films. The kineticenergy of sputtered particles to be depositedmay give an important effect on themicrostructure of NSB films.

On the other hand, a lot of research hasbeen devoted to determining the film forma-tion mechanism in sputter deposition as wellas in other ion-assisted techniques [32-35].It has become clear that the impact by ener-getic particles during deposition affects filmproperties significantly. The effect of theenergetic particles reflected from the targetsurface in sputter deposition has also beenstudied by several researchers [13, 14, 36].Until now, there have been few reportswhich describe the relation between the EDsof the sputtered particles and the film prop-erties.

From this point of view, the effect of theEDs of spu t te red par t i c les on f i lmformation, especially on alloy film forma-tion, was determined. Two representativesputtering methods used to prepare ternaryalloy films were examined; co-sputteringwith multiple elemental targets (co-sputter-ing) and sputtering with an alloy target(alloy sputtering). First, we measured theEDs of the ions sputtered from elementalnickel, silicon and boron targets and alsothose from an amorphous Ni-Si-B (NSB)target by ultra high vacuum secondary ionmass spectrometry (SIMS); the former cor-responds to co-sputtering and the latter toalloy sputtering. The results were comparedwith the microstructures of the films pre-pared by the corresponding methods.

As a result of SIMS analyses, Ni+ andNi2

+ ions were observed from the Ni target,but Ni2

+ was not detected. On the contrary,various kinds of ions such as Sin+, Si+, andSin

+ (n = 2, 3) were observed from the sili-con target. The boron target emitted onlyB+, except impurity ions. These EDs werecharacterized by the parameters describedabove, and the results are presented inTable 4. Most of the curves, except that forSi+, had lower N values than that given byThompson's theory.

It is noteworthy that the Em values of thesingly-charged monoatomic ions emittedfrom the alloy were lower than those from

1.9 (1.5) 2.7 (1.9) 2.5 (1.5)7.1 (6.9) 10.7 (9.8) 13.2 (8.6)1.6 (1.7) 2.0 (1.6) 1.4 (1.3)

17.1 (11.3) 15.5 (15.4) 18.4 (14.3)

Ni+ Si+ B+

Em (eV)FWHM (eV)NEa (eV)

Table 4 Results of the characterization of the EDsof sputtered ions.

8


the elemental targets. The lowering of theEm values resulted in a decrease in the Ea

values of the monoatomic ions. The presentresults showed that the Eb values of theatoms in the amorphous alloy target werereduced in comparison with those in the ele-mental targets.

NSB films were prepared by using thesputtering methods corresponding to theenergy distribution measurement. AnNi67Si21B13 target was used in alloy sputter-ing. An NaCl single crystal was used as asubstrate, on which almost 50 nm thickfilms were deposited. Also in this case, thecompositions of the films were determinedby AES and were found to be almost thesame, i.e. Ni69Si18B13 by co-sputtering andNi69Si21B10 by alloy sputtering. Structures of the NSB films were examinedby means of transmission electronmicroscopy (TEM). The transmission elec-tron diffraction analysis revealed that bothof the films were amorphous. Fig. 6 showsTEM images of the films. Obviously, thetwo films displayed different images. Thefilm prepared by co-sputtering was relative-ly uniform in low contrast. There aredomains about 20 nm in diameter whoseenvelop is seen clearly. On the contrary, thefilm prepared by alloy sputtering exhibited

Fig. 6 Transmission electron micrographs of the NBS films

prepared by the different deposition techniques:

(a) alloy sputtering; (b) co-sputtering.

rather a strong contrast and the diameter ofthe domains ranged from 5 to 20 nm.

The contrast could be attributed to massthickness fluctuation in the film. There arevarious candidates for the mass thicknessfluctuation: Fluctuation in geometricalthickness and that in composition. As aresult, it is recognized that the film deposit-ed by co-sputtering has a more uniformmass thickness than the film deposited byalloy sputtering. Two representative sputterdeposition methods used to prepare ternaryalloy films were examined.

Although it is still difficult to compare thedifference in film morphology directly withthe difference in the EDs of sputtered ions,the present results suggested that, in co-sputtering, the film grows under particlebombardment with higher energy than inalloy-sputtering. It is also suggested that thedifference in the energy of the depositingparticles affects their surface mobilities; infilm deposition by co-sputtering, particles tobe deposited would have a higher surfacemobility owing to their higher kinetic ener-gy, which might result in uniform film for-mation in co-sputtering. The difference inmass spectra, i.e. existence or absence ofmolecular ions such as NiSi+ and NiB+,might also affect the film structure. Ofcourse, uncertainty still exists concerningthe modification of EDs through the trans-port process. However, the difference in theEDs at the initial stage would affect the finalEDs. Further investigation will be neces-sary to clarify the deposition mechanism.

The kinetic energy of depositing the parti-cles, formed by sputtering process in agrowing film, strongly influences the struc-ture and property of the resulting films. The

9


6. Conclusions

electrical properties are particularly sensi-tive to the film nanostructure affected by thekinetic energy of the sputtered particles.Nowadays, it has become possible to controlthe film microstructure by the aid of thekinetic energies of both sputtered particlesto be deposited and reflected particles whichbombard the growing films. Practical appli-cations of high quality electrical films pre-sented in this review are really proofs of thesophisticated thin film technology of sput-tering. Further progress of thin film tech-nology must be considered in detail toexplore advanced thin film materials scienceand to ensure the field reliability of futureproducts.

[1] Sigmund, P. : Phys. Rev., 184(1969), 383[2] Pivin, J. C. : J. Mater. Sci., 18(1983), 1267[3] Martin, P. J. : J. Mater. Sci., 21(1986), 1[4] Taga, Y. : Proc. SPIE, 1275(1990), 110[5] Kaminsky, M. : Atomic and Ionic Impact on Metal

Surfaces, (1965), Springer-Verlag, Berlin[6] Carter, G. and Colligon, J. S. : Ion Bombardment

of Solids, (1988), Elsevier, New York[7] Berisch, R. : Sputtering by Particle Bombardment,

Topics in Applied Physics, Vol. 52, Ed. by R.Berisch, (1983), Springer-Verlag, Berlin

[8] Andersen, H. H. and Bay, H. L. : ibid. p.145[9] Coburn, J. W. : Thin Solid Films, 64(1979), 371[10] Stuart, R. V. and Wehner, G. K. : J. Appl. Phys., 35

(1964), 1819 [11] Kleinsasser, A. W. and Buhrman, R.A. : Appl.

Phys. Lett., 37(1980), 841[12] Olson, R. R. and Wehner, G. K. : J. Vac. Sci.

Technol., 14(1977), 319[13] Hoffman, D. W. and Thornton, J. A. : Thin Solid

Films, 40(1977), 355[14] Id. : J. Vac. Sci. Technol., 16(1979), 134[15] Kageyama, Y. and Taga, Y. : Appl. Phys. Lett., 55

(1989), 1035[16] Gotoh, Y. and Taga, Y. : J. Appl. Phys., 67(1990),

1030[17] Rudat, M. A. and Morrison, G. M. : Surf. Sci., 82

(1979), 549

[18] Takagi, T. : Physics of Thin Films, Vol. 13, Ed. byFrancome, M. H. and Vossen, J. L., (1987), 1,Academic Press, Boston

[19] Ohwaki, T. and Taga, Y. : Jap. J. Appl. Phys., 23(1984), 1466

[20] Thompson, M. W. : Philos. Mag., 18(1968), 377[21] Taga, Y. : Secondary Ion Mass Spectrometry SIMS

V, Ed. by Benninghoven, A., Colton, R. J., Simons,D. S. and Werner, H. W., (1985), 32, Springer,New York

[22] Taga, Y., Inoue, K. and Satta, K. : Surf. Sci., 119(1982), L363

[23] Inoue, K. and Taga, Y. : Surf. Sci., 140(1984), 491[24] Inoue, K. and Taga, Y. : Surf. Sci., 147(1984), 427[25] Motohiro, T. and Taga, Y. : Surf. Sci., 134(1983),

L494[26] Motohiro, T. and Taga, Y. : Thin Solid Films, 112

(1984), 161[27] Ohwaki, T. and Taga, Y. : Appl. Phys. Lett., 55

(1989), 837[28] Taga, Y. : MRS Symp. Proc., 128(1989), 41[29] Roth, J. : J. Appl. Phy., 52(1983), 91[30] Saidoh, M., Gnaser, H. and Hofer, W. O. : Appl.

Phys., A40(1986), 197[31] Sze, S. M. : J. Appl. Phys., 38(1967), 2951[32] Vossen, J. L. and Kern, W. : Thin Film Processes

(1978), Academic Press, New York[33] Chapman, B. N. : Glow Discharge Process, (1980),

Wiley, New York[34] Sawada, Y. and Taga, Y. : Thin Solid Films, 110

(1983), L129[35] Westwood, W. D. : Thin Film Physics, Vol.14, Ed.

by Francome, M. H. and Vossen, J. L., (1989),Academic Press, New York

[36] Rossnagel, S. M. : J. Vac. Sci. Technol., A7(1989),1025

10


References

Yasunori Taga was born in 1944. He received

the B. Eng. degree in fine measurement in 1968

and M. Eng. degree in applied physics in 1970,

both from Nagoya Institute of Technology. He is

Assistant General Manager of Electronics

Device Division, where he works on thin films

and surface and interface physics. He is a

member of Society of Applied Physics,

American Vacuum Society, Optical Society of

America and Materials Research Society.

modification of thin film properties by sputtered particles · sputtering process is discussed in...

Documents