Hoffman D.M., Singh B., Thomas J.H. (Eds). Handbook of Vacuum Science and Technology
Подождите немного. Документ загружается.
684
Rg.9.
Chapter
53:
Ion
Beam Technology
PRECLEAN AND
ASSIST
ION SOURCE
SUBSTRATE
SHUTTER
SUBSTRATE
HOLDER
SPUTTER
DEPOSITION
ION
SOURCE
VACUUM
PUMP
Typical configuration of ion beam sputter deposition (IBSD) system with ion-beam-assisted
deposition (IBAD) capability.
tion sytem layout is shown in Figure 9. Several properties of ion beam deposition,
such as the energetics involved, distinguish it from other deposition methods.
The energy of the sputtered atoms as they leave the target ranges from a few
up to tens of eV [45]. Due to the low-pressure regime of ion beam processing
(10
""^
torr), where the mean free path is on order of
a
meter these sputtered atoms
arrive at the substrate with most if not all of their original kinetic energy. This is
in contrast to the higher-pressure regime of magnetron sputtering, where the sput-
tered atoms thermalize with the background gas before reaching the substrate.
This lack of scattering has two direct consequences. First, the higher kinetic en-
ergy with which the atoms arrive imparts
a
high degree of mobility to the growing
film,
yielding good coverage at very small thicknesses, smoother films [46], and
enchanced microstructure. Second, the sputtered atoms maintain their direction
vector from the target to the substrate, allowing control over the atom's incidence
angle at the substrate, which can give excellent coverage of high-aspect-ratio ge-
ometries. Figure 10(a) shows a typical step coverage result
from
magnetron sput-
tering and Figure 10(b) shows step coverage of the same film material deposited
by ion beam sputtering.
5.3.3
Ion
Beam Sputter Deposition
685
Rg.lO.
(a)
5 fim
(b)
1 [im
(a) Typical step coverage for magnetron sputter deposition and (b) step coverage using ion
beam sputter deposition. Substrate material is Si02 and deposited film is "Sendust" (85% Fe/
10%
Si/5% Al) in both (a) and (b).
686 Chapter 5.3: Ion Beam Technology
Energy is also added to the growing film by reflected neutrals and has been
studied extensively
[45,47,48].
The higher energetics of ion beam sputtering makes
substrate heating unnecessary where it is often required in other thin-film deposi-
tion systems to obtain the desired film microstructure.
The amount of energy carried to the substrate by the sputtered atoms and re-
flected neutrals can be controlled over more than an order of magnitude by ad-
justing the beam energy and the ion species. Herein lies the true elegance of ion
beam processing, where the process parameters can be independently controlled
over wide ranges to optimize the desired film properties. In glow discharge-based
deposition systems such as magnetrons, it is difficult if not impossible to decouple
these variables, which are interlinked in a complex manner [49,50].
The major sources of contamination in ion beam deposition are from lack of
beam containment on the target and sputtered material from the ion source. Proper
design of the system geometry and proper ion source operation [51] should result
in contamination levels less than
1:10"*
for transitional metals. Target contamina-
tion is the highest from the immersion-type neutralizer and is minimized with the
plasma bridge and hollow cathode emitters.
Until recently, ion beam sputtering has often been characterized as useful only
for processes requiring low deposition rates and small substrate sizes. While low
deposition rates (~
1
A/s) are often desirable to control deposition of ultra-thin
films, Fe deposition by ion beam sputtering has recently been demonstrated [44]
at >20 A/s over 200 mm diameter wafers with excellent uniformity. In general,
the deposition rate from material to material follows the same trend as etch rate,
because they both are derived from the sputter yield.
AI2O3 films deposited with ion beam deposition have been shown to have su-
perior properties to e-beam and sputtered films [52]. These properties include
film stress an order of magnitude less than RF-sputtered films, nearly double the
AI2O3 electrical breakdown strength, and higher refractive index.
A current area of intense research is the deposition of new types of magnetic
thin films for higher-density recording. Soft magnetic materials with high magne-
tization saturation and permeability such as Fe-N and Fe-C are of interest for use
in high-flux-density write heads and have been studied using ion beam deposition
[53,54] because of the advantages previously discussed, especially for conformal
coverage of irregular topologies.
Thin-film magnetoresistance (MR) read heads are the current state of the art in
digital magnetic storage and has been studied extensively by ion beam sputtering
[55-57].
The ]VIR effect is the change in electrical resistance of a material in the
presence of a magnetic field and is used in the read head in magnetic storage de-
vices.
Giant magnetoresistance (GMR), dicovered in 1988
[58],
has been observed
in ultra-thin (—10 A) multiple (5-20) layer stacks of magnetic and nonmagnetic
materials [59-66] known as spin valves. These types of structures will likely be
used in the next generation of magnetic storage devices. The ability of ion beam
5.3.4 lon-Beam-Assisted Deposition
687
sputtering to put down these multilayer stacks in a precise, controlled fashion by
use of multiple targets is making it an attractive tool for research and possible
production.
5.3.4
lON-BEAM-ASSISTED DEPOSITION
Many advantages
due to the
energetic condensation
of
the growing film found
in
ion beam deposition
can
also
be
obtained
by
bombarding
the
growing film with
an energetic
ion
beam. This process
is
known
as
ion-beam-assisted deposition
(IB AD)
and can be
used
in
conjunction with
a
variety
of
deposition methods,
in-
cluding
ion
beam sputtering, electron beam evaporation, molecular beam epitaxy,
and pulsed laser deposition.
A
typical evaporation system layout with IBAD
ca-
pability
is
shown
in
Figure
11.
Fig.
11.
End hall
ion
source installation
in
commercial e-beam evaporation system
for
substrate pre-
cleaning and ion-beam-assisted deposition (IBAD).
688 Chapter 5.3: Ion
Beam
Technology
Several excellent reviews of IB AD have been recently written [67-71]. The
two parameters in ion-assisted film growth that most influence the film properties
are the ion energy and the ion-to-atom arrival rate ratio. As the energy and ion/
atom arrival rates are varied, different mechanisms come into play during film
growth. At low energies of 1-10 eV, surface desorption of adsorbed and chemi-
sorbed background gas is dominant. At this energy level, some displacement of
surface adatoms also occurs, which can affect surface morphology. The energy
range most commonly used in IBAD is 50-500 eV where significant enhance-
ment of film properties is achieved. The dominant mechanism in this regime is
often termed "densification" due to the number of altered film properties that are
related to film density. At still higher energies, effects from resputtering of de-
posited atoms and ion implantation will become more important. As the ion/atom
arrival rate is varied, different mechanisms can be brought into play. For example,
the amount of densification occurring at 20 eV for an ion/atom arrival ratio of 0.1
will be much less than that for a ratio of
1.
In general it appears that high ion/atom
ratios combined with low ion energy are most suited to minimize implantation
and below-surface defect production [71].
Evaporation alone tends to produce porous films with columnar grain structure
and is often supplemented with an ion assist source. Higher-temperature process-
ing of the substrate can improve these properties, but the disadvantages associ-
ated with heating the samples to several hundred degrees Celsius make this unat-
tractive in production systems. With the substrate at near room temperature, the
arriving atoms have a sticking coefficient near unity and very low surface mobil-
ity due to their low (~0.1 eV) thermal energies. As the ions collide with the sur-
face,
loosely packed clusters of atoms will be dissociated and given enough en-
ergy to diffuse along the surface and approach a more ideal packing density.
Intrinsic stress can develop in a growing film due to chemical mechanisms such
as the completion of oxidation reactions occurring below the surface and micro-
structure changes caused by defect generation or relaxation below the surface.
The stress induced chemically can be either compressive or tensile, depending on
the reaction, but the evolution of defects and microvoids generally results in ten-
sile stress. By bombarding the growing film with low-energy ions, compressive
stress can be introduced at the surface by forcing the atoms into closer proximity
to each other than in their relaxed state. This is similar to the ball peening that
is performed in metal working to induce compressive stress. This effect is some-
times referred to as "ion peening" and is used to counter the intrinsic tensile stress
that develops in the growing film. Proper adjustment of the ion beam parameters
can effectively neutralize the film stress.
This destabilization of surface atom clusters promotes step flow growth and
much smoother surface morphologies can be obtained [72]. This mechanism also
enhances the ability to obtain epitaxial growth at low substrate temperatures.
Crystallographic texturing of the growing film can be controlled in some cases
5.3.5 Ion Beam Direct Deposition 689
with IB AD by preferentially sputtering away one crystalline orientation. This has
recently become an important application in the deposition of superconducting
films. In this process, during the deposition of a yttria-stabilized-zirconia (YSZ)
buffer layer, IBAD is used to achieve in-plane texturing of the depositing YSZ.
High-quality superconducting films can be grown on single-crystal YSZ, which
acts as a seed layer. This texturing during IBAD allows the YSZ seed layer to
be deposited on a variety of materials prior to depositing the superconductor film
[73,74].
The control of film density has a large impact on properties such as refractive
index, hardness, and corrosion resistance. Optical coatings produced by IBAD
have higher refractive indexes and better temporal stability in air than those pro-
duced by other techniques. Electron beam evaporation coater manufacturers now
offer IBAD as either standard or optional equipment on their systems [75].
When depositing oxides such as AI2O3, Ti02, Ta205, or Si02, a low-energy
oxygen ion assist can be used to maintain proper stoichiometry of the deposited
film. Reactive IBAD can be used to deposit nitrides such as Si3N4, TiN, or BN
where the atomic element is deposited with simultaneous nitrogen ion beam bom-
bardment to complete the chemical reaction. Novel crystalline compounds such
as cubic-BN [76] and C-N [77], both of which have hardnesses on par with dia-
mond, have been studied by using IBAD.
Uniform coverage of irregular surface features such as was done in Figure 10
with ion beam deposition can also be performed with IBAD. Proper choice of ion
incidence angle on the sample can redistribute the deposited atoms over the fea-
tures and resputter the high spots of deposition.
In summary, IBAD allows the production of high-quality films by providing
the user with a very controllable method of changing the energetics at the surface
of the growing film. Both gridded and gridless ion sources are used for this appli-
cation, with gridless sources ideally suited due to their high current output at low
ion energies.
5.3.5
ION BEAM DIRECT DEPOSITION
The use of broad beam ion sources to deposit diamondlike carbon (DLC) has re-
cently found application in several areas. In this process, a carbon-carrying gas
such as methane is fed into the ion source and a low-energy (50-400 cV) beam is
directed at the substrate, forming a DLC film [78-82]. The term "direct deposi-
tion" is used since no sputtering target is present. Table 2 sunmiarizes some prop-
erties of DLC films grown by this method and compares them with diamond and
graphite. The films generally have high hardnesses, extremely low static and dy-
690 Chapter 5.3: Ion
Beam
Technology
Table
2
Properties of Diamondlike Carbon (DLC) Deposited by Direct Ion Beam Deposition.
Values for pure diamond and graphite are listed for comparison.
Property Diamond DLC Graphite
Hardness (GPa)
sp"*
Bonding (%)
Hydrogen Content (%)
Optical Gap (eV)
Coefficient of Friction
Refractive Index
100
100
0
5.45
0.05
2.4
15-30
40-50
30-50
1.2-4.0
0.05-0.1
2
—
0
0
0
0.1
—
namic friction coefficients, high transparency to the infrared and visible spec-
trum, and high indexes of refraction. Low deposition rates, similar to those found
in ion beam sputtering, characterize this method. Since the power density seen by
the substrate in this method is low, near-room-temperature film deposition is eas-
ily achieved with minimal cooling. This is often not the case in CVD plasma re-
actors where the substrate is immersed in the plasma. The DLC direct deposition
process can be done with both the end-hall-type sources and the gridded DC and
RF sources.
Currently this process is widely used to coat read/write heads for the magnetic
storage industry to provide a hard, low-friction surface that can also act as a chem-
ical diffusion barrier to prevent corrosion of the head. Since the flying heights
(head-media gap) of the head in the hard disk arena are currently on the order of
500 A, the DLC coating is on the order of 100 A or less and getting thinner as
flying heights decrease.
Another growing application for DLC coatings by direct ion beam deposition
is for coatings in the ophthalmic industry. Coatings on optical components from
sunglasses to supermarket scanner windows have been successfully coated with
DLC to provide a highly transparent scratch-resistant coating.
5.3.6
CONCLUSION
A broad range of ion source types and sizes are available commercially for a
number of ion beam processes. Due to the wide scope of ion beam technology, it
is not possible to cover many of the other types of ion beam sources and applica-
tions.
Broad ion beam technology applied to thin film etching allows a range of
material compositions and feature geometries to be etched in the same process
References ^91
due to its inherently physical nature. A reactive component can be added to this
process to enhance the rate and selectivity of the etch in
a
RIBE or CAIBE process.
Independent control of the ion beam parameters allows the isolation of key fig-
ures of merit in thin-film deposition processes. This is in contrast to the interlink-
ing of parameters in other methods such as RF diode sputtering, where it is difficult
to determine the importance of various mechanisms. This variable decoupling
allows precise control over the film properties. The increased energy of the film
growth in ion beam sputter deposition and ion-assisted deposition often can yield
superior results in a variety of properties over other deposition methods.
REFERENCES
1.
W. K. Chu and G. Langouche, MRS Bull. January (1993), p. 32.
2.
Donald L. Smith, Thin film deposition: Principles and practice (McGraw-Hill, New York, 1995),
pp.
453-555.
3.
Proceeding of 2nd Japan-U.S. Seminar on Focused Ion Beam Technology and Applications,
J.
Vac.
Sci. TechnoL, B9 (1991) 2461-2732.
4.
Handbook of Ion Implantation Technology, edited by J. F. Ziegler (North-Holland, Amsterdam,
1992).
5.
H. R. Kaufman, NASA Technical Note TND-585 (NASA, Washington, D.C., January 1961).
6. Charles E. Garner, NASA Tech Briefs (May 1995), p. 48.
7.
H. R. Kaufman, W. E. Hughes, R. S. Robinson, and G. R. Thompson, Nuclear
Instr.
and Meth.
in
Phys.
Research, B37/38 (1989) 98.
8. H. R. Kaufman and R. S. Robinson, American Institute of Aeronautics and
Astronautics
J., 20
(1982)745.
9. Harold R. Kaufman, Raymond S. Robinson, and Richard Ian Seddon, J. Vac. Sci. Technol., A5
(1987)2081.
10.
J. Reece Roth, Industrial Plasma Engineering, vol. 1 (Institute of Physics Pub., Philadelphia,
1995),
pp. 167-170.
11.
R D. Reader, D. R White, and G. C. Isaacson, J.
Vac.
Sci. Technol., 15 (1978) 1093.
12.
R Sigmund, Phys. Rev., 184 (1969) 383.
13.
E. Henshke, Phys. Rev., 106 (1957) 737.
14.
G. Wehner, /. App. Phys., 30 (1959) 1762.
15.
N. Laegreid and G. K. Wehner, J. Appl. Phys., 32 (1961) 365.
16.
A. N. Broers, W. W. Molzen, J. J. Cuomo, and N. D. Wittels, Appl. Phys. Lett., 29 (1976) 596.
17.
W. D. Sawyer, J. Weber, G. Nabert, J. Schmalzlin, and H.-U. Habermeier, J. Appl. Phys., 68
(1990)6179.
18.
M. Miyao, T. Sukegawa, and M. Hagino, J. Appl. Phys., 48 (1977) 1383.
19.
J. M. Millunchick, L. Hultman, and S. A. Barnett. J.
Vac.
Sci. Technol., A13 (1995) 1155.
20.
H.-C. Scheer, Rev. Sci. lustrum., 63 (1992) 3050.
21.
Ch. Steinbruchel, J.
Vac.
Sci. Technol., B2 (1984) 38.
22.
R. B. Jackman, Vacuum, 44 (1993) 239.
23.
M. Geis, S. W. Pang, N. E. Efremow, G. A. Lincoln, G. D. Johnson, and W D. Goodhue, in
Handbook of Ion Beam Processing
Technology:
Principles, Deposition, Film
Modification
and
Synthesis, edited by Jerome J. Cuomo, Stephen M. Rossnagel, and Harold R. Kaufman (Noyes
Publications, Park Ridge, New Jersey, 1989), pp. 219-240.
692 Chapter 5.3: Ion Beam Technology
24.
J. D. Chinn, I. Adesida, and E. D.
Wolf,
Appl. Phys. Lett., 43 (1983) 185.
25.
G. A. Lincoln, M. W. Geis, S. Pang, and N. N. Efremow, /
Vac.
Sci. Technol., Bl (1983) 1043.
26.
S. W. Pang, M. W. Geis, N. N. Efremow, and G. A. Lincoln, J.
Vac.
Sci. Technoi, B3 (1995) 398.
27.
S. W. Pang, G. A. Lincoln, R. W. McClelland, R D.
DeGraff,
M. W. Geis, and W. J. Piacentini,
J.
Vac.
Sci. Technoi, Bl (1983) 1334.
28.
Haruo Okano and Yasuhiro Horiike, Japanese J. of Appl. Phys., 21 (1982) 696.
29.
M. A. Bosch, L. A. Coldren, and E. Good, Appl. Phys. Lett., 38 (1981) 264.
30.
L Adesida, A. T. Ping, C. Youtsey, T. Dow, M. A. Khan, D. T. Olson, and J. N. Kuznia, Appl.
Phys.
Lett.,
65 (1994) SS9.
31.
A.T. Ping, L Adesida, and M. A. Khan, Appl. Phys. Lett., 67 (1995) 1250.
32.
R. Congxin, X. Deyuan, C. Guoliang, L. Xianghuai, and Z. Shichang, Nucl. Instr. and Meth. in
Phys. Res., B96{\995) 401.
33.
O. Wada, / Electrochem. Soc, 131 (1984)
2373.
34.
T. W. Drueding, S. C. Fawcett, Scott R. Wilson, and T. G. Bifano, Optical Engineering, 34
(1995)3565.
35.
Ren Congxin, Yang Jie, Zheng Yanfang, Chen Lizhi, Chen Guoliang, and Tsou Shichang, Nu-
clear
Inst,
and
Meth.
in Phys. Res., B19 (1987) 1018.
36.
S. Matsui, N. Takado, H. Tsuge, and K. Asakawa, Appl. Phys. Lett., 52 (1987) 69.
37.
J. Gao, Y. Boguslavskij, B. B. G. Klopman, D. Terpstra, G. J. Gerritsma, and H. Rogalla, Appl.
Phys.
Lett.,
59 (1991) 2154.
38.
J. Baglin, in Reference 23, pp. 279-297.
39.
T. Nakanishi, T. Toshima, K. Yanagisawa, and N. Tsusuki, IEEE
Trans,
on Magnetics, 15 (1979)
1060.
40.
T. Nakanishi, K. Kogure, T. Toshima, and K. Yanagisawa, IEEE
Trans,
on Magnetics, 16 (1980)
785.
41.
T. Toshima, T. Nakanishi, and K. Yanagisawa, IEEE
Trans,
on Magnetics, 15 (1979) 1637.
42.
S. Morishita and F. Okuyama, J.
Vac.
Sci. Technoi, A9 (1991) 167.
43.
L Miyamoto, S. T. Davies, and K. Kawata, Nuclear Inst, and Meth. in Phys. Res., B39 (1989)
696.
44.
M. J. Powers, personal communication, 1995.
45.
H. Coufal, Harold F Winters, H. L. Bay, and W. Eckstein, Phys. Rev., B44 (1991) 4747.
46.
O. Auciello, S. Chevacharoenkul, M. S. Ameen, and J. Duarte. /
Vac.
Sci. Technoi, A9 (1991)
625.
47.
E. Kay, F Parmigiani, and W Parrish, J.
Vac.
Sci Technoi, A5 (1987) 44.
48.
Harold F Winters, H. Coufal, C. T. Rettner, and D. S. Bethune. Phys. Rev., B41 (1990) 6240.
49.
J. A. Thornton, J.
Vac.
Sci Technoi, 11 (1974) 666.
50.
J. A. Thornton, Annu. Rev. Mater Sci, 7 (1977) 239.
51.
H. R. Kaufman, J.
Vac.
Sci. Technoi, A4 (1986) 764.
52.
M. Tan, Data Storage, 3(1) (1996) 35.
53.
T. Kobayashi, R. Nakatani, and N. Kumasaka. J. of Magnetism and Magnetic Mat., 81 (1989)
259.
54.
H. Jiang, Q. L. Wu, K. Tao, and H. D. Li, / Appi Phys., 78 (1995) 3299.
55.
Yasuhiro Nagai and Tomoyuki Toshima, J.
Vac.
Sci. Technoi, A4 (1986) 2364.
56.
Masaki Ueno and Shuji Tanoue, /
Vac.
Sci Technoi, A13 (1995) 2194.
57.
Yasuhiro Nagai and Masakatsu Senda, J. Appl. Phys., 64 (1988) 4108.
58.
M. N. Baibich, J. M. Broto, A. Fert, F Nguyen van Dau, F
Petroff,
P E. Etienne, G. Creuzet,
A. Friedrich, and J. Chazelas, Phys. Rev. Lett., 61 (1988) 2472.
59.
C. Kim, S. B. Qadri, H. Y. Yu, K. H. Kim, B. Maruyama, and A. S. Edelstein, /
Vac.
Sci Tech-
noi, AS (1990) 1407.
60.
C. Kim, S. B. Qadri, M. Twigg, and A. S. Edelstein, /
Vac.
Sci Technoi, A8 (1990) 3466.
References 693
61.
K. Inomata, Y Saito, and S. Hashimoto, J. of Magnetism and Magnetic Mat., Ill (1993) 350.
62.
R. Nakatani, T. Dei, T. Kobayashi, and Y. Sugita, IEEE
Trans,
on Mag., 28 (1992) 2668.
63.
Y. Wang, F. Z. Cui, W. Z. Li, and Y. D. Fan, J. of Magnetism and Magnetic Mat., 102 (1991)
121.
64.
S. B. Qadri, C. Kim, M. Twigg, P. Lubitz, and M. Rubinstein, J.
Vac.
Sci. Technol, A9 (1991)
512.
65.
T. Hamaguchi, H. Aida, S. Nakaqawa, and M. Naoe, J. Appl. Phys., 73 (1993) 6444.
66.
F Z. Cui, Y. Wang, H. Cui, W Z. Li, and Y. D. Fan, J. of Phys. D: App. Phys., 27 (1994) 2246.
67.
E. Kay and S. M. Rossnagel, in Reference 23, pp.170-218.
68.
J. E. Greene, Solid State Technology (April, 1987), p. 115.
69.
D. V. Vechten, G. K. Hubler, E. R Donovan, and F D. Correll, J.
Vac.
Sci. Technol., A8 (1990)
821.
70.
G. K. Hubler, D. V. Vechten, E. R Donovan, and C. A. Carosella, J. Vac. Sci. Technol., A8
(1990)831.
71.
J. S. Colhgon, J.
Vac.
Sci. Technol., A13 (1995) 1649.
72.
B. K. Kellerman, E. Chason, J. A. Floro, and S. T. Picraux, Appl. Phys. Lett., 67 (1995) 1703.
73.
X. D. Wu, S. R. Foltyn, R Arendt, J. Townsend, C. Adams, L H. Campbell, R Tiwari, Y Coul-
ter, and D. E. Peterson, Appl. Phys. Lett., 65 (1994) 1961.
74.
Paul C. Mclntyre, Kevin G. Ressler, Neville Sonnenberg, and Michael J. Cima, J.
Vac.
Sci. Tech-
nol., A14 (1996) 210.
75.
Satis Vacuum Corp., 10510 Olympic Dr., Dallas, TX 75220.
76.
D. J. Kester, K. S. Alley, D. J. Lichtenwalner, and R. F Davis, J.
Vac.
Sci. Technol., A12 (1994)
3074.
77.
K. J. Boyd, D. Marton, S. S. Todorov, A. H. Al-Bayati, J. Kulik, R. A. Zuhr, and J. W Rabalais,
J.
Vac.
Sci.
Technol.,
A13 (1995) 2110.
78.
J. Robertson, Surface and Coatings Tech., 50 (1992) 185.
79.
M. J. Mirtich, D. M. Swec, and J. C. Angus, Thin Solid Films, 131 (1985) 245.
80.
Dan Nir and Michael Mirtich, J.
Vac.
Sci, Technol., A4 (1986) 560.
81.
S. Aisenberg and R. Chabot, J. Appl. Phys., 42 (1971)
2953.
82.
M. J. Mirtich, D. M. Swec, and J. C. Angus, NASA Technical Memo, TM-83743, (NASA, Wash-
ington, D.C., 1984).