621

INDUSTRIAL APPLICATIONS OF ION IMPLANTATION

,.K. HTPAVrCIF'ý
Zymet, Inc., 33 Cherry Hill Drive
Danvers, MA 01923

ABSTRACT

The use of ion implantation for

non-semiconductor applications has evolved steadily
over the last decade. To date, industrial trials
of this technology have been mainly directed at the
wear reduction of steel and cobalt-cemented
tungsten carbide tools by high dose nitrogen
implantation. However, several other surface
sensitive properties of metals such as fatigue,
aqueous corrosion, and oxidation, have benefitted
from either i)direct ion implantation of various
ion species, ii)the use of ion beams to "intermix"
a deposited thin film on steel or titanium alloy
substrates, or iii)the deposition of material in
conjunction with simultaneous ion bombardment.

This paper will concentrate on applications

that have experienced the most industrial trials,
mainly high dose nitrogen implantation for reducing
wear, but will present the features of the other
ion beam based techniques that will make them
appear particularly promising for future commercial
utilization.

INTRODUCTION
It has been approximately ten years since the first
publications by Hartley and colleagues at Harwell of the effect
of high dose nitrogen implantation on wear reduction [1]. Since
that time, numerous laboratories have gotten involved in R&D
applications involving both mechanical and chemical
surface-sensitive properties, many of which are presented in
this proceedings. In spite of the large number of properties
(e.g., fatigue, aqueous corrosion, high temperature oxidation)
that have shown significant improvements due to ion
implantation, to date, the vast majority of industrial trials
has remained in the area of wear.

WEAR REDUCTION OF INDUSTRIAL TOOLING

Wear improvements have been seen in both steel as well as

Co-cemented WC tools under a variety of applications. It is
important to note that nitrogen implantation has been found
ineffective for those applications involving high operating
temperatures such as found in chip forming applications. This
is because of the instability of the nitrogen-defect structures
that are thought to be responsible for improving the wear
resistance. There is direct evidence from several groups
[2,3,41 that there is an anomalous retention of nitrogen in the
wear scar of implanted wear samples and a persistence of wear
resistance after the removal of material corresponding to 10-50
times the original range of nitrogen. There is however still

Mat. Res. Soc. Symp. Proc. Vol. 27 (1984) CElsevier Science Publishinq Co., Inc.
622

considerable debate as to what mechanism or mechanisms are

responsible for this persistence. Dearnaley et al. [2] have
suggested that the nitrogen decorates dislocation structures and
migrates under the locally high temperatures produced at
interacting asperities during wear. Baumvol and associates [51
have seen evidence for nitride and carbo-nitride formation
following implantation and subsequent dissolution at elevated
temperatures using conversion electron Mossbauer spectroscopy.
These results suggest that such precipitates may pin
dislocations, yet still be able to dissolve so that nitrogen
migrates inward during wear. Another possibility for this
anomolous retention is that nitrogen may be continuously
transferred from one wear surface to another as the wear front
advances.

Despite these uncertainties as to cause there are enough

industrial tests to demonstrate the efficacy of the process. To
date Harwell has worked with over a hundred companies in the
U.K. involving industrial trials of nitrogen implanted tools and
now several U.S. companies are initiating trials. As several
other papers in this proceedings show, other ion species such as
high dose Ti implantation can also have a marked effect on wear
reduction. However because few of these have yet been tested
industrially they won't be discussed here.

Implantation is an attractive technique for treating

industrial tooling where one or more of the following criteria
are met: i)there is a need to avoid dimensional changes of
the tooling, ii)the tooling can't accomodate the elevated
temperatures associated with other surface treatments such as
PVD or CVD because of possible distortion, or iii)conventional
antiwear coatings have delamination problems during use.

Table I is a compilation of several applications of nitrogen

implantati?9 for wea 5 resistance. The doses used range from
2X to 6X10 ions/cm at energies of 50keV-100keV. In addition
to the criteria given above most of the successful usages
involve relatively expensive tooling or high costs of lost
throughput associated with downtime of a production line. These
points are illustrated by the first examples listed in Table I.

An example of a precision tool costing several hundred

dollars is shown in Fig. 1. It is a D2 steel scoring die used
for the production of flip tops for aluminum beverage cans. One
large company currently evaluating implantation uses up to a
hundred of these in a week before rejecting them due to wear of
the fine pattern on the face of the tool. This particular tool
normally produces between three to five million tops before
wearing out whereas initial tests of N-implanted tools have
produced seven to seventeen million tops before wearing out.
Another tool shown in Fig. 2 is also made of D2 tool steel and
is used to form the bottom of aluminum beverage cans. It is
very highly polished and normally tends to pick up aluminum from
the can surface thereby weakening the can as well as requiring
time consuming repolishing several times during its lifetime.
An initial trial of a nitrogen-implanted tool has shown markedly
reduced metal pick-up during operation with further tests
underway.
 623

1-4 E
.,--I
0)
0
4) .- 4 41
0 Et,
E
0
Li 04 M
8) 0
c44
'0 0 >1

U) 0
4)- 4o
cc
44-,.-4 0
8) 0 0 Szn 0
Li L4 8)
0
,,-4
0
.C 0
w 4) 41
4-,-
-0 1-4 4-4
--0I
a)
(DU 0
0 (D
,-- '0 Cý -4
0
a)
0 4.1
0 0
4) 0 w-
0 4-4 04 I 0
x ý4 x 04 0
(N A (N
L1

04 041
-4 .,j 4- 'o
4)
Cl) 8j)
8) 8) 0
8) -4 1, 0
U)
U)
-4
8) 8) 0
Ci)
CI) 4) >
H 0 LW
8)
I-4
0'
0 4J
0 00 0
P 8)04 tio
04-
0 C.) E4
(N '0
8) 0
m
3) a) 04
41
8) cq
4) 541 0
0 04
'0 a)
0
0' t" U,
-4 e-4 ,-.4 .0
4J4 u
8) 8).
H
U) 4-4
0 8)• .) 4) 0
04 E)
)4. (a U)
U) 82)
8) 0 U)
z .,f. 5
0 .,-4 04
U) I04 C .0
0
ý-4 E 0 8).
0 4)
0M Cl) 0 4) (04
U)
En4
r_ C)
L-4
,,-- 0
C.
4)
1-4 U)
0 0
0 -4
p
624

0) w0 U)

.4
4-4 -- 4 X
8p
)

r8)8) - 41 8) 0

>)L-'- a)
'0
L C a) . C ,' )
8) -4) 018) z OA U
'04
0 .10z
'0 C) 4) C 8 .- 04 LA

0C LA *8 0 0 44 0 Li(0 04
0~ - •8
.0a
4J 0 '40 8) C
o 0 • -
Lfl 8)
OLA~~~-
Lir•
4 ~ >o
"
.0
-4
0
4)1•

x4)
4J48

04L 4 w x 0
m4 V
0 LO --L
0'0 U) C)
.08 0 0 ) 00 8
0
'04)~ LA 4 4N a)-

• >4 .C 8) LA C) 0

8)C i8) 0'8 0 4)-

•),) 44-1,-• -4 .0 4)
80 ,-0, 8)~ -0
04)M
0
><0 0
'- . aL O LA C 0
4a-8 4.) "
LA~4. ) 5- L8) 0 NC
5 -4.. . C '4 '

8) 04--0 . 4 .u0 m 0 0 4 1
0'"0 4 0
Cw 4)

0 (a U) w U,
•i0) ,) 4•) C)0
CC 0) 4)
8)0--1 0 44
C)
C)1 C) 8)4 8)-
(n Q4 8) '0 0q
U, -a
4 .44-
I 5 I I I H )
a-0 L8
048 )--
OLA 0 (N0
C) C)4 C) C) CU) C ) 0C 0) -1 w- *414

rA a8)u , Q
C U) 0'aC 4 8)
0 LA -40 8)
U).A1 4 0 0 w4) . - .-
-4 ) 8) .1-1 CU) U, . C)41
-4 1 Q) 8)--4 U, : n .)a
4-4 $-

8) c4
.0 U) 8)1- 04 0
8) z 0 0 0 H '0 0
LA
a41--4 01 -4 C) .4- 8) 4
-4
 625

Fig. 1 Scoring die used for Fig. 2 D2 Steel nose

producing lids of flip punch used for
top beverage cans forming bottoms
of aluminum
beverage cans.

Precision Co-WC punches and dies have previously been

reported to wear longer following N-implantation by Westinghouse
researchers[61. In industrial trials with a major electronics
manufacturing company, we have also found a 2X-4X improvement in
the lifetime of N-implanted precision WC-Co punches such as used
in the production of the so called lead frames that hold and
provide electrical connections to integrated circuit chips as
shown in Fig. 3. The cost of an individual punch as shown in
Fig. 3 is approximately $100-200 and the cost of an entire
progressive punch and die set for this application is typically
$50K-100K. Downtime is costly in terms of both resharpening
time (typically 1 shift) and lost production.

Fig. 3 Precision WC-Co punch and lead frame used in integrated

circuits.
626

In all of these examples, a 2X improvement in tool lifetime

would give a return-on-investment time of less than a year for a
dedicated implanter which is less than the two to three year
return on investment time for capital equipment that most
industries require. In addition to reduced wear, other side
benefits may be equally important to introducing implantation
technology into these industries. The use of nitogen
implantation for increasing the lifetime of Co-cemented WC
printed circuit board (PCB) drills is another application
previously reported by Westinghouse researchers (6]. In some
recent tests we have done in conjunction with Precision Carbide,
a major manufacturer of PCB drills, the surface temperature of
N-implanted drills was measured to be approximately 175 0 F as
compared to 3500F for untreated drills. This is another
observation of reduced friction of N-implanted Co-WC tools
reported initially by Dearnaley [2]. It is expected that
lowered drill temperatures would significantly decrease or
eliminate the smearing of the resin in the PCB on the walls of
the drilled holes, which presently requires a separate
"desmearing" or cleaning process utilizing hot sulphuric acid to
remove the smeared resin. The possiblility of eliminating this
toxic and corrosive process is expected to be another strong
driving force for adoption of the implantation process in this
industry. We have also observed extended life and reduced
friction of N-implanted WC dental burs. Clinical evaluation of
this application by several groups is currently underway. Here
again, reduced temperatures (i.e., reduced trauma) may be as
important or more important than reduced wear for acceptance.

It is obvious from a perusal of Table I that the diversity

of applications is extreme and acceptance will be based not only
by longevity increases but also by economics and industrial
attitudes. For example, circular cross section wire drawing
dies cost only a small fraction of shaped dies thereby
significantly influencing the economics of treating them. A two
inch diameter round die typically costs 3 to 5 dollars which is
comparable to implantation costs using dedicated equipment and
therefore wouldn't necessarily justify treatment costs whereas
shaped dies would. At this time, it is too early to accurately
gauge whether implantation technology for metals will spread
predominatly through service centers (e.g., heat treating
facilities) or via inhouse equipment. The first alternative
will obviously be most appealing to low volume users and many of
the industries who have traditionally had tooling both made and
surface treated out-of-house. It is interesting to note that
the initial purchasers of metals implanters in the U.S. are
firms intending to offer service work. Other industries have
great misgivings about having their expensive tooling out of
their direct supervision and also have concern about fast turn
around times for treatment of tools used in high volume
production runs. The combination of these concerns, coupled
with higher volume and concern with proprietary usage and
development, warrants buying dedicated equipment.

The majority of the applications listed in Table I involve

production tooling but the advantage of allowing processing on
an as-finished part without altering an existing production line
is also pertinent to consumer items. A particularly promising
application of this sort involves surgical implants such as hip
 627

joint prostheses, for which titanium alloys are being

increasingly used. The metal femoral component bears against an
acetabular cup of high molocular weight polyethylene. The
rubbing of the metal ball against the polyethylene pocket is
thought to remove protective oxides from the metal surface so
that it is attacked by body fluids thereby releasing
undesireable (metal) ions into the body. The wear rate is such
that a typical surgical implant will last less than 10 years and
surgical replacement poses significant problems. Dearnaley [101
has reported a dramatic reduction in the wear rate of Ti/6A1/4V
alloy used in surgical implants by both direct nitrogen
implantation, which gives a 10OX reduction in wear, or by ion
beam mixing of Sn, which gives a 1000X reduction. These results
were found in laboratory tests simulating the geometry and fluid
conditions found in the body. Williams et al. [13] have
reported similar findings using nitrogen implantation.
Considering there are approximately 100,000 operations of this
kind each year in the U.S. alone, even a small wear reduction
would have an immense benefit to recipients.

ION BEAM MIXING AND ION BEAM ENHANCED DEPOSITION

Due to the additional cost and complexity of implantation

equipment incorporating mass analysis, it is likely that future
commercial applications requiring nongaseous species will be
done using ion beams in conjunction with conventional thin film
deposition, such as in ion beam mixing, as an alternative to
direct high dose implantation.

The question of how effective ion beam mixing is versus

direct implantation must presently be addressed on a case to
case basis, but some studies have shown that higher alloy
concentrations can be achieved than in sputter-limited high-dose
implantation. An extension of this idea involves simultaneous
ion bombardment and thin film deposition. This technique has
been studied by a number of researchers [14,15,16,17] for the
purpose, at least partially, of understanding the role ions play
in conventional PVD processes. This combination allows thicker
alloyed regions to be produced than by either direct
implantation or ion beam mixing. It has been recognized for
some time that the presence of energetic ions has a synergistic
effect on thin film growth, and this topic has been recently
reviewed by Takagi[18]. There are several aspects of film
growth that have been beneficially influenced by ion bombardment
during deposition including i) film nucleation and growth[151,
ii) adhesion[15J, iii) internal stress[19], iv) morphology, v)
density[20], and vi) composition.

One of the most striking changes in structure produced by

ion beams has been the deposition of carbon films with
"diamond-like" properties observed either after ion bombardment
of carbon films or the collection of energetic ionized
hydrocarbons. It was first seen in 1971 by Aisenberg and
Chabot[211 and later by many other groups. The carbonaceous
films produced by these methods (termed i-C denoting the role of
ions) appear transparent with a high refractive index,
are quasi-amorphous, are very hard, and have high electrical
resistivity. Weissmantel et. al.[16] have reviewed this area of
producing hard coatings by ion beam techniques, including the
production of cubic boron nitride, another extremely hard
628

material second only to diamond in hardness and of obvious

interest for cutting surfaces. Weissmantel et. al. [22,23]
evaporated pure boron in a residual atmosphere of nitrogen, and
subsequently ionized and accelerated the resultant species at
0.5 - 3.0 keV onto various substrates. IR absorption spectra
confirmed that B-N bonding states predominate while transmission
electron microscopy and selected area electron diffraction
patterns indicate a quasi-amorphous structure; however high
energy deposits contained small crystallites of about 100
Angstrom diameter which had a lattice constant corresponding to
cubic BN. Shanfield and Wolfson[24] present hardness, x-ray,
and Auger analysis data supporting the production of cubic BN by
using an ion beam extracted from a borazine (B N3 H6 ) plasma.
Satou and Fujimoto have also recently publisheA results
suggesting the formation of cubic boron nitride by the
deposition of elemental boron and the simultaneous implantation
of 30 keV nitrogen ions[25].

These examples serve to show the extremely beneficial

effects ion beams can have when used in combination with
conventional deposition processes. Clearly, we are only at the
beginning of developing wear, corrosion and oxidation resistant
coatings. This development work will be accelerated now that
commercial equipment is available for such processing.

REFERENCES

1. N.E.W. Hartley, W.E. Swindlehurst, G. Dearnaley, and J.F.

Turner, J. Mater. Sci. 8, 900-904 (1973).

2. G. Dearnaley, Rad. Effects 63, 1 (1982).

3. S. Fayeulle, D. Treheux, P. Guiraldenq, T. Barnavon, J.

Tousset, and M. Robelet, Scripta Met. 17, 459-461 (1983).

4. F.Z. Cui, H. D. Li, and X. Z. Zhang, Nucl. Instr. and Meth.

209/210, 881-887 (1983).

5. C.A. Dose Santos, M. Behar, J.P. De Souze, and I.J.R.

Baumvol, Nucl. Instr. and Meth., 209/210, 907-912 (1983).

6. R.E. Fromson and R. Kossowsky in: Metastable Materials

Formation by Ion Implantation, S.T. Picraux and W.J. Choyke,
eds. (Elsevier, New York 1982) pp. 355-362.

7. N.E.W. Hartley, in: Ion Implantation Case

Studies-Manufacturing Applications, Harwell Report AERE-R
9065 (1978).

8. G. Dearnaley, private communication.

9. N.E.W. Hartley in: Metastable Materials Formation by Ion

Implantation, S.T. Picraux and W.J. Choyke, eds. (Elsevier,
New York 1982) pp. 355-362.

10. G. Dearnaley in: Proc. of NATO Advanced Study Institute on

Surface Engineering, Les Arcs France, July, 1982. R.
Kossowsky and S. Singhal, eds.
 629

11. G. Dearnaley in: Ion Implantation Metallurgy, C.M. Preece

and J.K. Hirvonen, eds. AIME, Warrendale, PA (1980).

12. A.H. Deutchman and R.J. Partyka in: Proc. of NATO Advanced
Study Institute on Surface Engineering, Les Arcs France
July, 1982. R. Kossowsky and S. Singhal, eds.

13. J.J. Williams, G.M. Beardsley, R.A. Buchaman, and R.K.

Bacon, This Proceeding.

14. J.M.E. Harper, and R.J. Gambino, J. Vac. Sci. Technol. 16

1901-1905 (1979).

15. L. Pranevicius, Thin Solid Films 63, 77-85 (1979).

16. C. Weissmantel, G. Reisse, H.J. Erler, F. Henny, K.

Bewilogua, U. Ebersback and C. Schurer, Thin Solid Films 63,
315-325 (1979).

17. J. Colligon, and A.E. Hill. 6th National Conference on

Interaction of Atomic Particles with Solids (Minsk USSR,
Sept. 1981 ).

18. T. Takagi, Thin Solid Films 92, 1-17 (1981).

19. D.W. Hoffman, and M.R. Gaerttner, J. Vac. Sci. Technol. 17

425-428 (1980).

20. R.F. Bunshah, J. Vac. Sci. Technol., 11, 633 (1974).

21. S. Aisenberg and R. Chabot. J. Appl. Phys. 42, 2953-2961

(1971).

22. C. Weissmantel, K. Bewilogua, D. Dietrich, H.J. Erler, H.J.

Hinneberg, S. Klose, W. Nowick and G. Reisse, Thin Solid
Films 72, 19-31 (1980).

23. C.J. Weissmantel, Vac. Sci. Technol. 18, 179 (1981).

24. S. Shanfield, S. and R. Wolfson, J. Vac. Sci. Technol.

(1983).

25. M. Satou and F. Fujimoto, Jap. Jour. of Appl. Physics. 22(3)

L171-L172, (1983).