Substrate
Selection for Tools Used with Hard Thin Film Coatings
By
Dr. Tohru Arai, Technical Advisor and Horst M. Glaser, Product Manager,
TD Center, Columbus, IN
(Reprinted
with permission from MetalForming Online, June 1998 issue)
The
use of thin hard coatings already has found a niche in the metalforming
industry. However, the full potential of thin hard coatings has
not been fully realized, especially when it comes to proper substrate
material selection.
Any discussion of tool steels for use with thin
hard coatings must take into consideration coating conditions, especially
high temperature processing, and post hardening conditions. Properties
of tool steels can change drastically during coating and heat treatment.
Therefore, substrate properties required when using coatings will
be different than those employed for noncoated applications--both
for tool development and use.
Thin Hard Coatings for Tooling
Methods used to produce thin hard coatings on tool
substrates include:
- CVD—chemical
vapor deposition.
- TD—thermal
diffusion.
- PVD—physical
vapor deposition.
- PACVD—plasma
assisted CVD.
Thin hard coating methods can be divided into two
groups, based upon coating process temperature: a) high temperature
process and b) low temperature process. This article will not discuss
the PACVD method because it has not been used by American industry.
As currently used in the United States, CVD and
TD are both high temperature processes. PVD is a low temperature
process. In Japan and Europe, low temperature CVD and TD, as well
as high temperature PVD processes, are available.

Fig. 1 - -Relation between
coating temperature for thin hard coating and that for ordinary
hardening.Coating & Substrate Hardening during Cooling (Q+T)
As shown in Fig. 1, high temperature
coatings are done at a temperature compatible with the hardening
(austenitizing) temperature of steels, and therefore need substrate
hardening to occur during the cool down from coating temperature
or with reheating. Process sequences for toolmaking are summarized
in Fig. 2.

Fig. 2 -- Process sequences
for tool making by thin hard coating.
CVD coating, especially for large tools, usually
is done by sequences (C) and (E) since high enough cooling rate
for quench hardening of most steels cannot be accomplished during
cooling from the coating process temperature. Sequences (D) and
(E) usually are applied to loosely toleranced tooling, since the
size movement control can be more difficult due to the crystal structure
transformation of substrate steels.
Hardening after coating procedures is not required
for PVD coatings. However, the PVD temperature (not starting temperature,
but true temperature during coating) should be selected relative
to the optimum tempering temperature of each substrate steel used.
Otherwise, substrate hardness after the coating operation could
be lower than its original hardness.
CVD coating can be done using a wide variety of
coating temperatures, and on a wide variety of substrates. However,
switching of coating temperatures and gas composition to harmonize
with the substrate materials being used is not easy to accomplish
in production by commercial treaters. Therefore, commercial CVD
coatings usually remain within D-series die steels, high-speed steels
and cemented carbides.
Tooling Materials
Table
1 lists the types of tooling materials used for metalforming,
their capability for thin hard coating, hardness in use and applications.
Most tooling materials can be coated by at least one and usually
all of the thin hard coating processes.

Key in selecting substrates and types of coating
used is the substrate hardness after coating. It must be high enough
to ensure support of the coating when a mechanical load is applied
to the tooling during forming. In rare cases, coating of materials
with low substrate hardness, such as copper alloys and cast iron,
without hardening is successful when forming pressures are very
low.
Material Properties Related to Tooling
Failures
The most critical point of attention for all thin
hard coatings is the role of plastic deformation of substrates.
The potential for large plastic deformation of the substrate by
high loads applied in metalforming can induce cracking in coatings
that have low ductility. This damage results in peeling of the coating
leading to wear and galling problems.
These types of failures are thought to be a failure
of the surface coating, but in actuality are failures of the substrate
surface directly under the thin hard coating. This misunderstanding
usually results in discontinuing the use of coatings rather than
changing the substrate to one that will support the coating.
Another failure in thin hard coated tooling results
in spalling (peeling) or chipping of the coated layer. This causes
damage to the product material that may not have happened with softer
uncoated tool surfaces. A spalling problem frequently is encountered
with PVD. Chipping will occur near the end of practical tool life
with CVD and TD.
The plastic strain of a substrate in which cracking
of the coating was confirmed can be concluded to be about one percent
and is the same for both TD-coated steels and PVD-coated steels.
Similar values can be expected for CVD-coated steels. It has been
confirmed that there is not a large difference of the critical strain
between TD-VC coatings formed on various steels, ranging from carbon
tool steels to high-speed steels.
A shorter life of the thin hard-coated tooling compared
with uncoated tooling is mainly attributable to spalling from poor
adhesion strength of the coating, especially in PVD coatings. This
is the result of either improper processing conditions or a failure
of the coatings caused by plastic deformation of the substrates.
These two important facts always should be kept in mind to obtain
successful application of thin hard coatings.

Fig. 3 --
Correlation in occurrence of failures in coating, substrate and
interface of thin hard coated materials.
Another point that should be made is
the strong correlation in failures and damage phenomena between
coating substrate and interface, as exemplified in Fig.
3. It is important to keep in mind that failures of tooling
are induced by very complicated phenomena, and therefore the effect
of substrate material selection on successful applications is very
complicated.
Relation of Substrate and Properties
of Coated Tooling
Roughly speaking, the surface properties of thin
hard coated materials usually are independent of the type of substrates
used so long as the substrates have enough strength. The change
of substrate materials results in a change of the micro-structures
of coated layers, such as size of crystal grains, preferred orientation
of crystal growth, surface roughness, etc. Strictly speaking, the
surface properties can change with the same kind of coating--even
by the same coating method--when the substrate is changed. However,
this problem will be ignored here since the problem may be very
complicated because of the correlation between coating conditions
and substrate materials.
Major properties, such as toughness and fatigue
strength, are determined by those properties found in the substrate
materials, which are inherent to each uncoated material. In some
cases, the effect of coating procedures and the existence of a strong
coating on substrates should be taken into consideration.
Interface properties can be determined by the cleanliness
of substrates, especially with PVD, and the reaction between the
coating and substrate during the coat formation, as well as the
kind of substrate and coating material. From this point on, only
those problems that are highly related to substrate selection will
be discussed.

Fig. 4 -- Relation between
compression strength and hardness of tool steels and cemented carbide.
The first consideration for good selection of substrate
materials is that substrates have the hardness (compression yield
strength) sufficient to prevent the substrates from being subjected
to large plastic deformation (more than one percent). This hardness
value can be roughly estimated from Fig. 4, which
explains the relationship between steel hardness and compression
strength; and Fig. 5, which introduces the hardness
range usually selected for various tooling materials. One percent
proof strength, not shown in Fig. 4, can be estimated roughly to
be larger than 20 to 30 percent of 0.2 percent proof strength.

Fig. 5 -- Approximate hardness
in use of various tooling materials.
A problem in substrate selection for metalforming
applications is the difficulty of estimating the compression stress
that actually is applied to each tool component. Note that the hardness
of substrates after coating may be somewhat different from the original
heat treating hardness.
Normally, it is said that PVD coating does not affect
substrate hardness. However, hardness drops can occur. A larger
hardness drop is observed on low alloy steels and no drop occurs
with the high-speed steels listed in Table
1.
Other published results show about HV 150 drop on
M2 high-speed steel; however, high-alloyed steels like high-speed
steels are better selections to minimize a possible hardness decrease.
Hardness decreases at the substrate surface can
be large with TD and CVD processing because of the decrease of carbon
(decarburization) for formation of the carbide coatings. This phenomenon
can be more serious in substrate materials containing only a small
amount of "free carbon" at coating temperatures. The free
carbon, carbon that easily can combine with atoms of carbide-forming
elements (V, Ti, etc.), means the carbon in the austenite matrix
of steels and carbon in the cobalt binder of cemented carbides.
The degree of hardness drop changes with substrate
materials and conditions of coating and post hardening. No drop
is observed in high-carbon, low-alloyed steel such as W1 and O1,
and a large drop is seen with low carbon, medium alloyed steel such
as S7.
In cemented carbide decarburization produces the
special crystal phase called "h
phase" and makes tooling very brittle. Selection of proper
thickness and carbide without large deterioration is important.
Decarburization of steels does not make any special
phase, for example an "h
phase." However, the lower substrate hardness, as well as large
tensile residual stress at the substrate surface decreases fatigue
strength. The harmful effect of decarburization can be recovered,
although not completely, during post hardening by thermal diffusion
of carbon from the core of the substrate toward the surface.

Fig. 6 -- Fatigue strength
of thin hard coated D2 steels
An improved strength derived from the existence
of hard coatings on soft substrates is obvious in the fatigue strength
results at cycles as shown in Fig. 6. CVD and PVD
effectively increased cycles to failure. TD also improved strength
of D2 but only when TD-coated D2 was tempered at high temperature.
In the case of TD-coated steels, post hardening is a very effective
method to improve fatigue strength. Therefore, post hardening is
highly recommended for tooling used in severe working conditions
where high compression stress is loaded onto tooling, resulting
in fatigue failure of the tool.
Considerations for Selection
Proper substrate selection is accomplished by considering
three major points. Remember, wear is the responsibility of the
thin hard coating, not the substrate.
- Tooling
Production Costs — lower substrate material cost, better
machinability and grindability, easier and less expensive heat
treatment cost, and availability.
- Coating
Procedure — easier coating operation and less distortion.
- Properties
of Coated Materials — strong adhesion strength, good tribological
properties of coating, high compression strength, high toughness,
and high fatigue strength.
Some discussion
has occurred concerning key properties. Now, we will present additional
information on properties and on important problems in other areas.
As previously
noted, substrate hardness sufficient to prevent plastic deformation
of the substrates by an applied load is the primary consideration
for successful thin hard coating applications.
Second, to minimize
substrate distortion, hardening will be the biggest concern when
selecting a substrate for use with high-temperature processing conditions.
Fig.
7 Effect of cooling
rate in quenching from TD processing temperature on warping of D2
punches.
As shown in Fig. 7, deformation
(shape change) is highly affected by cooling rate in the quenching
operation. The slowest cooling rate can provide the smallest deformation.
Consequently, air hardening steels most often are recommended for
CVD and TD.
Deformation can be caused not only by improper quenching,
but also by the heterogeneity of the substrate material structure.
Powder metallurgy steels are highly recommended for tooling that
requires minimum out of roundness, because of their very uniform
distribution of fine carbide particles. Cemented carbides also are
recommended for the same reason.
Furthermore, cemented carbides are the best material
to minimize a size-movement problem. They eliminate the need to
match good coating condition with preliminary hardening condition.
This duplication of heat treating conditions is needed with steel
substrates to minimize the change in the amount of retained austenite
and martensite to control distortion before and after the coating
process.
Normally, surface properties of coated materials
are almost independent of the substrate material being selected.
PVD does have a relatively large dependency of bonding strength
between coatings and substrate materials. With PVD, the highly alloyed
steels are said to create the best bond strength. For TD and CVD,
dependency of bonding strength on substrate materials is so low
that it can be eliminated as a selection factor.
Fig. 8 -- Comparison
of wear resistance and toughness of some kind of tool steels (data
from Crucible Steel's catalog).
Since wear is the responsibility of the selected
coating, the information offered by material makers, as shown in
Fig. 8 and Fig. 9, should be used
to select materials with the highest quality in other properties
rather than wear resistance. For example, if the tooling will be
used under high impact conditions, A2 or CPM M4 are good candidates
with substrate hardness of HRC 60 and 64, respectively.

Fig.
9 -- Comparison of hardness and toughness of some kinds
of cemented carbides (data from Vista's catalog).
If the tooling fails due to fatigue failure, the
steels can be hardened to higher hardness by high-temperature tempering.
Tempering at high temperature decreases residual tensile stress
at the substrate surface to increase fatigue strength. Toughness
is improved a great deal.
Also, high-temperature tempering can decrease the
retained austenite in steel substrates. This results in higher compression
yield strength and makes size movement control much easier.
Only some cold working die steels, such as D2 and
D7, can be tempered at low and high temperature to reach hardness
levels of HRC 57-62. Modified steels recently developed in Japan
and the United States are highly recommended for TD and CVD coating.
Higher hardness can be obtained easily by high-temperature tempering.
Furthermore, they feature higher toughness, better machinability,
deeper hardenability, etc., when compared to standard D2.
The selection of good quality materials is more
important than the type of material selected for large tooling.
Very large differences, much larger than those obtained between
different materials, can be observed within a single type of steel.
These differences mainly are caused by differences in the size and
shape of carbide particles and their heterogeneous distribution
(segregation) in steels. Therefore, powder particle steels usually
are much better than wrought steels for large tooling.
Rating of Tooling Materials
Table 2 shows
a rating of various tool materials after TD coating in terms of
fatigue strength, toughness and compression strength. This can be
a useful starting point for tool steel selection, when used in conjunction
with other factors such as price, availability, machinability, ease
of heat treatment, etc. Please keep in mind the ranking is a simple
guideline to be used as a rough comparison of materials, and that
the properties of materials are highly influenced by heat treating
condition and the quality of bar stocks, etc.
Table
2 -- Rating of Material TD Coated
| Hardness
RC |
Material |
Fatigue |
Toughness |
Compression |
| 54 |
CPM
9V* (HT) |
C |
A+ |
E
|
| D2
(HT) |
C |
B |
E |
| S7
(LT) |
D |
A |
E |
| S7*
(LT) |
D |
A |
E |
| VANADIS
4 (HT) |
B |
A+ |
E |
| VASCO
TUF (HT) |
C |
A+ |
E |
| 56 |
D2*
(HT) |
C
|
B |
D |
| VANDIS
4 (LT) |
D |
A |
D |
| 58 |
A18
(HT) |
B |
B |
B |
| A2
(LT) |
D |
B |
D |
| CPM
10V (HT) |
B |
B |
B |
| CPM
15V (HT) |
B |
C |
C |
| D2
(LT) |
D |
C |
C |
| MATRIX
I* (HT) |
B |
A+ |
B |
| VANADIS
10 (HT) |
B |
A |
B |
| VANADIS
10 (LT) |
C |
A |
B |
| VASCO
DIE (HT) |
B |
B |
B |
| VASCO
TUF* (HT) |
B |
A+ |
B |
| 60 |
CPM
10V* (HT) |
A |
A |
B
|
| ASP
23 (HT) |
A |
A |
B |
| D2*
(LT) |
C |
C |
B |
| MATRIX
II* (HT) |
B |
A+ |
B |
| VANADIS
4* (HT) |
A |
A+ |
B |
| VASCO
WEAR (HT) |
B |
B |
B |
| 62 |
VASCO
WEAR* (HT) |
B+ |
B |
B |
| ASP
23* (HT) |
A+ |
A+ |
B |
| CPM
10V* (HT) |
A |
C |
B |
| CPM
15V* (HT) |
A |
D |
B |
| CRU
WEAR* (HT) |
B+ |
B |
B |
| M2*
(HT) |
B+ |
C |
B |
| M4*
(HT) |
B+ |
D |
B |
| VANADIS
10* (LT) |
A |
A |
B |
| VANADIS
4* (LT) |
A |
A+ |
B |
| 64 |
ASP
30* (HT) |
A+ |
A |
A |
| CPM
M4* (HT) |
A+ |
A |
A |
| M50*
(HT) |
A |
B |
A |
| M52*
(HT) |
A |
B |
A |
| VANADIS
10* (HT) |
A+ |
A |
A |
| 66 |
CPM
REX 20* (HT) |
A++
|
B |
A+
|
| CPM
REX 45* (HT) |
A++ |
B |
A+ |
| CPM
T 15* (HT) |
A++ |
B |
A+ |
| MICRO
MELT T15* (HT) |
A++ |
B |
A+ |
| MICRO
MELT HS20* (HT) |
A++ |
B |
A+ |
| 68 |
ASP
60* (HT) |
A++
|
B
|
A++
|
| CPM
REX76* (HT) |
A++ |
B |
A++ |
| M42*
(HT) |
B+ |
E |
A++ |
| MICRO
MELT HS76* (HT) |
A++ |
B |
A++ |
| T15*
(HT) |
B+ |
E |
A++ |
| >68 |
Cemented
Carbide |
? |
E |
A+++ |
* Post Hardened
The rating primarily
was based upon the rating of ordinarily hardened materials. However,
the following information also was taken into consideration:
A)
In TD coating not accompanied with post hardening, there is a
tendency that substrate hardness may be slightly lower than that
in its originally hardened state. This is due to the slower quenching
rate employed in TD processing and possibly a larger amount of
retained austenite, etc.
B)
Carbon consumption at the substrate surface resulting in lower
hardness at the surface than inside the substrate can occur with
some steels.
C)
Residual tensile stress may occur at the substrate surface in
some steels under some TD coating conditions.
D)
Strengthening by the existence of the coating (items A through
C) may somewhat deteriorate the compression yield strength, toughness
and fatigue strength.
This can be mitigated by post hardening. Item D
may be effective only by high-temperature tempering and post hardening.
The ratings for TD-coated materials are not far
from that of CVD-coated materials. In CVD, unlike TD, carbon atoms
for carbide formation are supplied from both the substrates and
the coating medium (gas). However, fairly large amounts of carbon
atoms in substrates are consumed to develop a carbide layer. This
can cause a decarburized layer and residual tensile stress at the
substrate surface. Therefore, the ratings for TD-coated materials
shown in Table 2 can be used
for CVD coating with little or no modification.
The rating for ordinarily hardened materials, shown
in Table 3, can be applied
to PVD with the addition of an evaluation on bonding strength of
each material, which also is highly dependent on proper PVD coating
conditions, as long as the PVD coating could be done with no substrate
hardness drop.
Table
3 -- Rating of Material Ordinarily Hardened
Hardness
RC |
Material |
Fatigue |
Toughness |
Compression |
| 56 |
D2 (HT) |
D |
B |
D |
| S7 (LT) |
D |
A+ |
E |
| 58 |
A18 (HT) |
C |
B |
C |
| MATRIX
1 (HT) |
B |
A+ |
C |
| VASCO DIE
(HT) |
C |
B |
C |
| VASCO TUF
(HT) |
C |
A+ |
C |
| 60 |
D2 (LT) |
C |
C |
C |
| MATRIX
II (HT) |
B |
A+ |
B |
| VANADIS
4 (HT) |
B |
A+ |
B |
| 62 |
A2 (LT) |
B |
A |
C |
| ASP23 (HT) |
B |
A |
B |
| CPM 10V
(HT) |
B |
B |
B |
| CPM 15V
(HT) |
B |
D |
B |
| CPM M4
(HT) |
B |
B |
B |
| CRU WEAR
(HT) |
B |
B |
B |
| VANADIS
10 (LT) |
B |
A |
B |
| VANAIDS
4 (HT) |
B |
A+ |
B |
| VANADIS
4 (LT) |
B |
A+ |
B |
| VASCO WEAR
(HT) |
B |
B |
B |
| 64 |
ASP 30
(HT) |
A |
B |
A |
| M2 (HT) |
B |
C |
A |
| M4 (HT) |
B |
D |
A |
| M50 (HT) |
A |
B |
A |
| M52 (HT) |
A |
B |
A |
| VANADIS
10 (LT) |
A |
A |
B |
| 66 |
CPM REX
20 (HT) |
A |
B |
A+ |
| CPM RES
45 (HT) |
A |
A |
A+ |
| CPM T 15
(HT) |
A |
B |
A+ |
| MICRO MELT
HS 30 (HT) |
A |
B |
A+ |
| MICRO MELT
T15 |
A |
B |
A+ |
| T15 (HT) |
A |
D |
A+ |
| 68 |
ASP 60
(HT) |
A+ |
C |
A+ |
| CPM
REX76 (HT) |
A+ |
B |
A+ |
| M42 (HT) |
A+ |
D |
A+ |
| MICRO MELT
HS76 (HT) |
A+ |
B |
A+ |
| > 68 |
Cemented
Carbide |
? |
E+ |
A++ |
It is quite
difficult to make a similar ranking of cemented carbides and other
tooling materials, due to a lack of information on the properties
of each cemented carbide after thin hard coating. The ranking of
these materials may not be as important as with steels. These are
worth using only in the limited areas, for example, when the tooling
is to be used under very high loading or the tooling requires extremely
tight dimensional tolerance.
However, it
should be determined which commercial brands of cemented carbide
are more susceptible to brittle "h
phase" formation in CVD and TD coating.
Conclusions
Thin hard coating
processes (CVD, PVD and TD) can produce coatings of pure carbides
and nitrides onto a substrate surface. Because of the excellent
resistance of these coatings to wear and galling, they effectively
eliminate surface damage problems to product produced by metal form
tooling.
With the elimination
of wear and galling problems by thin hard coatings, substrate selection
can be focused on solving tooling failures caused by insufficient
toughness and fatigue strength.
Selection of
proper substrate materials for fatigue and toughness combined with
proper selection of coating and post hardening conditions should
offer the optimum in tooling performance.
Since the surface
properties of coated materials are independent of the substrate
materials used, a large possibility exists for cost reductions through
the selection of less-expensive substrate materials that have less
wear and galling resistance. These less-expensive substrate materials
should have better machinability and heat treatability, along with
good toughness and fatigue properties.
|