Mechanisms Producing Metallic Bonds in

Cold Welding 2
Q.

o
—i
UJ
>
Scanning electron microscope investigation shows two LLI
Q
\
coalescence mechanisms are involved in the cold welding X
o
oc
<
of aluminum to aluminum LU

BY N. BAY

Q_
O

ABSTRACT. In t h e c o l d w e l d i n g o f metals sure Y of t h e w e l d interface surface: f o r m s of surface preparations — m e c h a n i -

that are scratch-brushed b e f o r e joint _ AT - A0 cal, thermal and chemical t r e a t m e n t s —
plastic d e f o r m a t i o n , t w o basic b o n d i n g X h a v e b e e n investigated (Ref. 3, 4); these
or coalescence mechanisms exist. O n e is Ao s h o w degreasing f o l l o w e d b y scratch- X
o
f r a c t u r e of t h e brittle c o v e r layer f o r m e d AT - AQ brushing w i t h a rotating steel brush t o b e cc
b y scratch-brushing, extrusion o f base
Y = 0) t h e most e f f e c t i v e , giving t h e smallest <
At UJ
material t h r o u g h t h e cracks, and b u i l d u p t h r e s h o l d surface e x p o s u r e . Experimental </)
of real c o n t a c t a n d coalescing w i t h base w h e r e A 0 is t h e initial a n d A n t h e final area investigations h a v e also s h o w n the pres-
LU
tx
material of t h e o p p o s i t e surface. For AI-AI of t h e surface — Table 1 . In t h e case o f sure distribution at t h e w e l d interface
c o l d w e l d i n g this m e c h a n i s m is applicable rolling t w o metal sheets t o g e t h e r or c o m -
surface has an influence o n w e l d strength
t o 6 0 % of t h e w e l d area. In t h e r e m a i n - pressing t w o metal cylinders e n d t o e n d ,
(Ref. 1, 5-8).
ing area n o brittle c o v e r layer is present, Y = R, t h e r e d u c t i o n .
In this p a p e r t h e coalescence m e c h a - Q.
and w e l d i n g is h e r e established b y frac- Figure 1 s h o w s t h e w e l d strength in nisms are investigated, a n d a theoretical o
_J
t u r e o f t h e c o n t a m i n a n t film of oxides p u r e shear as a f u n c t i o n of t h e surface UJ
m o d e l f o r w e l d strength is d e v e l o p e d ,
and water vapor. e x p o s u r e f o r a n u m b e r of m e t a l c o m b i - >
explaining t h e influence o f t h e a b o v e -
nations w e l d e d b y rolling (Ref. 1 , 2).
A theoretical m o d e l f o r b o t h m e c h a - m e n t i o n e d variables.
Coalescence is n o t o b t a i n e d until a
nisms s h o w s t h e basic influence o f b o t h X
t h r e s h o l d surface e x p o s u r e has b e e n o
surface e x p o s u r e a n d n o r m a l pressure o n
r e a c h e d . B e y o n d this t h r e s h o l d value Metallic Bond Producing
oc
w e l d strength. It is in g o o d a g r e e m e n t
w h i c h is d e p e n d e n t o n t h e metal c o m b i -
<
Mechanisms. UJ
w i t h e x p e r i m e n t a l results.
n a t i o n , t h e w e l d strength increases r a p i d - v>
Experiments w i t h milled surfaces w e r e Brittle Cover Layer UJ
ly w i t h Y a n d reaches a steady progress
p e r f o r m e d t o investigate t h e influence o f oc
corresponding t o the strength of the
t h e c o n t a m i n a n t film w i t h n o brittle c o v e r Earlier investigations in c o n v e n t i o n a l
layer being present. The results s h o w a
w e a k e r m e t a l . This relationship b e t w e e n
light m i c r o s c o p e (Ref. 1) and in scanning
z
w e l d strength a n d surface e x p o s u r e is UJ
t h r e s h o l d surface e x p o s u r e i n d e p e n d e n t
o f n o r m a l pressure in a c c o r d a n c e w i t h
f u n d a m e n t a l f o r all c o l d w e l d i n g process- s
es. H o w e v e r , t h e t h r e s h o l d surface e x p o - Q.
t h e t h e o r y a n d an influence o f n o r m a l
sure for a g i v e n metal c o m b i n a t i o n o
pressure o n w e l d strength that is m u c h
appears t o b e d e p e n d e n t o n t h e t y p e o f
less p r o n o u n c e d . Table 1—Symbols and Definitions
process.
X
Surface p r e p a r a t i o n b e f o r e c o l d w e l d - X —Surface expansion of weld interface o
ing is a n o t h e r variable that has a basic surface oc
Introduction <
influence o n t h e w e l d strength. Various Y - Surface exposure of weld interface UJ
surface Crt
C o l d w e l d i n g is a solid phase w e l d i n g Y ' —Threshold surface exposure for ui
process characterized b y t h e large n u m - oc
contaminant film
ber o f possible metal c o m b i n a t i o n s . (1) C u - C u AT — Final area of weld interface
200 (2) Cu-Fe
Coalescence is o b t a i n e d b y joint plastic A 0 — Initial area of weld interface
(3) AI-AI
d e f o r m a t i o n of t h e t w o metals. A basic (A) Cu-Al A ' —Area corresponding to Y = Y '
x Q.
variable g o v e r n i n g coalescence is t h e P ',60 (5) Cu-Ni <r0 — Yield stress of basic material
d e g r e e of d e f o r m a t i o n expressed as t h e
e>
z (6) Cu-Ag aB — Weld strength o
_l
UJ (7) Zn-Zn CBC — W e l d strength, mechanism C
surface expansion X or t h e surface e x p o - tr 120 UJ
t- <rBF —Weld strength, mechanism F >
co i/Vc — Fraction of brittle cover layer to total UJ
o 80- C.
-I area
UJ \pf — Fraction of film layer to total area \
3 X
Paper presented under the title of "Bonding 40 (S-fp o
Mechanisms in Cold Pressure Welding" at the PE —Extrusion pressure oc
63rd AWS Annual Meeting in Kansas City, p — Normal pressure at base metal surfaces <
Missouri, during April 26-30, 1982.
oH p c ~ N o r m a l pressure between end surfaces Ui
0.3 0A 0.5 0.6 0.7 0.8
tf)
Y- SURFACE EXPOSURE of extruded base metals
N. BAY is Senior Lecturer- Dept. of Mechani- UJ
R — Reduction oc
cal Technology, AMT, Technical University of Fig. 7 — Weld strength in rolling as a function of RE — Extrusion reduction
Denmark, Lyngby, Denmark. surface exposure (Ref. 1 and 2)

W E L D I N G RESEARCH S U P P L E M E N T 1137-s

0.05 mm
' / ,
Fig. 2 — Weld interface surface after fracture.
Y = 0.375, p/<r0 = 1.9, <TB/(T0 = 0.02
electron microscope (SEM) (Ref. 7-11)
have indicated that a basic mechanism
governing coalescence is fracture of the
work-hardened surface layer produced
by scratch-brushing and exposure of the
base material.
Figures 2-4 show SEM micrographs of
the weld interface surface of Al — Al
(1100 alloy) cold welds after fracture by
tension testing. At a surface exposure
Fig. 4- Weld interface surface after fracture. Y = 0.72, p/o0 = 1.8, trB/tr0 = 0.71
138-s | M A Y 1983
Y = 0.375 (Fig. 2), fracture in the brittle
surface layer (characterized by horizontal
scratches from brushing) has occurred.
The underlying base material is exposed
and beginning to extrude up through the
cracks. The weld strength, however, is
small, only 0.02 times the yield stress o0
of the deformed material, and no weld
interface fractures are visible.
At a little larger surface exposure
Y = 0.383 (Fig. 3), base material has
extruded through the cracks. Real con-
tact and coalescence with base metal of
the opposite surface has been estab-
lished. Weld fracture is ductile and occurs
after local necking. Figure 4 correspond-
ing to Y = 0.72 shows extended areas of
base material and coalescence, unwelded
regions of brittle surface layer being con-
fined to small isolated islands. Here weld
strength is as = 0.71<ro.
Figure 5 shows a micrograph of a
section through the weld interface. The
brittle surface layer is broken under 45
deg in a shear fracture. Measurement of
the fragment length and the crack width
show that the fracture is brittle. The
fragments are lying like pearls on a
string.
Figure 6 shows a SEM micrograph of a
scratch-brushed Al surface before cold
welding. The topography is characterized
by an undulating country with long
tongue-shaped hills. It is these "tongues" Fig. 3 — Weld interface surface after fracture.
that are recognized as being the brittle Y = 0.383, p/cr0 = 2.0 aB/<r0 = 0.09
surface layer cracking when being
deformed — Figs. 2-4. They are easy to
distinguish by the regular pattern of hori- This seems supported by the micro-
zontal ridges from the scratch-brushing graph in Fig. 7 showing a section through
whereas the remaining areas in Fig. 6 the scratch-brushed surface. Brushing is
have a more irregular topography. The done from left to right. The tongue con-
width of each tongue is on the order of sists of a severely deformed structure
50-100 gm (2-4 X 10" 3 in.). The diame- which, at the right end, is teared over the
ter of each wire in the steel brush is original surface so that a crack is appear-
approximately 0.3 mm (0.012 in.), indicat- ing between the tongue and the base
ing that each tongue might be made by a material. This is generally seen, and some-
single wire. times the crack is more than half the total
tongue length.
Microhardness measurements were
made both on sectioned specimens after
welding and directly on top of the
scratch-brushed surface. They indicated a
hardness from 178 to 193 k p / m m 2 of the
tongues and 38 to 55 k p / m m 2 (Vickers,
load 15 g) of the base material (in initial or
deformed state).
Both methods resulted in a hardness
ratio between tongues and basic metal of
3:7. This explains the brittle behavior of
the tongues. They probably consisted of
deformed material mixed with oxides,
and it is believed that they might have
been formed by Al particles that were
picked up by brushing, adhered to the
wire ends of the brush, and later were
rejoined to the surface by cold welding.
This would explain the rather rough sur-
face topography found next to the
tongues —Fig. 6.
From the above observations it follows
that the role of scratch-brushing is not
just to clean the surfaces. It produces a

'.'V O.aRmyi
Fig. 5—Section through Al — Al cold weld
hard brittle surface film; this will crack
when being deformed and thereby
uncover underlying base material kept
clean after exposure due to high vacuum
in the cracks. The high normal pressure
between the contacting surfaces will
allow no admission of air, thereby keep-
ing the uncovered surfaces clean.
The surface film on top of the tongues
consists of an oxide layer on the order of
100 A thick and of a contaminant film
consisting of water vapor, grease and
gases that is about 30 A thick. This film is
very thin compared to the thickness of
the tongues (1-5 X 105 A). This means
that the contaminant film should not be
able to get in touch with the base mate-
rial in the crack. This also means that the
uncovered material, when getting into
contact with uncovered material from
the opposing metal, will bond immediate-
ly as observed in Fig. 3. In addition, weld
strength will increase with increasing nor-
mal pressure building up the contact area
by plastic deformation of the surface
asperities of the t w o rough surfaces. The
brittle surface layer of tongues is, there-
fore, here termed the cover layer.
Fig. 7-Section through scratch-brushed Al surface
Contaminant Film
As seen in Figs. 6 and 7 the brittle
tongues do not cover the whole surface.
Measurements on larger areas of scratch-
brushed surfaces showed the brittle cov-
er layer to be present as a fraction
\pc = 0.355 of the total area. In the
remaining area where I/'F = 1 — \pc =
0.645, coalescence is hindered by the
contaminant film. Placing t w o scratch-
brushed surfaces against each other, the
areas of cover layer " C " and of film layer
" F " of the t w o surfaces can be assumed
to be combined randomly as C — C,
C — F, F — C and F — F. In the areas
where one or t w o brittle cover layers are
present it can be assumed that the mech-
anism of metallic bonding will be as
described above, i.e., extrusion of base
material through cracks of the cover
layer. Vaidyanath, et al. (Ref. 1), have
shown that t w o opposing cover layers
will break up as one single layer.
In the remaining area, F — F coales-
cence is also obtained and can be seen in
Figs. 4 and 8. Figure 8 shows a surface
with Y = 0.45, a surface exposure large
enough to obtain clearly visible bonds.
The areas of coalescence are seen in the
cracks of the brittle cover layer. Howev-
er, they are extended uninterrupted in a
band from the crack into the F zone. This
phenomenon is also noticed in Fig. 4
WELDING RESEARCH SUPPLEMENT 1139-s
where Y = 0.72 and broad bands of
coalescence have been established.
Bands of coalescence extending into
the F zones from the C zones are gener-
ally found. This means that they are also
established in the F — F zones. The rea-
son for this is assumed to be that the
large local surface exposure due to crack-
ing of the cover layer will continue in an
extended band into the F zone. This is
because it will be less energy-consuming
than when establishing uniform deforma-
tion in the F zone; the latter would
require a large velocity discontinuity in
the deformation pattern at the boundary
of the tongue.
This leads to the conclusion that the
t w o mechanisms schematically outlined in
Fig. 9 are applicable (Ref. 12):
Mechanism C: Fracture of brittle cover
layer, extrusion of base material through
the cracks and establishment of real con-
tact and coalescence between base
materials.
Mechanism F: Fracture of contaminant
film and establishment of real contact and
coalescence between base materials.
Mechanism F is only functioning in the
contact zones F — F which are confined
to:
0 = iAF2 = 0.6452 = 0.4, or 40% of the
total area.
 a /Cover layer,-

B a a a a a
•^T^jfTxssa
g B ^ g e S g i g B g

contaminant film

onset of extrusion

local thinning of contaminant film

Fig. 9 —Schematic outline of coalescence mechanisms

Fig. 8-Weld interface surface after fracture. Y = 0.45, p/o0 = 1.8,
<rB7<r0 = 0.24

Theoretical M o d e l for W e l d
Strength
A theoretical model for the weld
strength based upon the first coalescence
mechanism — namely, fracture of brittle
cover layer —has been described in the
literature (Ref. 7, 8). A model taking both
mechanisms into account was later sug-
gested (Ref. 12). It was developed by
considering each mechanism separately
as if it were applicable to the whole
surface, calculating the weld strength in
each case, and then combining the t w o
expressions obtained for the weld
strength. The analysis is briefly shown
below. A more comprehensive descrip-
tion is given in the literature (Ref. 12).
Mechanism C
At a threshold surface exposure
dependent on the normal pressure p,
extrusion of base material through the
cracks of the cover layer is initiated. The
pressure PE necessary to start the extru-
sion is estimated using Johnson's (Ref. 13)
and Hill's (Ref. 14) slipline analysis of plane
strain extrusion through square dies.
Applying their results to the present
extrusion model and introducing the sur-
face exposure Y instead of the extrusion
reduction RE = 1 — Y yields the relation-
ship between pE and Y shown in Fig.
10.
When the normal pressure p reaches
the required pressure pE for extrusion,
the base material is forced through the
cracks in the cover layer. The extruded
material of the t w o metals meet, and real
contact is established. If the pressure p is
further increased, the normal pressure pc
and thereby the real coalescence area
between the rough end
increases; pc is given by:
Pc PE
In order to determine the nominal
bond strength <rBc (corresponding to the
final area AT), it is necessary to estimate
the true strength of the welds. The SEM
micrographs showed that the weld frac-
ture in tension testing was ductile. Con-
rad and Rice (Ref. 15) have investigated
the adhesion between clean surfaces of
Ag, Al, Cu and Ni, using the technique of
cold welding specimens previously frac-
tured in an ultrahigh vacuum. They found
that the weld strength obtained was
almost equal to the compression load
applied in cold welding. Similar results
were found by Upit and Manik (Ref.
16).
It is reasonable to assume (Ref. 7, 8, 12)
that the weld fracture is established by
plastic deformation in the welds and that
the stress necessary to break the welds
corresponds to the normal pressure nec-
0.2 CK 0.6
Y -SURFACE EXPOSURE
Fig. 10-Extrusion pressure as a function
surface exposure
surfaces essary to further increase of the coales-
cence or weld area (by plastic deforma-
tion of surface asperities). This normal
(2) pressure equals the maximum applied
compressive stress pc on the base metal
surfaces.
Force equilibrium gives the relationship
between the weld strength <rBc corre-
sponding to the nominal area AT and the
true weld strength p c corresponding to
the true virgin area:
O-BC A, = p c (AT - A 0 )
or by inserting (1) and (2):
«5c = YP_ PE
(3)
As shown earlier (Ref. 8, 12), t w o
maximum limits might be applied to the
expression above:
a /fraav 1 -
'max 1 = ,_ '
V3
telling that the maximum strength
obtained is equal to the yield strength of
the coalescence bridges formed in the
cracks between the cover layer. And:
C >max2-1
telling that the maximum strength equals
the yield stress of the deformed metal.
The correct solution is believed to lie in
between the t w o as discussed in the
literature (Ref. 8 and 12).
0.8 1.0
Mechanism F
of In the areas F — F where no cover layer
is present, coalescence is obtained when

140-s | MAY 1983

 a V °0
1.0 t /} = 0.415
THEORY V'= 0 35
t-
p / o • 1.85
O 08
UJ EXPERIMENTS

H 06 _ o Al • Al

UJ 04

0.1 0.2 0.3 0.4 0.5 0.6 0.7 06 09 1.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
y - SURFACE EXPOSURE Y- SURFACE EXPOSURE

Fig. 11 — Theoretical weld strength as a function of surface exposure Fig. 12 - Weld strength as a function of surface exposure. Al - Al,
and normal pressure Al - Al scratched-brushed, p/tr0 = 1.85

a threshold surface expansion Y ' neces-
sary to break the contaminant film has
been reached. This has been estimated to
0.30 < Y ' < 0.35 in case of Al - Al cold
welding as discussed later.
Assuming no cover layer but only a
contaminant film to be present the actual
area of base metal being decovered will
be A i — A ' where A ' is the area corre-
sponding to Y = Y ' . Similar to the calcu-
lation above for mechanism C, a force
equilibrium gives the relationship be-
tween the weld strength crBF correspond-
ing to the nominal area AT and the true
weld strength corresponding to the area
of base material surfaces AT — A ' . The
true weld strength equals p, the normal
pressure in the base material surfaces
according to the same arguments as
above based upon previously reported
results (Ref. 15, 16):
<rBF AT = p(Ai - A ' )
From equation (1) it is seen that:
1
— =
Ao " 1 - Y '
Together with (1) and (4) this gives:
O-BF Y - Y' p
1-Y'a0
Combined Theory for Mechanism C and F
As mentioned earlier, the contaminant
film fracture mechanism is acting in a
fraction 13 = 0.4 of the total area whereas
the cover layer fracture mechanism is
acting in the remaining area. This puts
forward the following suggestion for the
combined theory for the weld strength —
see equations (3) and (5):
^=(i-^)Yp^B+/3i^r.ii
a0 a0 1—Y
In case of a normal pressure p < PE, set
p = pE, and incase of Y < Y ' set Y = Y ' .
It follows that the first term in equation (6)
corresponds to mechanism C which
comes into action when p > pi. The
second coalescence mechanism is func-
tioning when the surface exposure Y
exceeds the threshold value Y ' .
Figure 11 shows the theoretical model
(4)
for Al — Al cold welding when Y ' = 0.35
and 8 = 0.4. At low normal pressures,
p/o-Q < 2.0, welding is initiated when
Y > Y ' by fracture of the contaminant
film. For p/a0 = 1.5, there is a bend on
the curve at Y = 0.49 corresponding to
the starting point of extrusion through
the cracks and weld formation between
the segments of cover layer. This is in
accordance with Fig. 10 showing the
(5)
necessary extrusion pressure.
For p/<r0 > 2.0 welding is initiated by
fracture of the cover layer, extrusion
through the cracks and weld formation in
between the segments of cover layer.
Mechanism F, the fracture of the contam-
inant film later comes into action, namely
at Y = Y ' ; this explains the bend at
Y = 0.35 of the curves for p/a0 > 2.0.
Experimental Results and
Discussion
(6)
aa In order to test the theoretical model
for weld strength, experiments were per-
formed in equipment that allowed inde-
pendent variation of surface exposure
and normal pressure when cold welding
Al — Al. The experiments as described in
the literature (Ref. 7, 8) were done by
plane strain compression in a channel
forming tool. Weld strength was esti-
mated by a specially developed tension
testing technique that allowed six small
test pieces to be cut from each specimen.
The specimen material was annealed
1100 alloy.
Two series of experiments were per-
formed. One with scratch-brushed sur-
faces allowed both coalescence mecha-
nisms to function. For the other series of
experiments, surfaces were milled with-
out a lubricant; this produced a surface
with a contaminant film of oxides and
water vapor (formed during the time

I
o o«

ft 0.415 1-
THEORY t 0.30 , 0.35
n 0 6
U 04 P/o0=5.1 -i 5 milled p/o 0 =l.33
UJ
5 0.4 + milled p/o 0 a3.85
EXPERIME >ITS
o s c r a t c h - b r u s h e d p / o 0 = 1.45
o Al - Al
o® 0.2 9l O 0 =3,85 _
t h e o r y /3-l,V'=0.3
- aa} 8 oo I o 00'—9-&
0.2 03 04 0.5 0.6 0.7
Y-SURFACE EXPOSURE
0 0.1 0.2 0.3 0.4 05 06 0.7 08 0.9 1.0
Y- SURFACE EXPOSURE

Fig. 14 - Weld strength as a function of surface exposure and normal

pressure, Al — Al. Comparison between theory and experiments with
Fig. 13 - Weld strength as a function of surface exposure. Al - Al, milled surfaces, p/a0 = 1.33 and 3.85. Experiments with scratch-
scratch-brushed, p/tr0 =5.1 brushed surfaces with p/<r0 = 1.45

WELDING RESEARCH SUPPLEMENT 1141-s

between milling and welding) but with no
brittle cover layer, i.e., only allowing the
second coalescence mechanism to func-
tion. Welding in both series was per-
formed after 5 minutes (min) exposure to
the atmosphere.
Figures 12 and 13 show a comparison
of the theoretical calculated weld
strength with experiments with scratch-
brushed surfaces. Surface preparation
consisted of degreasing with toluene fol-
lowed by scratch-brushing with a rotating
steel brush; the brush had a 180 mm (7
in.) outer diameter, 0.3 mm (12 X 10~ 3
in.) wire diameters, a 1450 rpm running
speed and 0.2 mm/s (8 X 10~3 in./s)
feed. Figure 12 corresponds to a mean
normal pressure p/aQ = 185 and Fig. 13
to p/a0 = 5.1. Although the scatter in the
experimental results is quite large, it
seems in reasonable accordance with the
theory; also, the sudden rise in weld
strength with surface exposure in Fig. 13
is found experimentally at about
Y = 0.35.
Figure 14 shows the results from the
series with milled surfaces. After degreas-
ing, a 0.1 mm (4 X 1 0 - 3 in.) chip was cut
from each of the bonding surfaces 5 min
before cold welding. Experiments were
conducted at t w o different mean normal
pressures — p/cr0 = 1.33 and 3.85.
Control experiments were conducted
with scratch-brushed specimens at p /
a0 = 1.45 for comparison with the earlier
scratch-brushing results (Ref. 7, 8).
Despite the lower normal pressure
obtained with the control experiments
(Fig. 12), the results lie in the same range.
The lower normal pressure is due to
more careful lubrication in these experi-
ments. The reason that the weld
strengths obtained are still in the same
range as for p/<r0 = 1.85 (in Fig. 12) has
yet to be explained. It might be due to
the fact that the relative humidity was
very low (down to 27%) when the last
experiments ( in Fig. 14) were carried
out.
From Fig. 14 it appears that milling
gives larger weld strengths than scratch-
brushing. Increasing normal pressure will
increase the weld strength also in case of
milled surfaces; however, the tendency is
much less pronounced than theory indi-
cates. The threshold surface exposure for
film break d o w n in case of milled surfaces
seems to be about Y ' = 0.3 independent
of normal pressure. This independence
142-s | M A Y 1983
was also adapted in the theory. The
threshold value seems somewhat smaller
than for the F — F zones of scratch-
brushed surfaces (Fig. 13) where
Y ' = 0.35 seems a better fit.
Conclusions
The SEM investigations of the weld
interface surface after fracture of Al — Al
cold welds show t w o coalescence mech-
anisms to be active:
1. Fracture of the brittle cover layer
formed by scratch-brushing, extrusion of
base material through the cracks of the
cover layer and establishment of real
contact between base materials. Calcula-
tions show that this mechanism is active
over 60% of the total area. In the remain-
ing 40% of the area, no cover layer is
present and welding is here obtained by
fracture of the contaminant film of oxides
and water vapor and establishment of
real contact and bonding between
exposed base material; in case of Al — Al
cold welding, this will occur when the
surface exposure Y exceeds Y ' = 0.35.
2. Based upon the observed coales-
cence mechanisms a theoretical model
for weld strength as a function of the
surface exposure and normal pressure
has been developed. The model also
takes into consideration the variation of
the threshold surface exposure before
break down of the contaminant film Y '
and the fraction 8 of the weld area with
no cover layer.
3. Experiments with conventional
scratch-brushed surfaces at varying sur-
face exposures and normal pressures
show good accordance with the theoret-
ical model. They confirm earlier experi-
mental observations that not only the
surface exposure but also the normal
pressure is a basic variable governing
coalescence; this explains the rather
unnoticed fact that the threshold surface
exposure to obtain coalescence can vary
from process to process. The variation of
the threshold with the metal combination
can also be explained since Y ' and 8 will
vary with the metal combination and
since the ductility of the cover layer may
depend on the base metal.
Experiments with milled surfaces show
larger weld strengths than with scratch-
brushed. This contradicts results in the
literature (Ref. 3). The threshold surface
exposure for breakdown of the contami-
nant film appears to be independent of
normal pressure in accordance with the
theoretical considerations. Normal pres-
sure has a less pronounced influence on
coalescence in case of milled surfaces
than in case of scratch-brushed, and the
effect is much smaller than theory
states.
Acknowledgment
The author wishes to thank Mr. O.
Jagert for helping with the experimental
investigations.
References
1. Vaidyanath, L. R., Nicholas, M . C , and
Milner, D. R. 1959. Pressure welding by rolling.
Brit. Weld. J. 6:13-28.
2. McEwan, K. J. B., and Milner, D. R. 1962.
Pressure welding of dissimilar metals. Brit.
Weld. I 9:406-420.
3. Vaidyanath, L. R., and Milner, D. R. 1960.
Significance of surface preparation in cold
pressure welding. Brit. Weld. J. 7:1-6.
4. Wodara, J. 1963. Einfluss der oberfla-
chenvorbereitung auf die Kaltpressschweiss-
barkeit von Metallen. Schweisstechn.
13(12):548-552.
5. Gumm, P. 1964. Kombination von
Umformung und Kaltpressschweissen beim
Fliesspressen und Rohrziehen. Dissertation.
Technische Hochschule Carolo-Wilhelmina,
Braunschweig, GFR.
6. Vorm, T., Bay, N., and Wanheim, T.
1973. Influence of the hydrostatic stress com-
ponent on critical surface expansion in forging
compound products. Weld. Res. Int. 3(3):16-
25.
7. Bay, N. 1976. Metallisk friktion og kold-
tryksvejsning. Ph.D.-thesis, AMT, Technical
University of Denmark.
8. Bay, N. 1979. Cold pressure welding. The
mechanisms governing bonding. Trans. ASME,
I Engn. Ind. 101(2):121-127.
9. Eggers, H., Krause, E., and Ruge, J. 1970.
Zum Mechanismus des Kaltpressschweissens-
Bruchflachenuntersuchungen mit dem Raster-
elektronenmikroskop. Schweiss. Schneid.
22:241-244.
10. Cave, J. A., and Williams, J. D. 1973.
The mechanism of cold pressure welding by
rolling. / Inst. Met. 101:203-207.
11. Wright, P. K., Snow, D. A., and Tay, C.
K. 1978. Interfacial conditions and bond
strength in cold pressure welding by rolling.
Met. Techn.: 24-31.
12. Bay, N. 1981. Cold pressure welding. A
theoretical model for the bond strength. Proc.
Instn. Metall. Conf.: joining of Metals-Practice
and performance. Univ. Warwick, U.K.: 1-16.
13. Johnson, W . 1954-55. Extrusion through
wedgeshaped dies, Part I, /. Mech. Phys. Sol.
3:218.
14. Hill, R. 1950. The mathematical theory
of plasticity: 185. Oxford: Clarendon Press.
15. Conrad, H., and Rice, L. 1970. The
cohesion of previously fractured FCC metals in
ultrahigh vacuum. Metall. Trans. 1:3019-3029.
16. Upit, C. P., and Manik, J. J. 1969. The
effect of load on the cohesive strength of
virgin surfaces of plastic metals. Wear 13:77-
84.