WELDING RESEARCH

SUPPLEMENT TO THE WELDING JOURNAL, OCTOBER 1992

Sponsored by the American Welding Society and the Welding Research Council

Diffusion Brazing in the Nickel-Boron System

Process variables are taken into account in developing

a model for the four phases of diffusion brazing

BY J. E. RAMIREZ A N D S. LIU

ABSTRACT. The principles of diffusion Diffusion Brazing 1 atm (1 4.7 lb/in. 2 ), and heated to the
brazing were investigated using the brazing temperature in an inert atmo-
nickel-boron system. Mathematical Diffusion brazing, also known as eu- sphere or vacuum (Ref. 2). Once the
modeling of the process was carried out, tectic bonding, activated diffusion bond- brazing temperature is reached, the in-
followed by experimental verification of ing, and transient liquid phase bonding, terlayer may melt, or a liquid may form
the model developed. The kinetics of was first reported by Peaslee and Boam as the result of element diffusion be-
melting and solidification in diffusion (Ref. 1) in 1952. The process combines tween the interlayer and the base metal.
brazing was thoroughly characterized as the manufacturing ease of brazing with The liquid, by capillary action, fills the
a function of process variables such as high joint strength, achievable only by joint clearance and eliminates potential
temperature, time, interlayer thickness, solid-state diffusion bonding. The basic voids at the interface between the two
filler metal composition, and the base features of diffusion brazing are shown parts. W h i l e the parts are held at the
metal composition. Boron diffusion in in Fig. 1. An interlayer of a specific com- brazing temperature, diffusion of alloy-
the base metal was the controlling mech- position and melting point is inserted be- ing elements occurs between the liquid
anism during isothermal solidification of tween the parts to be brazed. The inter- and the base metal leading eventually to
the liquid zone. The activation energy layer thickness is generally less than 250 isothermal solidification of the braze
for boron diffusion in nickel between Um (9.8 X 1 0- 3 in.). The parts are held metal. Maintaining the brazed compo-
1125° and 1 225°C was estimated to be together under slight pressure, less than nent at the brazing temperature after so-
39.7 kCal/mol. While liquid metal grain
lidification w i l l produce a joint with
boundary penetration increased the ki-
chemical composition and microstruc-
netics of the solidification process, boron
ture closely equivalent to those of the
concentration change in nickel near to
base metal.
the interface decreased the chemical KEY WORDS Diffusion brazing has been described
composition gradient and driving force
for diffusion of boron into nickel. The as a process comprised of four stages
Diffusion Brazing (Refs. 3, 4): 1) dissolution and melting
results of this work can be used to assist Nickel-Boron System
in the development of interlayer materi- of the interlayer; 2) homogenization and
Mathematical Model widening of the liquid layer; 3) isother-
als for diffusion brazing of other more
Boron Diffusion mal solidification of the liquid zone; and
complex nickel alloy systems.
Isothermal Solidi- 4) homogenization of the solidified bond
fication region. Figures 2 and 3 show schemati-
Pack Cementation cally the variation of composition in the
J. E. RAMIREZ and S. LIU are with the Cen-
ter for Welding and Joining Research, De- Process Variables joint region as the different stages of dif-
partment of Metallurgical and Materials En- Reaction Kinetics fusion brazing take place.
gineering, Colorado School of Mines, Chemical Composition
Golden, Colo. Activation Energy Purpose of This Investigation
Paper presented at the AWS 71st Annual
Meeting, held April 22-27, 1990, in Ana- This research was conducted to study
heim, Calif. the controlling mechanisms and the re-

W E L D I N G RESEARCH SUPPLEMENT I 365-s

 Assembly ready
for bonding

Heated t o bonding
t e m p e r a t u r e — Interlayer
melts
Brazing
At bonding t e m p e r a t u r e
diffusion causes joint
t o isothermally
solidify

Completion of
solidification — homogenization
continues

Bonded assembly wth

completed C C/3/.Co
CZa Lfi
homogenization
Percent of B

Fig. 1 — Functional description of the diffusion brazing process (Ref. Fig. 2 — Typical binary equilibrium phase diagram with an eutectic
2). reaction.

action kinetics of diffusion brazing. boronizing conditions were performed Following the brazing experiments,
Nickel, being the base element of many at 600°, 750°, 900° and 1000°C (1112°, the specimens were examined metallo-
superalloys, and boron, an element with 1 382°, 1 652° and 1 832°F) for one, three graphically using Marble's reagent to de-
high diffusivity, form the material sys- and six hours. Metallographic analyses termine the advance of the solid-liquid
tem used in this study. The process was were done to characterize the thickness, interface at a given time and tempera-
modeled mathematically and experi- uniformity and adherence of the surface ture. The thickness of the remaining liq-
ments were carried out to investigate the layer. X-ray analyses were also used to uid layer in each sample was also deter-
influence of temperature, interlayer identify the phases found in the boride mined, as an average of the measure-
thickness and composition, and the coating. ments taken every 80 to 1 60 u.m (3.1 to
phase diagram of the alloy system on Brazing of the boron-coated nickel 6.2 X 10~3 in.) along the entire interface.
diffusion brazing. specimens was carried out in a vacuum The concentration profile of nickel
furnace at the temperatures of 1125°, across the bond region was determined
Experimental Procedure 1175° and 1225°C (2057°, 2147° and using SEM-WDS (scanning electron mic-
Nickel coupons of commercial pu- 2237°F) and a vacuum of 1.33 X 10~ft roscope-wavelength dispersive spec-
rity (99.9 wt-%) were boronized in a kPa (1 X 1 0 - 5 torr). The specimens were troscopy). Attempts of direct determina-
graphite dish using the pack cementa- held together under the pressure of a tion of boron concentration with WDS
tion process. The surfacing heat treat- hand-tightened clamp for different peri- were unsuccessful due to the low boron
ment was carried out in an electric fur- ods of holding time (from minutes to concentration (2 to 3 wt-%) in the bond
nace with argon atmosphere. Each hours), followed by cooling in the vac- region. The strong interaction of the
nickel coupon was separated from the uum chamber. Aluminum oxide sheets boron characteristic x-ray radiation with
others by thin sheets of aluminum oxide were also inserted between the nickel the nickel matrix reduced drastically the
and only the top surface of each speci- specimens and the stainless steel clamp intensity of the boron x-ray radiation and
men was exposed to the boron powder. to prevent any contact reaction. The braz- exhibited extremely high noise-to-sig-
Preliminary tests to determine the ing thermal cycles are shown in Fig. 4. nal ratio.

<-C 0 03

A O

a. a.
C; :coL

(a) (b) (c) (d) (e)

Fig. 3 — Schematic representation of the variation of composition with time in diffusion brazing. A-C — Dissolution of the interlayer;
D — homogenization of the liquid region; E — isothermal solidification.

366-s I O C T O B E R 1992
 W E I G H T Vo
10 15 20 75 B

1400

1095C (Eutectic Tamparaturv)

Variable Holding Time

0 10 20 30 40 50 60 70 80 90 100 4-00
Ni. 10 zo 30 4-0 50 70 80 90 B
Time (min) ATOM Vo B
Fig. 4 — Schematic representation of the brazing thermal cycles. Fig. 5 — Nickel-boron equilibrium phase diagram (Ref. 6).

Results and Discussion tion and structure. Furthermore, Ni 3 B Dissolution and Melting of the Interlayer
and nickel form only a simple eutectic
Boride Interlayer system, instead of the multiple eutectic Considering an insert (or interlayer)
reactions that may occur in case that a of composition B positioned between
The structure, thickness and phase pure boron interlayer is paired with parts of composition A, under equilib-
composition of the boride coating pre- nickel in diffusion brazing. rium conditions, metal A w i l l dissolve
pared by diffusion saturation are strongly The mean thickness of the boride lay- C a L percent of B and metal B, (100 - CLn)
dependent of the composition of the sat- ers was measured as a function of time percent of A at temperature T 2 — Fig. 2.
uration medium, the composition and and temperature of the chemical heat The solid solutions, a and B, will be in
structure of the material being boronized treatment process. The growth of the dif- equilibrium with an intermediate liquid
(Ref. 5), and the saturation temperature fusion zone was found to obey the fol- phase of composition that may vary from
and time. As such, the phases encoun- lowing equation: C L a (in local equilibrium with a) to CLn
tered in the boride coating w i l l be de- (in equilibrium with |3). The initial com-
X = Ki" (1) position profile for this stage is indicated
termined by the kinetic conditions such
as rate of boron delivery to the surface by Fig. 3A. After a small amount of dif-
where X is the average thickness of the
and its diffusion rate into the metal, and fusion has occurred, a narrow layer of
boride layer, K is a proportionality con-
by the stability of the different boride liquid is formed on each side of the in-
stant and characteristic of the transport
phases at the saturation temperature. terlayer, producing a composition pro-
mechanism, and t is the pack cementa-
The nickel-boron equilibrium phase d i - file that is shown in Fig. 3B. The thick-
tion time. The exponent n was deter-
agram is shown in Fig. 5 to illustrate the nesses of the two liquid regions, as
mined to be 0.5, which suggests that the
many intermetallic compounds that may shown in Fig. 3B, will increase while the
growth of the boride layer was con-
exist between nickel and boron. thickness of the solid insert between
trolled by volume diffusion of boron in
them will decrease. At the moment that
X-ray analysis results showed that at nickel (Ref. 8).
the solid interlayer is completely con-
temperatures below 900°C, no stable
sumed, a single liquid zone of compo-
boride coating was formed on nickel. Modeling of Diffusion Brazing
sition ranging from C|_a to C L R, as shown
Above 1000°C, a thick layer of Ni 2 B and
in Fig. 3C, is formed.
Ni 3 B formed rapidly. At 900°C, only The development of the mathemati-
Ni 3 B was observed. Schoebel and cal modeling of diffusion bonding has In the case that boron is initially con-
Stadelmair (Ref. 7) indicated that Ni 2 B been presented by several authors (Refs. fined within a finite region, - h < X < +h,
was metastable at approximately 900°C 9-11) and w i l l not be detailed in this and considering unidimensional diffu-
and transformed readily into the more paper. However, the major results of the sion from a source of finite thickness,
stable Ni 3 B. Hence, all specimens were modeling work by Ramirez and Liu (Ref. Ramirez and Liu (Ref. 9) proposed that
boronized at 900°C to have an interlayer 9) and experimental verification of their the dissolution time (tj) of an interlayer
with well-defined chemical composi- model are reported below. of thickness, 2h, at a given temperature

W E L D I N G RESEARCH SUPPLEMENT I 367-s

 Ag-Cu Silver-Copper
Alomic Percentage Silver
20 30 40 50 60 70 80 90
lOOrv l 0 6 4 . 5 *

1000
961.93'
L
900
(Cu)

800
i«OOF 7.9 71.9 91.
\(Afl)
700 ' /

600 /

500 :

0 10 20 30 40 50 60 70 80 90 100 400
Cu 10 20 30 40 50 60 70 80 90 Ag
Interlayr Thickness, 2h (p,m) j.C.Chasion Weight Percentage Silver
Fig. 6 — Dissolution time as a function of interlayer thickness, as pre- Fig. 7 — Silver-copper equilibrium phase diagram (Ref. 12).
dicted by the mathematical model for the Ni-NijB system.

could be determined by the f o l l o w i n g uid, the value of D L = 1 0" 4 cm 2 /s (10 4

equation (Ref. 9): -« _= erf(K) (3)
ium2/s) was assumed. As a result,
(2hf C 0 and CLp can be defined in the N i -
(2) t„=l-x\0-i(2hf at 1125°C (4)
16K2DL Ni 3 B portion of the Ni-B equilibrium
phase diagram.
where D L is the liquid diffusivity of td = 5.1xlO"*(2A)J at 1150°C (5)
At 1125° (2057°F) and 1 150°C
boron and K is a constant that depends
(21 02°F), K was found to be 0.78 and In these two equations, t^ is expressed
on the alloy system. K can be obtained
1.10, respectively. Due to the lack of re- in seconds, and 2h in micrometers. As
by solving the following equation (Ref.
ported data for boron diffusivity in liq- can be seen in Fig. 6, the dissolution
9):

.
" K = 0.330 (10) CoL Solid
! K = 0.325 (11)
- K = 0.337 (Present Work) (a)
{u)
~w Liquid

E 3 - #(10)

X = 0

2 --
*-— Present Work

1 — i i . 1 . 1 . 1 , 1 . i . i . t .

0 10 20 30 4-0 50 60 70 80 90 100
Interlayer Thickness, 2h (/Am)
Fig. 8 — Dissolution time as a function of interlayer thickness, pre- Fig. 9 — Homogenization oi the liquid region. A — Initial conditions;
dicted by three mathematical models for the Ag-Cu alloy system at B — boundary conditions at t > 0.
82CPC.

368-s I OCTOBER 1992

 350

E 300
D
.ff -y
OJ
CJ
o 250
t
200
Experimental Data
:

150
D D
Z ^ " " .ff = S.rf
100 - •x/
50
7
n . t . , . 1

0 10 20 30 40 50 60 70 80 90 100 10 15 20
Interlayer Thickness, 2h (y>m) Time (min)
Fig. 10 — Homogenization time as a function of interlayer thickness, Fig. 11 — Displacement of the solid-liquid interface as a function of
as predicted by the mathematical model for the Ni-Ni3B system. time, during homogenization of the liquid region, in the Ag-Cu alloy
system.

time of an interlayer of thickness rang- extended source of solute was also as- nonstoichiometric Ni 3 B may also be pre-
ing from 0 to 100 (im (the range of prac- sumed. Notice that the major differences sent and affect boron diffusion toward
tical applicability in diffusion brazing) between these two approaches and the the base metal. The formation of a liq-
required only fractions of a second. approach taken in this work were the as- uid film also accelerates greatly the dif-
Thus, experimental observation of inter- sumed initial distributions of the diffus- fusion process (Ref. 14).
layer dissolution is very difficult and was ing species and the movement of the
not attempted in this work. solid interface. The different initial and Homogenization and Widening
Comparable results have been re- boundary conditions led to different ex- of the Liquid Region
ported in t w o other studies by Tuah- pressions for the evaluation of K in Equa-
Poku, ef al., and Liu, ef al. (Refs. 10, 11). tions 2 and 3. To determine the aptness When the liquid layer is in contact
of Equation 2 in predicting interlayer dis- with the adjacent base metal, further dif-
(2hf (6)
solution time, the equation was plotted fusion of alloying elements takes place.
\6K;D, together with Equations 6 and 7 using The elements present in the liquid phase
the Ag-Cu-Ag system for illustration. The diffuse into the solid base metal while
_ (2hf K value was determined for 820°C using elements from the base metal migrate
(7) the Ag-Cu phase equilibrium diagram in into the liquid phase, leading to the ho-
16K;D,
Fig. 7. The time of interlayer dissolution mogenization of the liquid region and
K1 in Equation 6, proposed by Tuah- and melting was calculated and plotted further melting of the base metal. Finally,
Poku, etal. (Ref. 10), can be evaluated in Fig. 8. Note that Equation 2 predicts a wider liquid region of uniform c o m -
from the solidus and liquidus composi- well the fast kinetics of the dissolution position CLrx results — Fig. 3D.
tions of the alloy system. These authors and melting of the interlayer during the The net transfer of elements from
assumed an initially extended source of diffusion brazing. solid to liquid, or vice versa, will depend
solute and a unidirectional displace- Taking into account that the "inter- on the relative amount and composition
ment of the solid-liquid interfaces into layer" used in this work (Ni 3 B) was stoi- of the solid and liquid. The rate of mass
the interlayer. Equation 7, on the other chiometric and that the nickel and boron transfer, however, is a function of the
hand, considered active dissolution of diffusivities are greatly reduced due to concentration gradients of the alloying
both the interlayer and base metal, with the chemical bonding between these elements in the system (Ref. 15). The
the liquid region advancing into both atoms. The accelerated kinetics ob- isothermal mass transfer from solid to
the base metal and the interlayer (Ref. served during this first stage of diffusion liquid in a concentration gradient can
11). In this equation, K2 is a dimension- brazing may be the result of high diffu- generally be considered to occur in two
less growth constant related to the dis- sion kinetics in the surface layers and sequential steps. First, a surface reaction
placement of the interfaces. An initially interfaces (Ref. 13). Small amounts of

Table 1 -- Composition Data and 7h Parameter in the Ni-Ni3B System Table 2— Dependence of 7s on Ci in the
Ni-Ni3B System
T(C) CoL(at.-%) CLa(at.-%) C„L(at.-%) Ci(at.-%) 4> C0(at.-%) •yh
1125 17.0 14.5 0.18 0 227.7 25.0 0.091 T(°C) ys Ci (ppm)
1175 18.8 12.7 0.16 0 27.7 25.0 0.212 1125 3.29 X 1 0 " 4 320
1225 17.7 10.4 0.12 0 2.7 25.0 0.279 1175 6.24 X 10~ 4 257
1225 7.05 X 10" 4 209

W E L D I N G RESEARCH SUPPLEMENT I 369-s

 Assuming that p 0 is equal to p L and re-
'La arranging Equations 11 and 1 2, homog-
enization time, t h , can be rewritten as:

Solid Liquid (a) I

(2hf (13)
4
7/,V D *
Ci
The data and numerical solution of y^
for the Ni-Ni 3 B system at 11 25°,11 75°,
and 1225°C are reported in Table 1.
Assuming that boron diffusivity in the
X == 0 liquid phase is the controlling mecha-
nism and that Det-,- = D L , Equation 1 3 can
be rewritten as:
C
L«
,,, =4.1xl0- 4 (2/i)" at 1125°C (14a)
CoL
th = 1.3xlO~*(2A)3 at 1175°C (14b)
Solid Liquid (b)
c /,, = 1.5xl0" 4 (2/;) : at 1225°C (14c)
aL
Fig. 12 — Isothermal As predicted by the model, the stage
Q of homogenization and widening of the
solidification. A —
Initial conditions; B 1 liquid zone required only a few seconds
(Fig. 10) and could not be observed ex-
— boundary condi-
tions at t > 0. perimentally. Figure 1 0 also shows that
increasing the bonding temperature
does not always reduce t h because y h is
a strong function of the materials system
w i l l occur in which atoms go from the and depends on the characteristics of
solid into the liquid phase. Second, the _C„L-CLa exp{-Y,; the corresponding phase diagram.
atoms accumulated in the liquid zone cLa-c, To confirm the adequacy of the
next to the solid-liquid interface will mi- model developed, experimental data for
grate into the bulk of the liquid (Ref. 1 5). -C, exp[-V,,-<p) 1
(9) silver-copper diffusion brazing (Ref. 10)
The w i d e n i n g of the liquid region or C -C 1- ''/(r„v>) \K,t> were used. From Fig. 7, C o L , C L a , and
solid-liquid interface displacement will C a L were determined to be 39.5, 29.8,
Deff is the effective diffusivity and is de-
be controlled by the slowest step (reac- and 12.3 wt-%, respectively, at 820°C
pendent of the controlling mechanism,
tion at the interface, surface diffusivity, (1 508°F). The liquid, surface, and lattice
that is, diffusion of atoms in the liquid
or diffusion in the liquid). At low liquid diffusivity coefficients (D L , DSurf and Ds)
phase or to the reaction at the interface,
diffusion rates, the surface reaction may of copper in silver (Refs. 16-18) are 1.2
during the process. In Equation 9, the
be fast enough to maintain the concen- X 1 f>5 / 1 x 1 0" 6 and 4.6 X 1 0" 1 0 cm 2 /s,
concentration terms can be determined
tration of solute atoms in the liquid at respectively. Using these data,Yn was
by using the equilibrium phase diagram,
the interface close to the equilibrium liq- calculated to be 0.235, and therefore,
(j) is the liquid to solid diffusivity ratio,
uid composition. However, if neither
DL/D,
step is much faster than the other, both
mechanisms may control the process. Aside from the initial thickness of the Z = 0.235T/4ZV* O5)
interlayer, the maximum width of the where Deft- is the effective diffusivity. As
The mathematical treatment of this homogenized liquid region, W m , was
stage was based on the analysis of dif- mentioned previously, the homogeniza-
found to be a function of the material
fusion in systems with moving bound- tion process occurs through two sequen-
system.
aries. The assumptions made were: 1) tial steps. Therefore, each of the follow-
unidimensional diffusion; 2) static liq- ing expressions for the displacement of
uid, with no convection effect; 3) con- 2/ip„C, the solid-liquid interface may be used
W„, = (10)
stant diffusion coefficient;4) equilibrium depending on w h i c h diffusion mecha-
Pfiu,
at the solid-liquid interface; and 5) semi- nism is considered to be the controlling
infinite media. The initial and boundary p 0 and p L are the densities of the solid one.
conditions are illustrated schematically interlayer and homogenized liquid
in Fig. 9. From the mathematical model zone, respectively. From Equation 8, the Z = 126.1x'th i f D # D, (16)
developed (Ref. 9), the displacement of homogenization time can be determined (17)
Z = 36.4 A t„ i f D „ = DS,
the solid-liquid interface, Z, during the as:
homogenization and widening of the Plotting Equations 16 and 1 7 with the
liquid region stage can be described by ZL experimental data, shown in Fig. 1 1 ,
Equation (8), (11)
4 clearly indicated that interface displace-
rA, ment in the stage of liquid zone homog-
(8) /here: enization and widening is mainly con-
z = r n 4D„, trolled by surface diffusion of copper in
Yh is a dimensionless parameter, char- silver. The good correlation shown in
W. - lh Fig. 11 also shows that by considering
acteristics of the system at a given tem- Z=
perature, that can be determined by nu- the relative kinetics of the t w o sequen-
merically solving the f o l l o w i n g equa- (2/0 tial steps that may occur, the mathemat-
tion: -1 (12) ical model developed is well able to pre-
2 PL
dict the displacement of the solid-liquid

370-s I O C T O B E R 1992
interface during the homogenization of (18)
the liquid region.

Isothermal Solidification of the tiquid Zone -C, e.\p{-Ys') 1

(19)
C„ •C^' \ + erf(y,)' 4^t
When the concentration of the liquid
zone reaches C L a the boundary condi- y s is a dimensionless parameter re-
tions are changed and the solute ele- lated to the solidification characteristics
ments in the liquid region w i l l now dif- of the diffusion brazing process and can
fuse entirely into the base metal. This be calculated by numerically solving
w i l l lead to a compositional change in Equation 19. D s is the boron diffusivity
the joint and will raise the melting point in solid nickel.
of the liquid layer. As a result, isother- The progress of solidification was
mal solidification occurs and reverses monitored by determining the amount
the direction of motion of the interface. of remaining liquid, Wf, in the braze
A n d as more diffusion takes place, the metal. Figures 1 3 and 1 4 illustrate the
liquid layer w i l l eventually disappear, change in thickness of the remaining liq-
as shown in Fig. 3E. Contrary to the pre- uid film as a function of holding time at
vious stages, isothermal solidification of several brazing temperatures. While at
the liquid region occurs rather slowly 1225°C, the nickel-boron-nickel bond
and is, thus, the controlling stage of dif- solidified after only 30 m i n , over 1 00 :
'I i il
fusion brazing. min were required for the 11 25°C spec-
To model isothermal solidification in imens to reach complete solidification. Fig. 13 — Thickness of the remain-
diffusion brazing as a diffusion problem The displacement of the liquid-solid in- ing liquid layer as a function of time
in systems with moving boundaries, the terface (indirectly, the width of the re- at 1175"C and at different holding
following conditions were assumed (Ref. maining liquid), shown in Fig. 14, fol- times. Top — / min; center — 4 min;
9): 1) unidimensional diffusion; 2) static lowed a square root law as a function of lower— 16 min.
liquid; 3) constant diffusion coefficient; time. However, solidification was ob-
4) semi-infinite medium; 5) Equilibrium served to occur at two different rates. At
the solid (C aL ) and liquid (C La ) at the in-
at the interface; and 6) constant solid- all temperatures, the liquid-solid inter-
terface at a given temperature (T) can be
liquid interface area. The initial and face advanced rapidly during the early
expressed as follows:
boundary conditions are illustrated in part of the process and became signifi-
Fig. 12. cantly slower at longer times. Attempts CaL=Ms(TitA-T) (20)
According to the above assumptions, to understand the basic mechanisms in-
Ramirez and Liu (Ref. 9) concluded that volved in isothermal solidification and CLa=M,\TMA-T) (21)
the displacement of the solid liquid in- to explain the t w o rate regimes in the where, M s and M L are the slopes of the
terface, Z, and the time for solidifica- controlling stage of diffusion brazing are solidus and liquidus lines, respectively.
tion, t s , could be determined by the fol- reported below. T M A is the melting temperature of metal
lowing equations (Ref. 9): The equilibrium concentrations of A. Using Equations 20 and 2 1 , the fol-

E 16
" V E
3 • 1225C
3
14 V 1175C
0J
• 1125C
c
CJ 12 t
CD

_C

cu 10 CD

6 E
\ ^\ CD
CJ

_o
'o 4 - \ \\ ta
E co
1\

s
(JJ
or 2
I
.

n T 7 * C I7""W W\ , •
10
0.5.
Square Root of Time (min ' ) Square Root of Time (min )
Fig. 14 — Width of the remaining liquid layer as a function of the Fig. 15 — Comparison of the predicted displacement of the solid-liq-
square root of time. uid interface with those obtained from experimental observation in
the Ni-Ni3B system.

W E L D I N G RESEARCH SUPPLEMENT I 371-s

 '

Base Metal

J L
(a)

<*
c
Fig. 16 — Grain boundary penetration observed during diffusion braz- o
ing in the Ni-Ni3B system.
2
c
lowing equation can be obtained. O Base Metal
Fig. 17 — Nickel con-
centration profile
c
C„,-C M, o
C -C M, -M, across the bond region o
as determined by elec-
C, tron microprobe analy-
"53 Liquid
(22) o
{M,-MS){TM-T) sis. A — 1 min holding
time at 1125°C;B— I
The term M,/(M L - Ms) is material de- min holding time at (b)
pendent and usually considered c o n - 1175°C;C — less than
stant for a chosen alloy system. The sec- 1 min holding time at -\—r
100

y\
ond term in the right-hand-side mem- 1225°C.
ber of Equation 22 is temperature de- c
pendent. However, since high-purity o 98
nickel coupons were used in this work,
2
Cj (initial boron concentration in the 96
c
base metal) can be assumed to be zero, CD Base Metal Base Metal
thus eliminating the temperature depen-
dency. Substituting Equation 22 in 19,
O
c
o
94 fc
Ys can be rewritten as: o
"tD
92 •L-- Liquid
M, e.xpi -ry) 1 I L
ML-M, \ + erf(ys)
(23) 90 _J_J

v^ (c)
and for the N i - N i 3 B system, y5 was de-
termined to be 6.2 X 10~ 3 .
Since boron diffusion in nickel is ex-
pected to be the controlling mechanism
and the diffusion coefficient has been
reported (Ref. 19) to be 3.6 X 10~6 cm 2 /s, Liquid Phase Grain Boundary Penetration ance at the interface can be expressed
the displacement of the interface can be as:
modeled using Equation 1 8 as the fol- The deviation observed during the
lowing expression: initial solidification rate regime can be dCs'
A, C, - C „ , Z = A l - £ > (25)
explained as a result of grain boundary dx
Z = [.SZ^ts (24) penetration of the liquid film as illus- A 0 is the initial base metal interlayer
trated in Fig. 1 6. Despite that the mod- interfacial area and A r is the total solid-
Comparison of the predicted and ex- eling of isothermal solidification as- liquid contact area after liquid metal
perimentally observed solid-liquid in- sumed that the solid-liquid interfacial grain boundary penetration occurred.
terface displacement in Fig. 15 shows area remained constant (see assumption When corrected for the increase in liq-
that the displacement predicted by the 6 in the isothermal solidification section uid-solid interfacial area, ys becomes y sc
developed model is close to that ob- above), penetration of the liquid phase with the following expression:
served during the initial stage of the so- into the base metal along grain bound-
lidification process. The small deviation aries increased considerably the solid- A^ C„, em-Ysc l (26)
indicates good accuracy of the mathe- liquid contact area and the solidifica- \.. CLa~C^ l+erf(Yx) 4n
matical model, and that boron diffusiv- tion. As much as 2500 |im of penetra- The increase in interfacial area that
ity in solid nickel is indeed responsible tion depth was observed in this research. provided greater boron transfer also re-
for the kinetics of regime 1 in the early Considering the effect of grain sulted in a solidification rate faster than
part of isothermal solidification. boundary penetration, the solute bal- that predicted by the model.

372-s I OCTOBER 1992

 100 -

K
c
q
5*J
'in
O
Q.
E
o
o
"5

Fig. 18 — Secondary electron image

and average nickel concentration of a
partially solidified bond. Small nickel-
rich islands are observed.

Chemical Composition Changes T o f u r t h e r i n v e s t i g a t e this effect, t h e o f t h e best fit l i n e o f t h e e x p e r i m e n t a l

at the Liquid-Solid Interface c h e m i c a l c o m p o s i t i o n s o f the s p e c i m e n s results in t h e first r e g i m e o f the isother-
t h a t h a d e x p e r i e n c e d s o l i d i f i c a t i o n in m a l s o l i d i f i c a t i o n stage o n a l o g Z v s .
T h e v e r y s l o w d i s p l a c e m e n t o f the i n - t h e s e c o n d r e g i m e , p a r t i c u l a r l y in t h e l o g t p l o t . K n o w i n g K , ys a n d y s c , b o r o n
terface d u r i n g the second regime of region next to the s o l i d - l i q u i d interface d i f f u s i v i t y in n i c k e l c a n be d e t e r m i n e d
i s o t h e r m a l s o l i d i f i c a t i o n m a y also be the w e r e d e t e r m i n e d . Figure 1 7 shows the via the f o l l o w i n g equations.
r e s u l t o f a c h a n g e in b o r o n c o n c e n t r a - c o n c e n t r a t i o n distribution of nickel
t i o n in n i c k e l , at o r near t h e l i q u i d - s o l i d across t h e b o n d r e g i o n . As e x p e c t e d , a D. (29)
interface. This w o u l d decrease the g r a d i e n t of b o r o n c o m p o s i t i o n w a s o b - 417,
c h e m i c a l c o m p o s i t i o n g r a d i e n t a n d the
served o n the s o l i d side o f t h e s o l i d - l i q -
d r i v i n g force for diffusion of b o r o n into
u i d i n t e r f a c e , w h i c h supports t h e a b o v e
n i c k e l . A c c o r d i n g t o E q u a t i o n s 19 a n d D =- — (30)
discussion. Additionally, nickel-rich
2 2 , y s , at a g i v e n t e m p e r a t u r e , is d e p e n -
" p r e c i p i t a t e - l i k e " islands w e r e o b s e r v e d
d e n t o f t h e c o n c e n t r a t i o n o f b o r o n in D s c is d i f f e r e n t f r o m D s that it takes i n t o
to f o r m in t h e r e m a i n i n g l i q u i d z o n e as
n i c k e l , C j , a n d c a n be expressed as: c o n s i d e r a t i o n the l i q u i d phase grain
the specimens were cooled from the
b o n d i n g t e m p e r a t u r e at t h e e n d of t h e b o u n d a r y penetration effect. T a b l e 3
ys = [constant - / ( C , , TJ\- s u m m a r i z e s t h e f i n d i n g s of n, K, D S a n d
b o n d i n g process — Fig. 18. This i n d i -
cates t h a t e l e m e n t p a r t i t i o n w a s in D s t . T h e v a l u e s o f n, b e t w e e n 0 . 4 9 a n d
exp{-y]) 1
(27) progress w h e n s o l i d i f i c a t i o n w a s inter- 0 . 6 1 , s h o w e d that regime 1 of solidifi-
1 + erf(ys) 4n cation indeed f o l l o w e d the square root
rupted.
Table 2 shows the strong dependence l a w and that b o r o n diffusion was the
of y s o n C j in the range a p p l i c a b l e to the Boron Diffusivity Prediction controlling mechanism. D s varied from
N i - N i 3 B system. U s i n g the best linear fit 1.8 X 10~ 6 to 5.2 X 1 0 - 6 c m 2 / s . D s c w a s
r e l a t i o n b e t w e e n t h e interface d i s p l a c e - A c c o r d i n g to the mathematical f o u n d t o v a r y f r o m 1 . 0 X 1 0~ 6 to 4 . 5 X
ment and square root of t i m e d u r i n g m o d e l d e v e l o p e d , the d i s p l a c e m e n t of 1 0 ~ 6 c m 2 / s , in close a g r e e m e n t w i t h t h e
r e g i m e 2 (later o n d u r i n g t h e s o l i d i f i c a - t h e i n t e r f a c e , Z , c a n b e e x p r e s s e d as: d i f f u s i o n c o e f f i c i e n t s r e p o r t e d in the lit-
t i o n p r o c e s s ) , s h o w n i n F i g . 1 5 , y5 w a s (see Equation 1 8) erature (Ref. 19).
f o u n d to d o u b l e w h i l e C j w a s r e d u c e d In a d d i t i o n to d i f f u s i v i t y , t h e a c t i v a -
f r o m 3 2 0 t o 2 1 0 p p m b e t w e e n 11 2 5 ° Z = K V? (18a) t i o n e n e r g y o f b o r o n d i f f u s i o n in n i c k e l
and 1 2 2 5 ° C . Thus, small changes of
and,
b o r o n c o n c e n t r a t i o n in n i c k e l c a n p r o -
d u c e large changes in the k i n e t i c s o f the logZ = log/c + nlogf (28)
d i s p l a c e m e n t of the interface.
K c a n be d e t e r m i n e d as the i n t e r s e c t i o n Table 4 —Diffusivity of Carbon, Boron and
Beryllium in Nickel<a*

Solute D Equation T(°Q

Table 3— Exponenial Coefficient (n), log K Values (from the Logarithmic Relation between Z
and t), D s , and Dsc Values C /-35700\ 860-1100
0.37expl RT I
Regime T(°C) n log/c D s (cm 2 /s) Dsc (cm 2 /s) B /-39700\ 1125-1225
6
1125 0.61 -4.775 1.84 X 10" 1.01 X 10" 6
1 1175 0.51 -4.563 4.88 X 1 0 - 6 2.97 X 10~ 6 Be /-46200\ 1020-1400
1225 0.49 -4.572 5.20 X 10" 6 4.48 X 10" 6 0.02exp^ RT J
(a) Refs. 20, 2 1 .

WELDING RESEARCH SUPPLEMENT I 373-s

 140

120

100

-4.5

o
E
-5.0 - o

-5.5
0 10 20 30 40 50 60 70 80 90 100
—1 4 —1
Temperature ( x 10 K ) Interlayer Thickness, 2h (p.m)
Fig. 19 — Solidification of the liquid region in the Ni-Ni3B system: Fig. 20 — Homogenization time as a function of interlayer thickness.
Log K as a function of the inverse temperature (1/T). The final concentration of boron at the centerline of the bond was
assumed to be 100 ppm.

was also determined in the present work. 21 56°F) (Ref. 19). The temperature range and
K from Equation 18a can be rewritten of 11 25° to 1 225°C investigated in the
as: present w o r k was considerably larger = erf(K2) (39)
than those found in the literature.
The activation energy determined C r is the desired composition after ho-
K= n A ,4D = y^AD0 exp\j^;j (31)
seems to be reasonable when compared mogenization. Figure 20 gives the time
where D Q is the frequency factor in with the activation energy reported for required to reduce the boron concentra-
cm 2 /s and Q is the activation energy of carbon (Ref. 20) and beryllium (Ref. 21) tion at the centerline of a N i - N i 3 B - N i
boron diffusion in Cal/mol. R is the uni- in nickel, Table 4. The activation energy diffusion brazed joint to 1 00 ppm as a
versal gas constant and T is the process- for atom diffusion in a given matrix in- function of the initial thickness of the in-
ing temperature. The logarithmic form creases as the atomic radius of the dif- terlayer.
of Equation 31 is: fusing element increases. The atomic ra-
dius of carbon, boron, and beryllium in- Conclusions
1 crease in that order and thus, an increase
loeK" = loa/c (32)
is expected for the activation energy. Based on the modeling and experi-
where, mental results obtained, the f o l l o w i n g
Homogenization of the conclusions are made:
(33) Solidified Bond Region 1) Pack cementation is efficient in
providing a thin and uniform layer of
and Holding the solidified joint at a high Ni 3 B on nickel for diffusion brazing.
temperature will lead to extensive diffu- 2) The mathematical model proposed
(34) sion of the elements in the braze region describes well the kinetics of the four
2* 2.3ft into the base metal, which ensures a uni- stages of diffusion brazing, as a function
D 0 and Q were evaluated by plotting form distribution of the alloying ele- of processing temperature, interlayer
log K as a function of 1/T in the form of ments throughout the brazement. Thus, thickness and composition, and the
an Arrhenius plot (Fig. 19) and had val- the final composition and microstruc- phase diagram of the Ni-Ni 3 B system.
ues of 3.27 cm 2 /s and 39,700 Cal/mol, ture of the brazed joint w i l l closely re- 3) The first t w o stages, dissolution
respectively. Thus, semble that of the base metal. and melting of the Ni 3 B interlayer, and
According to the mathematical homogenization of the liquid zone are
D = 3.27 exp -39.700 (35) model developed, the time necessary to extremely fast, requiring seconds or less
RT reduce the concentration of the solute for completion.
This result is different from that reported at the centerline of the joint, from C 0 to 4) Isothermal solidification of a N i -
in the literature (Ref. 19). a required level C r , is given by the fol- N i 3 B - N i diffusion bond occurs slowly
lowing equation. with boron diffusion in solid nickel as
19,000
D = 2.3xl0">.vp - (36) the controlling mechanism.
RT I 1 V-h)
(37) 5) The kinetics of isothermal solidifi-
Disagreement between the two re- 16D, cation can be divided into two regimes:
sults is possibly because of experimen- a faster initial regime and a sluggish final
tal errors in a relatively narrow range of where, regime. Liquid phase penetration along
diffusion brazing temperature, for ex- the grain boundaries provides an in-
ample, 1140° to 1180°C (2084° and ^ = «/« (38) creased solid-liquid interfacial area for

374-s I O C T O B E R 1992
b o r o n transfer, a n d it is r e s p o n s i b l e f o r fusion bonding process I. Report of Center stresses in solids. Protective Coatings on Met-
t h e faster kinetics r e g i m e . for the joining of Materials, Carnegie-Mellon als. Vol. 5, pp. 247-252.
6) T h e v a r i a t i o n o f b o r o n c o n c e n t r a - University,Pittsburgh, Pa. 14. Semenov, A. P., Pozdnyakov,V. U.,
5. Ed. V. I. M a t k o v i c h . 1977. Boron and and Waposhina, L. B. 1 972. Contact eutec-
t i o n at the interface d u r i n g s o l i d i f i c a t i o n
Refractory Borides, Springer-Verlag. tic melting as a method of producing surface
c h a n g e s t h e k i n e t i c s o f the process a n d
6. Moffat, W . G. 1984. The Handbook of coatings. Protective Coatings on Metals. Vol.
is r e s p o n s i b l e f o r t h e s l o w e r s e c o n d 4, pp. 216-220.
Binary Phase Diagram, Vol. 1, General Elec-
regime. tric Co., Schennectady, N.Y. 15. Lammel,). M., and Chalmers, B. 1959.
7) T h e a c t i v a t i o n e n e r g y f o r b o r o n 7. Schobel, ] . D., and Stadelmair, H. Z. The isothermal transfer from solid to l i q u i d
d i f f u s i o n in n i c k e l , e s t i m a t e d to be 3 9 . 7 1965. Das zweistoffsystem nickel-boron, Z. in metal systems. Trans, of the Metallurgical
kCal/mol, compared well with diffusivi- Metallkunde, 56, pp. 856-859. Society of AIME, pp. 4 9 9 - 5 0 8 .
ties of carbon a n d b e r y l l i u m in n i c k e l . 8. Gasele, V., and Tu, K. N. 1982. Growth 16. Yamarura, T., and Ejima, T. 1973.
kinetics of planar binary diffusion couples. lapan Institute of Metals, V o l . 37, pp.
Acknowledgement Journal of Applied Physics, 53(4):3252-3260. 901-907.
9. Ramirez, J. E. 1989. Modeling of diffu- 17. Butrynowicz, D. B., M a n n i n g , J. R.,
sion bonding in nickel-boron system. M.S. and Read, M. E. 1974. Diffusion in copper
A u t h o r , J. E. R a m i r e z , g r a t e f u l l y a c -
thesis T - 3 7 5 1 , Colorado School of Mines, and copper alloys.,] Phys. Chem., Ref. Data,
k n o w l e d g e s t h e f i n a n c i a l s u p p o r t of t h e
Golden, Colo. Vol. 3, pp. 327-602.
Instituto C o l o m b i a n o d e l Petnoleo. 10. Tuah-Poku I., Dollars, M., and Mas- 18. Hall, M. C , and Haworth, C. W .
salski, T. B.. 1988. A study of the diffusion 1969. The diffusion of copper in silver(rich)-
References bonding process applied to a Ag/Cu/Ag sand- copper alloys. Trans. AIME, V o l . 245, pp.
wich joint. Metallurgical Transactions - A, V. 2476-2479.
1. Peaslee, R. L , and Boam, W . M. 1952. 19A, pp. 6 7 5 - 6 8 6 . 19. Liping, D., Fengzhi, Y., and Yugin, G.
Design properties of brazed joints for high 1 1 . Liu, S., Olson, D. L., Martins, G. P., 1 982. The diffusion bonding of superalloys
temperature applications. Welding Journal and Edwards, G. R. 1990. The use of coating K1 8 and the diffusion behavior of element
31(8):651-662. and interlayer technology in brazing. Weld- boron. Hanjie Xuebao. 3(3):84-98.
2. Paulonis, D. F., D u v a l l , D. S., and ing Journal 70(8):207-s to 215-s. 20. Smith, R. P. 1966. The diffusion of
O w c z a r s k i , W . A.. 1972. U.S. Patent 12. Eds. J. M. Poate and J. M. Mayer. 1978. carbon in gamma iron-nickel alloys.. Trans.
3,678,570. Ag-Cu phase diagram. Thin Films-lnterdiffu- AIME. Vol. 236, pp.1224-1227.
3. Sekerka, R. F. 1980. O n the modeling sion and Reactions, John Wiley, N.Y. 2 1 . Grigorev, G. V., and Pavlino, L. V.
of diffusion bonding. Proc. on Physical Met- 13. Podstrigach, Y. S., and Shevchuk, P. 1968. Diffusion of Be in Fe and Ni. Fiz. Metal.
allurgy, p. 1 , TMS-AIME, St. Louis, M o . R. 1973. Effect of thin coatings and interme- Metalloved, 25(5):836-839.
4. Lesoult, C. 1 976. Modeling of the dif- diate layers on diffusion processes and

Stress Indexes, Pressure Design and Stress Intensification

Factors for Laterals in Piping
By E. C. Rodabaugh

WRC Bulletin 360

January 1991

The study described in this report was initiated in 1987 by the PVRC Design Division Committee on Piping,
Pumps and Valves under a PVRC grant to E. C. Rodabaugh following an informal request from the ASME Boiler
and Pressure Vessel Committee, Working Group on Piping (WGPD) (SGD) (SC-II) to develop stress indexes and
stress intensification factors (/-factors) for piping system laterals that could be considered by the ASME Commit-
tee for incorporation into the code.
In this study, the author has considers all existing information on lateral connections in concert with existing
design guidance for 90-deg branch connections; and has developed compatible design guidance for lateral con-
nections for piping system design. As a corollary bonus, he has also extended the parameter range for the "B"
stress indexes for 90-deg branch connections from d/D = 0.5 (the present code limit) to d/D = 1.0. Therefore,
this report should be of significant interest to the B31 industrial piping code committees, as well as the ASME
Boiler and Pressure Vessel Committee.

Publication of this bulletin was sponsored by the Committee on Piping, Pumps and Valves of the Design Divi-
sion of the Pressure Vessel Research Council. The price of WRC Bulletin 360 is $30.00 per copy, plus $5.00 for
U.S. and $10.00 for overseas, postage and handling. Orders should be sent with payment to the Welding Re-
search Council, Room 1301, 345 E. 47th St., New York, NY 10017.

WELDING RESEARCH SUPPLEMENT I 375-s

 American Welding Society
Conference Planner

October 19-21, 1992 - Columbus, OH

Eighth North American Welding Research Conference: Recent
Developments in the Joining of Stainless Steels and High Alloys
This conference is directed to the producers and users of stainless steels and
high-performance Ni-based and Co-based alloys. Special topics will include
austenitic stainless steels, duplex stainless steels, and high-performance alloys.

November 3-6, 1992 - Orlando, FL

Int'l Conference on Computerization of Welding Information IV
Topics will highlight data formats and searchable standards, weld sensing for real
time control, quality and non-destructive examination, welding engineering
applications, weld controllers and control systems, and databases and welding
procedures. A hands-on computer exhibition will also be featured.

November 9-11, 1992 - Pittsburgh, PA

ISO 9000 - A Quality System Workshop
This intensive workshop, designed for fabricators who use welding procedures,
will compare the new European Community standards for welding procedures and
welder performance qualifications with existing U.S. welding design and
fabrication standards. Registration is limited to 20.

February 3-5, 1993 - San Francisco, CA

Golden Gate Materials Technology Conference
Topics will include advanced welding methods and equipment, structural welding
high-strength and conventional steels, advancements in metals and composites,
modern approaches to corrosion prevention, improvements in nondestructive
inspection technologies and strategies, new approaches to materials
characterization and process modeling, and improvements in productivity,
reliability and profitability. An exhibition will be featured.

For further information and registration details, call or write AWS Conferences, 550 N.W. LeJeune
Road, Miami, Florida 33126, 800-443-9353, Ext. 278 .

376-s I OCTOBER 1992