ArticlePDF Available

Analytically Solvable Differential Diffusion Equations Describing the Intermediate Phase Growth

Authors:
  • Rauf Ablyazov East European University

Abstract

Analytical method to solve differential diffusion equations of describing the growth of the phase wedge during the intermetallic compound formation with a narrow concentration range of homogeneity in bicrystals is proposed. A model of describing the diffusion phase growth from point sources inside polycrystals grains is regarded. Analytical method to solve differential diffusion equations for such model is suggested. Parabolic, cubic, fourth power diffusion regimes for different scales from nanometers to micrometers and millimeters are analyzed. Key words: diffusion, reaction, phase growth law, intermetallic compounds, grain boundaries.
ДЕФЕКТЫ КРИСТАЛЛИЧЕС
КОЙ РЕШЁТКИ
PACS numbers: 61.72.Cc, 64.75.Op, 66.30.Dn, 66.30.Ny, 66.30.Pa, 68.35.Fx, 68.35.Rh
Analytically
Solvable Differential Diffusion Equations
Describing
the Intermediate Phase Growth
M.
V. Yarmolenko
Kyiv
National University of Technologies and Design, Cherkasy Branch,
Faculty
of Market, Information and Innovation Technologies,
241/2
V. Chornovola Str.,
UA
-18028 Cherkasy, Ukraine
Analytical
method to solve differential diffusion equations describing the
growth
of the phase wedge during the intermetallic-compound formation
with
a narrow concentration range of homogeneity in bicrystals is proposed.
A
model describing the diffusion phase growth from point source inside the
polycrystal
grains is regarded. Analytical method to solve differential diffu-
sion
equations for such a model is suggested. Parabolic, cubic, and fourth
power
diffusion regimes for different scales from nanometers to micrometers
and
millimeters are analysed.
Key
words: diffusion, reaction, phase-growth law, intermetallic compounds,
grain
boundaries.
Зап
ропоновано аналітичну методу розв’язування диференційного рів-
няння,
що описує кінетику утворення інтерметалевої фази вздовж межі
між
зернами з одночасним проникненням у самі зерна. Розглянуто мо-
дель,
який описує кінетику утворення інтерметалевої фази з точкового
джерела
всередині полікристалічних зерен. Запропоновано відповідну
аналітичн
у методу розв’язування диференційного рівняння такого моде-
л
ю. Проаналізовано дифузійні режими (параболічний, кубічний, четвер-
того
степеня) для різних масштабів від нанометрового до мікрометро-
вого
та міліметрового.
Ключові
слова: дифузія, реакції, закон зростання фази, інтерметалеві
сполуки, міжфазні межі.
Corresponding author: Mykhaylo Viktorovych Yarmolenko
E-mail: yarmolenko.mv@knutd.edu.ua
Citation: M. V. Yarmolenko, Analytically Solvable Differential Diffusion Equations
Describing the Intermediate Phase Growth, Metallofiz. Noveishie Tekhnol., 40, No. 9:
12011207 (2018), DOI: 10.15407/mfint.40.09.1201.
Ìåòàëëîôèç.
íîâåéøèå
òåõíîë.
/
Metallofiz.
Noveishie
Tekhnol.
2018
, т. 40, 9, сс. 12011207 / DOI: 10.15407/mfint.40.09.1201
Îттиски
доступнû непосредственно от издателя
Ôотокопирование
разрешено только
в
соответствии с лицензией
2018 ÈÌÔ (Èнститут металлофизики
им. Ã. Â. Êурдюмова ÍÀÍ Óкраинû)
Íапечатано в Óкраине.
1201
1202 M. V. YARMOLENKO
Предлагается
аналитический
метод
решения
дифференциального
урав-
нения,
которое описûвает кинетику образования интерметаллического
соединения
вдоль границû между зёрнами с одновременнûм проникнове-
нием
в сами зёрна. Рассматривается модель, которая описûвает кинетику
образования
интерметаллического соединения из точечного источника
внутри
поликристаллических зёрен. Предлагается соответствующий ана-
литический
метод решения дифференциального уравнения такой модели.
Àнализируются
диффузионнûе режимû (параболический, кубический,
четвёртой
степени) для разнûх масштабов от нанометрового до микро-
метр
ового и миллиметрового.
Ключевые
слова: диффузия, реакции, закон роста фазû, интерметалли-
ческие
соединения, межфазнûе границû.
(Received March 12, 2018)
1. INTRODUCTION
Analytical method of interdiffusion problems was presented in [1]. The
researchers analysed concentration profile of Zn in the diffusion re-
gion of ZnCu alloy (α-brass, solid-state solution, concentration of Zn
was less than 30%). This system has several intermediate phases too
(β-brass, concentration of Zn is about 50%, γ-brass, concentration of
Zn is about 68%, ε-brass, concentration of Zn is about 84%). These
phases are formed between α-brass and Zn during diffusion. Approxi-
mation of constant diffusion flux along the diffusion direction within
the width of each phase is used (so-called constant flux method) for de-
scribing the growth kinetics of the phases which was theoretically
grounded in [2]. This technique necessitates no allowance for the con-
centration dependence of D(C). Deviations from the parabolic law of
phase growth in cylindrical and spherical samples were analysed in [3]
using this method. This method was applied for describing the growth
kinetics of thin γ-brass and ε-brass layers in a cylindrical sample at
400°C (Cu was in the centre of the cylindrical specimens). The γ-brass
layer grew slower and the ε-brass layer grew more rapidly than in the
planar sample [4]. Model of the growth of an intermediate phase be-
tween low-soluble components on diffusion at grain boundaries involv-
ing outflow was suggested in [5] and criteria for a transition from the
Fisher regime t1/4
to a parabolic one were established. It was proved in
[6] that perpendicular grain boundaries do not influence phase growth
kinetics in B-regime. This result allows us to use the well-known model
of a polycrystal as a 3D array of grain boundaries to be perpendicular
to the interface for describing the phase growth. There were no expla-
nations in [5, 6] how one can solve differential diffusion equations be-
cause of a very complicated method. The formalism suggested was ex-
tended to the case of the growth of a solid-state solution with an expo-
ANALYTICALLY SOLVABLE DIFFERENTIAL DIFFUSION EQUATIONS 1203
nential concentration dependence of the diffusion coefficient. Analyt-
ical solution and Monte Carlo modelling of the Kirkendall effect were
suggested in [7]. Grain boundary (GB) diffusion parameters determi-
nation using A-kinetics of intermetallic layer formation was proposed
in [8]. Experimental data on Cu5Zn8 (γ-brass) diffusion growth kinetics
were used for separate determination of the volume diffusion activa-
tion enthalpy and the GB activation enthalpy. Alternative models of
competition of voiding and Kirkendall shift during compound growth
in reactive diffusion were analysed in [9]. One can improve the meth-
ods to solve the diffusion equations for the growth of intermediate
phase in bicrystals, polycrystals and inside grains.
2. MODELS AND METHODS
Model 1. The model of the phase layer growth during the intermetallic
compound formation with a narrow concentration range of homogenei-
ty, DC1, in bicrystals is based on the following assumptions [5, 6]:
1. An intermediate phase forms at first on the base of the grain
boundary; the latter, transforming from the boundary A–A to the
boundary 1–1, remains, due to easy influx with a diffusion coefficient
Db and having a thickness of δ 1 nm (i.e., the GB is not overgrown
with a new phase and does not bifurcate).
2. Formed phase 1 broadens normally to the GB due to volume diffu-
sion with a diffusion coefficient D << Db.
3. At all the points of the formed 1–A phase boundary between the
broadening phase 1 and the matrix A the concentration of the compo-
nent B is C1 on the side of phase 1 and is zero on the side of phase A
(solubility of B in A is ignored).
4. Outflow from the GB is the same at all GB points:
11
1
2
() , ( ,0) .
(, ) (,0)
C DC
C Cx xt t
x xt y xt C
DD
∂D
= = =
(1)
5. A flow in the volume of a phase wedge normal to the GB is con-
stant along x (a corresponding property is proved in [2]) in a reference
system associated with the moving nose of the wedge, y(t).
The equation for y(t) has such a form [5, 6]:
() ()
,
()
dyt A yt
B
dt y t t
=
(2)
where
11 11
/ , (1 / ) 2 / .
b
ADCCB DCC
=D =δD
There were no explanations in [5, 6] how equation (2) can be solved
as a very complicated method was used. A simpler method can be point-
ed out.
1204 M. V. YARMOLENKO
Method 1. One can simplify equation (2) by the following way
() 2
2 ( ),
dz t B
A zt
dt t
=
(3)
where
2
0
() () () ().zt utvt y t= =
One can transform equation (3) into
() () 2
() () () 2 .
du t dv t B
vt ut vt A
dt dt t

+ +=


(4)
Assumption
leads to
( ) exp( 4 ).vt B t=
Next step gives:
4 44
0
2
() 2 .
4
Bt Bt Bt
A AA
u t A e dt te e C
BB
B
= = −+
(5)
General solution of Eq. (3) is as follows:
4
0
2
() .
4
Bt
A AA
zt t Ce
BB
B
= −+
(6)
Using initial conditions z(t = 0) = 0 one can obtain finally:
2
( ) (1 ex p( 4 ))
4
AA
zt t B t
BB
= −−
(7)
or
2
( ) (1 ex p ( 4 )) .
4
AA
yt t B t
BB
= −−
(8)
Equation (8) shows the Fisher diffusion regime:
22
1
4
1
() 2
b
DC
yt t
DC
δD
=
(9)
for
() .
22
bb
DD
yt
DD
δ< < δ
Model 2. A model of the phase layer growth during the intermetallic
ANALYTICALLY SOLVABLE DIFFERENTIAL DIFFUSION EQUATIONS 1205
compound formation with a narrow concentration range of homogenei-
ty inside grains is based on the following assumptions:
1. An intermediate phase 1 forms inside grains from a point source
of substance A that is surrounded by substance B. The point source has
a diameter of δ 1 nm. The dislocations steps can be the point sources
in nanometers scale.
2. Dislocation pipe is easy path for A-atoms to go from substance A
to the dislocations steps with a diffusion coefficient Dd Db and a di-
ameter of δ 1 nm.
3. Formed spherical phases 1 broadens in 3D space from the disloca-
tions steps due to diffusion with a diffusion coefficient D1
(D < D1 < Dd).
Method 2. One can use constant flux method [3] to get differential
equation for intermetallic compound growing inside polycrystals
grains from a point source and forming small spherical particles
(which form the polycrystals [10] and 3-dimensional integrated cir-
cuits [11]):
2
11
sph 1
4 () ()
4 () , ()
2 () 2
RtD C dR t
J R tC Rt
R t dt
D δ
= = p >>
−δ
(10)
or
211
1
() () 2
DC
dR t Rt
dt C
D
= δ
, (11)
and the solution
11
3
1
3
() .
2
DC
Rt t
C
=
(12)
3. ANALYSIS
It was proved in [6] that perpendicular grain boundaries do not influ-
ence phase growth kinetics in B-regime. This result allows us to use the
well-known model of a polycrystal as a 3D array of grain boundaries to
be perpendicular to the interface for describing the phase growth. The
growth phase layer law in polycrystals for diffusion time
22
1
11 3
1
42 8
b
DC
tDC
δ
>D
(13)
is parabolic because volume diffusion is more pronounced than GB dif-
fusion.
Parabolic diffusion regime is valid (in micrometers and millimetres
1206 M. V. YARMOLENKO
scales [4]) for y(t) > (Dbδ)/2D [5, 6]:
1
1
2
() .
DC
yt t
C
D
=
(14)
Parabolic diffusion regime is valid in nanometres scale and the
growth phase layer law is as follows:
1
1
2
() .
b
DC
yt t
C
D
=
(15)
A comparison of Eqs. (12) and (15) show that
23
1
11 5
1
23
.
2
b
CD
tCD
δ
D
(16)
One can find:
22
1
11 3
11
34 6
b
DC
tDC
δ
D
and
23
11
1
34
.
2
b
D
yt D


≈δ




(17)
4. SUMMARY
The growth law of the phase layer during the intermetallic compound
formation with a narrow concentration range of homogeneity is para-
bolic for diffusion time
23
1
5
1
.
2
b
CD
tCD
δ
<D
The growth phase layer law inside polycrystals grains is proportion-
al to
3
t
in about 100 nanometres scale for diffusion time
2 3 22
11
53
111
.
26
bb
C D DC
t
CD DC
δδ
<<
DD
The growth phase layer law in bicrystals in B-regime is the same as
the Fisher solution: the phase wedge is proportional to
4
t
for diffusion
time:
ANALYTICALLY SOLVABLE DIFFERENTIAL DIFFUSION EQUATIONS 1207
δδ
<<
DD
22 22
11
33
11 1
.
68
bb
DC DC
t
DC DC
The phase wedges and roughness are smoothed during phase growth
[4, 6]. Smoothing rate is the more pronounced, the smaller the rough-
ness radius [3]. The growth phase layer law in polycrystals in microme-
ters and millimetres scales for diffusion time
22
1
3
1
8
b
DC
tDC
δ
>D
is parabolic because volume diffusion is more pronounced than GB dif-
fusion.
REFERENCES
1. H. Cho, K.-M. Yamada, and T. Okino, J. Mod. Phys., 9, No. 2: 130 (2018).
2. K. P. Gurov, A. M. Gusak, and M. V. Yarmolenko, Metallofizika, 10, No. 3: 91
(1988) (in Russian).
3. A. M. Gusak and M. V. Yarmolenko, J. Appl. Phys., 73, No. 10: 4881 (1993).
4. V. V. Bogdanov, A. M. Gusak, L. N. Paritskaya, and M. V. Yarmolenko,
Metallofizika, 12, No. 3: 60 (1990) (in Russian).
5. M. V. Yarmolenko, A. M. Gusak, and K. P. Gurov, J. Eng. Phys. Thermophys.,
65, Iss. 3: 876 (1993).
6. M. V. Yarmolenko, Defect Diffusion Forum, 143147: 1567 (1997).
7. M. V. Yarmolenko, Defect Diffusion Forum, 143147: 509 (1997).
8. M. V. Yarmolenko, Solid State Phenom., 72: 251 (2000).
9. T. V. Zaporozhets, N. V. Storozhuk, and A. M. Gusak, Metallofiz. Noveishie
Tekhnol., 38, No. 10: 1279 (2016).
10. K. P. Gurov, A. M. Gusak, V. V. Kondratev, and M. V. Yarmolenko, Fiz. Met.
Metalloved., 66, No. 1:34 (1988) (in Russian).
11. K. N. Tu and A. M. Gusak, Scr. Mater., 146: 133 (2018).
... It agrees with in situ TEM observation of Si precipitate dissolution through the dislocation in the aluminium grain at 623K [17]. So, the diffusion law R=(αt) 1/3 from a point source (practically, through dislocation end point) obtained mathematically in [16] was proved experimentally in [17]. ...
... It agrees with in situ TEM observation of Si precipitate dissolution through the dislocation in the aluminium grain at 623K [17]. So, the diffusion law R=(αt) 1/3 from a point source (practically, through dislocation end point) obtained mathematically in [16] was proved experimentally in [17]. Dislocation diffusion energy for the diffusion of Si in an Al grain was calculated: ...
Article
Full-text available
Four main diffusion laws: 1D diffusion in a planar bulk sample or random walks along a straight line x=α1t1/2; 3D diffusion or random walks from a point source and forming small spherical particle: x=α2t1/3; 1D+1D diffusion or random walks along a straight plane with simultaneous outflow into balk: x=α3t1/4; 1D+2D diffusion or random walks along a straight line with simultaneous outflow into balk: x=α4t1/6 are analysed theoretically using mathematical modelling and appropriate physical models. Convex shape of the diffusion profile near the top along a dislocation pipe with simultaneous outflow into balk is predicted. It is shown that the cone angle near the top is increasing with time. Literature experimental data are used for analysis.
... тобто глибина проникнення пропорційна кореню шостого степеня з часу дифузії.Метод 5.Ми можемо також описати процеси перенесення речовини вздовж міжфазних меж у полікристалах. Замість рівняння (4.2) отримаємо таке рівняння, яке може бути розв'язане аналітично[9]: ПЛАТФОРМА 2. ІННОВАТИКА В НАУЦІ ...
Conference Paper
Full-text available
The article considers methods to solve some Physics problems that are absent in books for higher education students: dissolving of a metal anode, diving a pencil into water, determining the viscosity of a liquid by the time it flows out through a horizontal capillary from a vertical vessel, the process of transfer of matter along the dislocation tube with simultaneous penetration into the volume, the process of transfer of matter along the interfacial boundaries with simultaneous penetration into the volume.
... A method of dislocation pipe diffusion parameter determination during the type B diffusion kinetics was suggested by the model of dislocation pipe diffusion involving outflow [6,20]. The method involves diffusion dislocation pipe kinetics for two different annealing times at the same temperature during the type B kinetics and dislocation pipe kinetics for one annealing time at other lower temperature during the type C kinetics. ...
Chapter
Full-text available
Our investigations show that electrochemical corrosion of copper is faster than electrochemical corrosion of aluminium at temperatures below 100°C. Literature data analysis shows that the Al atoms diffuse faster than the Cu atoms at temperatures higher than 475°C, Al-rich intermetallic compounds (IMCs) are formed faster in the Cu-Al system, and the Kirkendall plane shifts towards the Al side. Electro-chemical corrosion occurs due to electric current and diffusion. An electronic device working time, for example, depends on the initial copper cover thickness on the aluminium wire, connected to the electronic device, temperature, and volume and dislocation pipe diffusion coefficients, so copper, iron, and aluminium electro-chemical corrosion rates are investigated experimentally at room temperature and at temperature 100°C. Intrinsic diffusivities ratios of copper and aluminium at different temperatures and diffusion activation energies in the Cu-Al system are calculated by the proposed methods here using literature experimental data. Dislo-cation pipe and volume diffusion activation energies of pure iron are calculated separately by earlier proposed methods using literature experimental data. Aluminium dissolved into NaCl solution as the Al 3+ ions at room temperature and at temperature 100°C, iron dissolved into NaCl solution as the Fe 2+ (not Fe 3+) ions at room temperature and at temperature 100°C, copper dissolved into NaCl solution as the Cu + ions at room temperature, and as the Cu + and the Cu 2+ ions at temperature 100°C. It is found experimentally that copper corrosion is higher than aluminium corrosion, and the ratio of electrochemical corrosion rates, k Cu /k Al > 1, decreases with temperature increasing, although iron electrochemical corrosion rate does not depend on temperature below 100°C. It is obvious because the melting point of iron is higher than the melting point of copper or aluminium. It is calculated that copper electrochemical corrosion rate is approximately equal to aluminium electrochemical corrosion at a temperature of about 300°C, so the copper can dissolve into NaCl solution mostly as the Cu 2+ ions at a temperature of about 300°C. The ratio of intrinsic diffusivities, D Cu /D Al < 1, increases with temperature increasing, and intrinsic diffusivity of aluminium could be approximately equal to intrinsic diffusivity of copper at a temperature of about 460°C.
... We should note that were XK is the Kirkendall shift. It needs to be pointed out that Gurov's and Gusak's method can be applied to describe phase formation rate along GB with outflow in volume [7,8,9] (Fisher's model [10]) and along dislocation pipe with outflow in volume [11] (Le Claire's and Rabinovich's model [12]). ...
Article
Full-text available
Electric corrosion of aluminium and copper is investigated experimentally. It is found that the electric corrosion of copper is higher than the electric corrosion of aluminium. It is also clarified that the intrinsic diffusion coefficient of Cu is higher than the intrinsic diffusion coefficient of Al in each phase, so inert markers move to Cu. Copper has a higher electric conductivity, higher thermal conduction, and lower material cost than gold, so it is possible to use Cu instead of Au for wire bonding in microelectronics packaging, because the thin Al pad (1.2 µm thickness) can prevent gold and copper corrosion. Intermetallics disappearance and Kirkendall shift rates calculation methods are proposed. Methods involve mass conservation law and concentration profiles change during mutual diffusion. Intermetallics disappearance and Kirkendall shift rates in Al-Cu (Al is thin layer on Cu), Cu-Al (Cu is thin layer on Al), Al-Au, Zn-Cu, and Cu-Sn systems are analyzed theoretically using literature experimental data. Diffusion activation energies and pre-exponential coefficients for Cu-Sn system were calculated combining literature experimental results.
... Reaction rates of phases formation at temperature 400 o C were measured: [10], and at temperature T = 250 o C (Sn is liquid) [11]. Parabolic growth constants for the layer thicknesses were measured in [6], also the range of homogeneity of each phase were measured, and the values of the mutual diffusion coefficients for the Cu3Sn (phase 1) and Cu6Sn5 (phase 2) phases between 463 K and 493 K (190 o C and 220 o C) were calculated too: [12] or by "constant flux method" (Gurov's and Gusak's method) [13][14][15][16][17][18][19][20][21][22][23] or by other methods [26,[28][29][30] ...
Article
Full-text available
Intermetallics disappearance rates and intrinsic diffusivities ratios in the Cu-Zn system at temperature 400oC and in the Cu-Sn system at temperatures from 190oC to 250oC are analyzed theoretically using literature experimental data. Diffusion activation energies and pre-exponential coefficients for Cu-Sn system are calculated combining literature experimental results.
... Authors [16] didn't calculate diffusion activation energies and the pre-exponential factors, so we can do it using calculated data by k. p. gurov's and a. m. gusak's method or "constant flux method" [3,4,[18][19][20][21][22][23] (Table 2) and eqs. 27: ...
Article
Full-text available
Copper and aluminium electric corrosion rates are investigated experimentally at room temperature and at temperature 100oC. It is founded that copper corrosion is higher than aluminium corrosion, and ratio of electric corrosion rates, kCu/kAl , decreases with temperature increasing. It is calculated that copper corrosion rate is approximately equal to aluminium corrosion at temperature about 300oC due to Cu2+ ions are less mobile than Cu+ ions. It is obvious physically: the higher temperature is, the grater atoms’ displacements in crystal lattice, Cu atoms can diffuse without two electrons, and Cu2+ ions more strongly interact with crystal lattice than Cu+ ions. A theoretical method to calculate intrinsic diffusivities ratio in double multiphase systems is proposed. The method involves the Kirkendall plane displacement and the general phases thickness only. Intrinsic diffusivities ratios in the Al-Cu system are calculated using literature experimental data. Diffusion activation energies and pre-exponential coefficients for the Cu-Al system are calculated combining literature experimental results. Analysis of literature data shows that the Kirkendall shift changes sign at temperature about 460oC in the Cu-Al system because of intrinsic diffusivities ratio, DCu*/DAl*, dependence from temperature. Such result agrees with copper and aluminium electric corrosion rates investigation.
Article
Full-text available
Abstract: Our investigations show that electrochemical corrosion of copper is faster than electrochemical corrosion of aluminium at temperatures below 100°C. Literature data analysis shows that the Al atoms diffuse faster than the Cu atoms at temperatures higher than 475°C, Al rich intermetallic compounds (IMCs) are formed faster in the Cu-Al system, and the Kirkendall plane shifts toward the Al side. Electrochemical corrosion occurs due to electric current and due to diffusion. An electronic devise working time, for example, depends on initial copper cover thickness on aluminium wire, connected to the electronic devise, temperature, and volume and dislocation pipe diffusion coefficients, so copper, iron, and aluminium electrochemical corrosion rates are investigated experimentally at room temperature and at temperature 100°C. Intrinsic diffusivities ratios of copper and aluminium at different temperatures and diffusion activation energies in the Cu-Al system are calculated by proposed here methods using literature experimental data. Dislocation pipe and volume diffusion activation energies of pure iron are calculated separately by earlier proposed method using literature experimental data. Aluminium dissolved into NaCl solution as the Al3+ ions at room temperature and at temperature 100°C, iron dissolved into NaCl solution as the Fe2+ (not Fe3+) ions at room temperature and at temperature 100°C, copper dissolved into NaCl solution as the Cu+ ions at room temperature and as the Cu+ and the Cu2+ ions at temperature 100°C. It is found experimentally that copper corrosion is higher than aluminium corrosion, and ratio of electrochemical corrosion rates, kCu/kAl>1, decreases with temperature increasing, although iron electrochemical corrosion rate does not depend on temperature below 100°C. It is obvious, because the melting point of iron is more higher than the melting point of copper or aluminium. It is calculated that the copper electrochemical corrosion rate is approximately equal to aluminium electrochemical corrosion at temperature about 300°C, so copper can dissolve into NaCl solution mostly as the Cu2+ ions at temperature about 300°C. The ratio of intrinsic diffusivities, DCu/DAl<1, increases with temperature increasing, and the intrinsic diffusivity of aluminium could be approximately equal to the intrinsic diffusivity of copper at temperature about 460oC. Intrinsic diffusivities ratios in the Cu-Zn system at temperature 400°C and in the Cu-Sn system at temperatures from 190°C to 250°C are analyzed theoretically using literature experimental data. Diffusion activation energies and pre-exponential coefficients for the Cu-Sn system are calculated combining literature experimental results.
Article
Full-text available
Growth kinetics of second phase, which appearing between first phase and second component after first component exhausting, is analysed. As shown, a diffusion homogenization of the first phase leads to near linear second phase growth laws in planar and cylindrical samples. As founded, the second phase velocity is higher in cylindrical samples then in planar samples because of internal stress relaxation arising due to first phase growing with negative dilatation. Method to estimate a time of exhausting one of the components is proposed too. Key words: diffusion, reaction, phase-growth law, intermetallics, grain boundaries.
Preprint
Full-text available
Our investigations show that electrochemical corrosion of copper is faster than electrochemical corrosion of aluminium at temperatures below 100 o C. Literature data analysis shows that the Al atoms diffuse faster than the Cu atoms at temperatures higher than 475 o C, Al rich intermetallic compounds (IMCs) are formed faster in the Cu-Al system, and the Kirkendall plane shifts toward Al side. Electrochemical corrosion occurs due to electric current and due to diffusion. An electronic devise working time, for example, depends on initial copper cover thickness on aluminium wire, connected to the electronic devise, temperature, and volume and dislocation pipe diffusion coefficients, so copper, iron, and aluminium electrochemical corrosion rates are investigated experimentally at room temperature and at temperature 100 o C. Intrinsic diffusivities ratios of copper and aluminium at different temperatures and diffusion activation energies in the Cu-Al system are calculated by proposed here methods using literature experimental data. Dislocation pipe and volume diffusion activation energies of pure iron are calculated separately by earlier proposed method using literature experimental data. Aluminium dissolved into NaCl solution as the Al 3+ ions at room temperature and at temperature 100 o C, iron dissolved into NaCl solution as the Fe 2+ (not Fe 3+) ions at room temperature and at temperature 100 o C, copper dissolved into NaCl solution as the Cu + ions at room temperature and as the Cu + and the Cu 2+ ions at temperature 100 o C. It is founded experimentally that copper corrosion is higher than aluminium corrosion, and ratio of electrochemical corrosion rates, kCu/kAl>1, decreases with temperature increasing, although iron electrochemical corrosion rate doesn't depend on temperature below 100 o C. It is obvious, because melting point of iron is more higher then melting point of copper or aluminium. It is calculated that copper electrochemical corrosion rate is approximately equal to aluminium electrochemical corrosion at temperature about 300 o C, so copper can dissolve into NaCl solution mostly as the Cu 2+ ions at temperature about 300 o C. Ratio of intrinsic diffusivities, DCu/DAl <1, increases with temperature increasing, and intrinsic diffusivity of aluminium could be approximately equal to intrinsic diffusivity of copper at temperature about 460 o C.
Article
Full-text available
A method of dislocation pipe diffusion parameters determination during the type B diffusion kinetics is suggested. Proposed method involves diffusion dislocation pipe kinetics for two different annealing times at the same temperature during the type B kinetics and dislocation pipe kinetics for one an-nealing time at other (lower) temperature during the type C kinetics. Transition time for type B kinetics to type A kinetics (volume diffusion) and kinet-ics law t 1/6 for cone top rate are used in this method. Literature experimental data are used for separate determination of the volume diffusion activation energy and the dislocation pipe diffusion activation energy. Запропоновано метод визначення параметрів дифузійних процесів уз-довж дислокацій, використовуючи кінетику типу В. У методі використо-вується дифузійна кінетика типу В вздовж дислокацій для двох різних часів відпалу при одній і тій самій температурі та дифузійна кінетика ти-пу С вздовж дислокацій для одного часу відпалу за іншої (нижчої) темпе-ратури; застосовується перехідний час від В-режиму до А-режиму (об'єм-на дифузія) і кінетичний закон t 1/6 для швидкості руху вершини конуса. Літературні експериментальні дані використовуються для визначення енергії активації дифузії окремо в об'ємі та вздовж дислокацій.
Article
Full-text available
The simultaneous growth of both the phase-layer thickness and the void sizes during the intermetallic-compound formation with different mobilities of components and with a narrow concentration-range of homogeneity is described. This is done with account of competition for extra vacancies between dislocation steps and interfaces (K-sinks leading to Kirkendall shift) and voids (F-sinks providing Frenkel voiding). Three alternative models for three alternative places of preferential voids' formation are formulated and compared. Possibilities of control over Kirkendall shift versus Frenkel voiding competition are discussed.
Article
Full-text available
An analytically solvable model for diffusion phase growth along the grain boundary in a bicrystal is suggested. Criteria for the transition between diffusion regimes and corresponding phase wedge lengths are found. It is shown that the phase wedge must have a convex shape near the top although the concave one is determined far from the top. It is proved that a perpendicular grain boundary does not influence phase growth kinetics in B-regime. This result allows us to use the well-known model of a polycrystal as a 3-D array of grain boundaries to be perpendicular to the interface for describing the phase growth.
Article
Full-text available
A method of grain boundary (GB) diffusion parameters determination during A-regime of intermetallic layer formation is suggested. Experimental data on Cu5Zn8 diffusion growth kinetics are used for separate determination of the volume diffusion activation enthalpy and the GB activation enthalpy.
Article
Full-text available
An analytical solution and computer simulation of mutual diffusion in a binary system are introduced for the case of different diffusion coefficients.
Article
Full-text available
It is shown that during diffusional growth of the intermediate phase between substances A and B for cylindrical and spherical samples, the total diffusion flux relative to the center, -S(r)D~∂c/∂r, where S is the area of the interface, varies by a small amount in the radial direction, despite the concentration dependence of the interdiffusion coefficient D~(c). In this context we propose a constant‐flux method which allows the kinetics of the diffusional phase growth in cylindrical and spherical samples to be described in a rather simple way. Deviations from the parabolic law of phase growth in cylindrical and spherical samples are analyzed.
Article
Classical cellular precipitation is an incomplete reaction, meaning not all the supersaturated solutes can be precipitated out. Consequently, the kinetic parameters such as lamellar spacing and growth velocity cannot be defined. Recently, in 3-Dimension Integrated Circuits (3D IC) microbump technology, the porous Cu3Sn formation is a complete cellular precipitation; the kinetic parameters can be defined. This is rare and unique in phase transformations. In this letter, kinetic models of both precipitations are analyzed for a deeper understanding.
Article
An analytically solvable model of the growth of an intermediate phase between low-soluble components on diffusion at grain boundaries involving outflow is suggested. Criteria for a transition from the Fisher regime t1/4 to a parabolic one are established. The formalism suggested is extended to the case of the growth of a solid-state solution with an exponential concentration dependence of the diffusion coefficient.
  • K P Gurov
  • A M Gusak
  • V V Kondrat'ev
  • M V Yarmolenko
K. P. Gurov, A. M. Gusak, V. V. Kondrat'ev, and M. V. Yarmolenko, Fiz. Met. Metalloved., 66, No. 1:34 (1988) (in Russian).