Computer simulation of temperature profiles of a twolayer sample during heating by the electron beam
Table of contents: The KazakhAmerican Free University Academic Journal №4  2012
Authors: Alontseva Darya, East Kazakhstan State Technical University in honor of D. Serikbayev, Kazakhstan
Krasavin Alexander, East Kazakhstan State Technical University in honor of D. Serikbayev, Kazakhstan
The efficiency of
advanced technology of getting protective coating by means of pulsing plasma
jet deposition of Nibased powders onto steel items often falls due to the porosity of the received
coatings and their poor adhesion to the substrate [1,
2]. To eliminate these disadvantages the coatings are modified by the
plasma jet or electron beam [1]. The processes of diffusion and formation of
new phases in materials under the influence of electron irradiation happen very
quickly, the temperature being one of the main factors influencing these
processes. However, the temperature measurement under irradiation conditions is
difficult and unreliable. Development of a mathematical model of temperature
distribution in a material depending on irradiation parameters makes it
possible to assume the kind of structures and phases that form in the material
during irradiation (on the basis of the received values of temperature and the
known phase diagrams). Based on this model, one can choose the parameters of
irradiation so as to develop sufficiently high temperatures on the boundary of
the coating to the substrate to accelerate the diffusion processes in order to
improve adhesion of the coating to the substrate. The sources devoted to the
development of such a model [3, 4] testify the relevance of this problem, but
they do not provide a comprehensive solution.
The aim of this work is to
propose a model of temperature distribution in twolayer metal absorbents
during irradiation by a direct current electron beam depending on the energy
and beam current density; on the basis of a simulation experiment on the calculation
of temperature profiles to recommend specific irradiation modes; to carry out
the exposure to radiation according to these modes.
Results and Discuss
Experiment and modeling
The protective coatings with a thickness of 150 to 300 microns were
formed on a substrate of quality carbon steel St3 (20x30x10 mm^{3 }samples) using "Impulse6"
plasmadetonation facility. They were deposited with the PG10N01 and PGAN33
(Russian standards) Nibased powder alloys.
The irradiation of samples on the side of the surfaces according to
the calculated modes was carried out in vacuum by an electron beam on
"U212" generator with an accelerating voltage of 30 kV. The scan is
sawlike; the beam travel speed in the horizontal direction is 360 mm/min; the
diameter of the electron beam on the sample is 10 mm; the current amperage is 2030mA.
The need for a detailed explanation of the coating structure scheme
stems from the fact that in order to develop a mathematical model of
temperature distribution in the coating during irradiation we have to justify
the choice of material and thickness of the irradiated layers. Resting on
reliable experimental data [57] we proposed a layered scheme of the coating
structure [8]. A thin layer (less than 5 microns) with mostly Cr oxides and
carbides forms on the coating surface. Then comes the main layer of the
Nibased coating, 100300 microns thick, then a layer of Fe (substrate), 10 000
microns thick. Because of the small thickness of the Cr layer on the surface,
this layer was neglected when calculating the temperature profile during
electron irradiation, and NiFe doublelayer coatings irradiated from the Ni
side were considered.
In order to formulate the problem of describing the heating of the
coated sample by a moving electron beam as a boundary problem of heat
conductivity theory, it is necessary to specify the density of heat sources in
a composite solid body. Since the thickness of the coating layer in which the
electron beam is almost completely absorbed is very small compared to the
thickness of the coating, and we are interested primarily in the temperature
field at the boundary surface between the coating and the substrate, we
simulate a moving beam of electrons by a moving flat normalcircular source of
a given power, i.e. we assume that the specific heat flux at a distance r from
the point of intersection of the symmetry axis of the beam with the sample
surface is given by expression (1) (without considering losses):
(1)
where q_{max}=kN/π (N beam power, N=U_{k}I, where U_{k} cathode voltage, and I – the beam amperage), and the heat flux
concentration ratio k is correlated with the heating spot radius R_{b} (the beam radius) by the formula
k=1.125/R_{b}^{2}. The analytical solution of the problem of
heating a plate of finite thickness with a moving normalcircular source
presented in the literature [9] makes it possible to roughly estimate the
maximum heat value of the points on the unheated sample surface. The corresponding
calculations for the given ranges of beam energies and the geometrical dimensions
of the sample show that the maximum heating (the difference between the maximum
temperature reached by a point and the initial temperature of the sample) for
the points on the ends and the "back" side of the plate does not
exceed 3° C. Thus, the nature of the heat exchange with the environment on the
unheated plane of the substrate and the ends of the sample has little effect on
the temperature distribution in the contact area of the substrate and coating;
and we simulate a sample by an infinite plate of thickness h lying on
the surface of the semiinfinite space filled with a material with desired
thermal characteristics.
Introducing the Cartesian coordinates by the method indicated in
Fig. 1 (X and Y axes lie in the plane of the surface coating, Z axis
points into the sample), we believe that at the time t_{0}=x_{0}/v a normally circular source begins to operate at the surface, its center moves
uniformly with velocity v along the axis X, and switches off at
time t_{1}=t_{0} (and at time t=0 corresponds to
the passage of the beam center point with the coordinates O(000).
Since the heating occurs in vacuum, we believe that the only
mechanism of heat loss from the heated surface of the coating is the heat
emission described by the StefanBoltzmann equation
(2)
where ð  beam
surface power density [W/m^{2}] s  the StefanBoltzmann constant, ε –
the emissivity factor for the coating material.
Fig.
1. Schematic representation of a twolayer sample with a moving spot during heating
by the electron beam, indicating the choice of the coordinate system
Thus, we have the following problem of heat conductivity theory:
find function T_{1 }(x,y,z,t) (temperature of the coating) and T_{2 }(x,y,z,t) (substrate temperature), as defined in areas S_{1} and S_{2} respectively (area S_{1} is defined by
the
inequalities 0 ≤ z ≤ h, t_{0 }≤ t ≤ t_{1}, while are S_{2} is
defined by the inequalities h ≤ z ≤ ∞ and t_{0 }≤
t ≤ t_{1}, at that for both areas õÎ(¥, ¥) and yÎ(¥, ¥), that comply in these areas with the differential equations (3)
and (4):
(3)
(4)
where l_{1}=l_{1}(T) the thermal conductivity of the
coating material, considered as a function of temperature and l_{2}=l_{2}(T) the thermal conductivity of the substrate material, also considered
as a function of temperature. In the calculations for computing the values of
the functions l_{1}(T) and l_{2}(T) we used polynomal interpolation on tabulated values of thermal
conductivity of nickel and iron, c_{1}=c_{1}(T) and c_{2}=c_{2}(T) – specific heat capacity of the coating and the substrate, respectively,
also considered as a function of temperature; r_{1} and r_{2} the density of the coating and
the substrate materials (the constants), when the initial and boundary
conditions described below are met: the initial conditions: T_{1}(x,y,z,t_{0})=T_{0} and T_{2}(x,y,z,t_{0})=T_{0}, where T_{0}  the initial temperature of the sample set equal to T_{0}=20°C; the boundary conditions (5), (6), (7) è
(8): at the boundary z=0 (the coating surface) – condition (5)
(5)
where P(x,y,0) – the point on the surface of the coating, and (T_{1})_{p}=T_{1}(x,y,0) and respectively,
the values of the temperature and the normal derivative of temperature at the
point , – thermal
conductivity of the coating material (depending on the temperature), – the distance
from point P to the center of the normally circular source (X_{ö}(t)=X_{0}+vt); at the boundary
between the coating and the substrate (plane z = h) must be met the two
conditions (6) and (7):
(6)
where the thermal
conductivity of the substrate material, considered as a function of temperature
(7)
that is, for all x, y and any t,
belonging to the interval (t,t_{0}) at z tending to
infinity, the temperature tends to the initial temperature of the sample Ò_{0}, condition (8)
(8)
Experiment results
The problem was being solved by the finitedifference method. We
used the data [10] for the values of the thermal conductivity, emissivity,
specific heat and density of Ni and Fe. Fig. 2 shows the dependence of the
temperature at the point with the coordinates (0,0,h) (the point that
lies at the boundary surface between the coating and the substrate) on the time
at the following design parameters: coating thickness h=300 µm, the beam power N=300W (cathode voltage U_{k}=30 kV , beam amperage I=20 mA),
beam radius R_{b}=5 mm, beam velocity v=0.004 m/s,
calculation time interval t_{1}t_{0}=14s (t_{0}=7.0s),
correspondingly x_{0}=28 mm. Fig. 2b displays the corresponding temperature dependence on the z coordinate for the point with the
coordinates x=0, y=0 at the time t=0 at the above
calculated parameters (the source switches on at the time t_{0}=7.0 s,
time t=0 corresponds to the center of the source passing the point with
the coordinates (0,0,0)).
The samples of Nibased coatings were additionally irradiated according
to the modes recommended in the result of numerical simulation calculation:
electron beam current density – 20 mA/cm^{2}, accelerating voltage – 30
kV, in the continuous exposure regime.
For practical calculations at low electron energies we needed to turn
to the experimentally obtained patterns. The empirical evidence [11] suggests
that the 12 mm thick Ni layer at the energies of the electron beam of 30 keV
is being completely absorbed. Since the depth of the total absorption of
electrons is extremely small in comparison with the thickness of coatings, a
model of surface distributed sources of heat can be taken for the construction
of the temperature profile in the sample. In our proposed model not only a high
temperature in the boundary zone is achieved, but also long enough, the order
of several seconds, holding of the area in the high temperature diapason of
400° C is provided, which allows for diffusion processes. The model enabled to
choose low current density values, which allows one to save energy for further
processing, without penetration into the coating or substrate.
a b
Fig. 2.
Dependence of the temperature of a sample point on the boundary of the substrate
and the coating on the time at the surface heating by a moving beam of electrons
(a) and the corresponding temperature dependence on the coordinate z (b)
Based on the model of temperature distribution in twolayer
absorbents with the surface distribution heat sources, the temperature profiles
were calculated according to the irradiation parameters and conditions. The
choice of materials and thicknesses of absorbent layers is based on the
experimentally developed scheme of the structure of thick plasmadetonation
powder coatings. Basing on the calculations we proposed the modes of exposure
leading to the formation of high temperatures in the coating  substrate
contact zone to accelerate diffusion processes.
Acknowledgments
This research was funded by the National Agency of Technology Development
of Kazakhstan for the projects No. 389, “Development of technology for surface
modification of by irradiation to produce nanostructured multifunctional protective
coatings with high performance properties.”
REFERENCES
1. Kadyrzhanov K.K., Komarov F.F.,
Pogrebnyak A.D., Russakov V.S., Turkebaev T.E. Ionbeam and Ionplasma Modification
of Materials.  MGU: Moscow, 2005, 425c.
2. Celik E., Ozdemir I., Avci
E., Tsunekawa Y. “Corrosion behavior of plasma sprayed coatings”, Surface and
Coatings Technology, vol. 193, 2005, pp. 297302.
3. Ganeev G.Z., Kislitsin S.B.
“The calculation of temperature fields, thermal stresses, thermal erosion
during pulsed irradiation by an electron flow”, in: Abstract Book of 7th
International Conference Nuclear and Radiaion Physics, Almaty, Kazakhsnan,
2009, p. 34.
4. Pischasov N.I., Nikolaev A.V.
“Modification of the structure and properties of hard alloys of WCCo system by
highcurrent beams of charged particles”, Communications of Omsk University 2,
1996, pp. 3943.
5. Alontseva D.L., Bratushka
S.N., Borysenko A.A., Prokhorencova N.V., Shablya V.T. Formation of Microand
Nanostructured Phases in Coatings Based on NiCr and CoCr, Their Structure and
Properties”, Metallofizika I Noveishie Teknologii, vol. 33, 2011, pp. 721745.
6. Alontseva D. “The study of
using plasma technologies for the deposition of NiCr based coatings on steel
substrate and modification of their properties by duplex treatment”,
Przegląd Elektrotechniczny, vol. 86, N7, 2010, pp. 4244.
7. Pogrebnjak A.D., Ruzimov
Sh.M., Alontseva D.L. “Structure and properties of coatings on Ni base
deposited using a plasma jet before and after electron a beam irradiation”,
Vacuum, vol. 81, 2007, pp. 12431251.
8. Alontseva D.L., “Creating the
Firm, WearProof and CorrosionResistant Coatings on Base Of NiCr and Co–Cr
Powder Alloys by Plasma Detonation Method”, in: Abstract Book of the Fourth
International Conference “Deformation & Fracture of Materials and
Nanomaterials”, Moscow, 2011, pp. 446447.
9. Rykalin N.N., Calculations of
thermal processes in welding. Moscow: Mashgiz, 1951, p. 297.
10. Smithells C.I., Metals
Reference Book, Moscow: Metallurgiya, 1980 p. 447.
11. Abroyan I.A., Andronov A.N.,
Titov A.I., Physical basis of electronic and ionic technology, Higher, 1984 p.
320.
Table of contents: The KazakhAmerican Free University Academic Journal №4  2012

