APM Papers

Modeling of Microporosity, Macroporosity, and PipeShrinkage Formation during the Solidification of Alloys Using a Mushy-Zone Refinement Method: Applications
to Aluminum Alloys
Ch. PEQUET, M. GREMAUD, and M. RAPPAZ A microporosity model, based on the solution of Darcys equation and microsegregation of gas, has been developed for arbitrary two- (2-D) and three-dimensional (3-D) geometry and coupled for the first time with macroporosity and pipe-shrinkage predictions. In order to accurately calculate the pressure drop within the mushy zone, a dynamic refinement technique has been implemented: a fine and regular finite volume (FV) grid is superimposed onto the finite-element (FE) mesh used for the heat-flow computations. For each time-step, the cells, which fall in the mushy zone, are activated, and the governing equations of microporosity formation are solved only within this domain, with appropriate boundary conditions. For that purpose, it is necessary to identify automatically the various liquid regions that may appear during solidification: open regions of liquid are connected to a free surface where a pressure is imposed, partially closed liquid regions are connected to an open region via the mushy zone, and closed regions are totally surrounded by the solid and/or mold. For partially closed liquid pockets, it is shown that an integral boundary condition applies before macroporosity appears. Finally, pipe shrinkage (i.e., shrinkage appearing at a free surface) is obtained by integration of the calculated interdendritic fluid flow over the open-region boundaries, thus ensuring that the total shrinkage (microporosity plus macroporosity and pipe shrinkage) respects the overall mass balance. This very general approach is applied to Al-Cu and Al-Si alloys.
I. INTRODUCTION
POROSITY is one of the major defects in castings, which results in a decrease of the mechanical properties, in particular the fatigue and ultimate tensile strengths.[1] It is induced by two mechanisms, solidification shrinkage and gas segregation, which occur concomitantly but with different intensities.[17] Solidification shrinkage, which induces a negative volume variation during the phase transformation of most alloys, has to be compensated for by interdendritic liquid flow (i.e., feeding) to avoid porosity. Feeding induces a pressure decrease in the mushy zone, which combined with a decrease in temperature, lowers the limit of solubility of dissolved gases in the liquid.* In addition, during solidifi*For some alloys, the pressure might even fall below the vapor pressure of some volatile solute elements (e.g., zinc).
developed mainly in one-dimensional (1-D) geometry,[2,6,8] occasionally in two dimensions[4,9,10] but rarely in three dimensions.[11,12] Lee et al.[13] recently made a fairly extensive review of existing microporosity models for Al-Si alloys. Models were classified according to the proposed approach (analytical, criteria functions, numerical model, etc.), and their advantages and drawbacks were pointed out. Recently, Vo et al.[14] developed a two dimensional (2-D) axisymmetric model, which is coupled with ABAQUS*.
*ABAQUS is a registered trademark of Hibbit, Karlsson & Sorenson, Inc., Pawtucket, RI.
cation, the gas rejection at the solid/liquid interface leads to an increased gas concentration in the liquid. If this concentration reaches a critical value, based on the limit of solubility of gas in the liquid, then pores can nucleate and grow (Figure 1). Quantitative information on the development of porosity as a function of alloy variables and casting parameters is particularly important for the foundryman, in order to control and limit the formation of such a defect. Detailed modeling of microporosity formation has been
Ch. PEQUET, PhD. Student, and M. RAPPAZ, Professor, are with the Institute of Materials, Faculty of Engineering, Ecole Polytechnique Federale de Lausanne, CH-1015 Lausanne, Switzerland. Contact e-mail: michel.rappaz@epfl.ch M. GREMAUD, Managing Director, is with Calcom SA, CH-1015 Lausanne, Switzerland. Manuscript submitted October 11, 2001.
METALLURGICAL AND MATERIALS TRANSACTIONS A
Microporosity predictions are limited to qualitative information because of restrictive assumptions of criteria functions. Sabau and Viswanathan[15] have attempted to solve the equations governing the pressure in the entire liquid and mushy regions, which introduces inherent numerical difficulties: the dynamic pressure in the fully liquid zone (usually smaller than 1 Pa) is significantly lower than the pressure drop in the mushy zone (usually on the order of several KPa). Bounds et al.[11] proposed a model for macrodefect predictions based on the coupling of the free-surface flow, heat transfer, and solidification. Numerical schemes for solving the Navier Stokes equation are complex and convergence may be difficult to obtain. Furthermore, some of the assumptions underlying the physics of microporosity formation are unclear. On the other hand, pipe shrinkage, i.e., solidification shrinkage appearing at a free surface, and macroporosity, i.e., solidification shrinkage appearing in a closed-liquid pocket (Figure 2(ab)), can be predicted by most commercial casting software but with information obtained mainly from
VOLUME 33A, JULY 20022095
Fig. 1Schematic representation of (a) a 1-D columnar mushy zone; (b) temperature and pressure profiles (gravity g is toward the left); and (c) gas concentration, [H]l, and gas solubility limit in the liquid, [H]*(T, p, cl). l
Fig. 2(a) Schematic representation of pipe shrinkage at a top free surface and of a partially closed liquid pocket. (b) A macropore, which might start to form in a partially closed liquid region (black area) when this liquid region is closed. (c) A magnified view of the mushy zone shows that only few FE nodes fall within its thickness (filled circles). (e) The FE elements are refined into small regular squares or cells (a), and those falling within mushy elements (gray elements in (d )) are activated.
the thermal field. Up to now, such calculations have never been coupled with microporosity prediction in a consistent way. Moreover, the transition between open regions of liquid (i.e., liquid regions connected to a free surface) and closed regions (i.e., liquid regions totally surrounded by the solid or the mold) has never been taken into account. At some stage, a liquid region may become totally surrounded by the mushy zone but not by the solid: for such partially closed liquid pockets, feeding can still occur via the mushy zone. The accuracy is another problem arising in microporosity prediction for geometry other than 1-D. One of the main advantages of unstructured meshes (typical of the finiteelement (FE) method) over structured ones (typical of the finite-difference (FD) method)* is that fewer elements and
*The so-called finite-volume (FV) method can be based on either structured or unstructured meshes.
while improving the accuracy in the mushy zone, a mushyzone refinement technique has been developed (Section II). The volume elements of the refined grid (hereafter, called cells)* are activated as the mushy zone moves across the
*This denomination of cell is introduced to clearly distinguish an element of the refined structured grid from an element of the unstructured coarse mesh (called element).
nodes are required to accurately describe the geometry and solidification of complex cast parts.[16] However, as a consequence, fewer nodes are also present across the mushy zone to calculate the pressure drop and microporosity formation. On the other hand, structured meshes used to calculate solidification in complex geometry are fine everywhere: this drastically increases the total number of nodes and central processing unit (CPU) time, without necessarily producing enough nodes across the mushy zone. Although the present contribution is based on a fairly classical approach of porosity modeling, first initiated by Piwonka and Flemings,[2] it introduces several new concepts. First, in order to keep the advantages of unstructured meshes
2096VOLUME 33A, JULY 2002
coarse unstructured mesh, and the equations governing microporosity formation are solved only for those cells. Second, since the governing equations of the problem, which are presented in Section III, are only solved in the refined mushy region, conditions must be defined at the boundaries of this zone (Section IV). For that purpose, it is necessary to distinguish the nature of various liquid pockets that may form during solidification (open, partially closed, and closed). It is shown that an integral boundary condition (IBC) applies for partially closed liquid pockets. In Section V, the numerical method is briefly described, whereas numerical results validating the method and showing the influence of various parameters are presented in Section VI. Although the method is mainly applied to aluminum alloys, it is generally valid and can be easily extended to other systems. II. MUSHY-ZONE REFINEMENT TECHNIQUE In the present approach, it is assumed that macroscopic phenomena occurring at the scale of the whole casting, such
as heat and mass transfer, can be described using a fixed and coarse unstructured mesh. A FE method formulation of the average conservation equation is applied, but equivalent methods based on unstructured meshes, such as FV method, work as well. However, at any given time-step, the mushy zone, sl , may only occupy a small fraction of the whole domain, , occupied by the metal (volume, V ) (Figure 2). This is particularly true for alloys solidifying with a short solidification interval, To, in a strong thermal gradient, G. In order to accurately describe the pressure drop and microporosity formation in such a zone, a refinement of this zone rather than an adaptive grid method[17] is selected because it is much simpler to implement, especially for three-dimensional (3-D) geometry, and is much faster. On the other hand, this approach is close to that previously used for the modeling of grain structure formation (Cellular Automation-Finite Element (CAFE) model[18]), and therefore, some existing numerical tools could be adapted to our needs. In the present case, the cell size, a, of the fine grid is not directly connected to the microstructure (e.g., secondary dendrite-arm spacing as in the CAFE model) but is dictated by the required number of nodes across the mushy zone, Nsl , along the thermal gradient (typically 10 to 102). If Gmax is the maximum thermal gradient in the casting and To is the solidification interval, a is given by a T0 1 Gmax Nsl [1]
pores). The local pressure in the liquid, pl(x, t), is made of several components: pl pa pm pd [2] where pa is the atmospheric (or external) pressure, pm is the metallostatic contribution, and pd is the dynamic contribution responsible for the movement of the liquid in the mushy zone (not to be confused with the much smaller dynamic pressure in fully liquid regions). In Figure 1, the negative slope of the pressure profile close to the dendrite tips is due to gravity (i.e., slope of pm), whereas the contribution of pd is clearly visible near the roots of the dendrites. Assuming that Darcys equation describes the flow in the mushy region, the superficial velocity of the interdendritic liquid, v(x, t), is given by v K grad pd K [grad pl
lg]
[3]
where K is the permeability of the solid skeleton, is the dynamic viscosity of the liquid, l is its specific mass, and g is the gravity vector. The dynamic viscosity is expressed as a function of temperature, T,
0
exp
E RT
[4]
The total number of cells in the casting is, therefore, equal to V a d, where d is the dimensionality of the problem (2 or 3). The fine and regular FV grid, covering the whole domain filled with metal, is generated and superimposed onto the FE mesh prior to the calculations (Figure 2). For each element to which they belong, the cells are numbered in a sequential order according to a scan in the x-, y-, and possibly z-directions. Since the mushy zone, sl(t), evolves with time, only those cells located in the mushy elements must be activated at a given time. An element (e) is considered to be mushy when its temperature interval, [min T, max T ](e), defined by the minimum and maximum temperatures of all its nodes, overlap the solidification interval defined by the liquidus temperature, TL , and the solidus or eutectic temperature, TS/TE . The number of activated cells during one time-step, Nc , is of the order of Nsl (V )(d 1)/d a(1 d) if the gradient is fairly strong (i.e., ToG 1 (V )1/d ) or (V a d) if the whole casting is mushy. During solidification, an inactive cell can become active, while an active cell can remain active or become inactive. III. MICROPOROSITY MODEL As in most previous works,[2,4,6,810] the present model of microporosity formation is based on two main phenomena: pressure drop in the mushy zone, on one hand, and segregation of gas/precipitation of gas bubbles (cavitation), on the other. A. Pressure Drop in the Mushy Zone The motion of liquid metal in the mushy zone is supposed to be governed by the pressure field only (in the absence of
Here, 0 is the viscosity at very high temperature, E is an activation energy, and R is the universal gas constant. The permeability, K, which reflects the resistance of the solid pattern to the fluid flow, has been deduced here from the KozenyCarmans equation:[8] K(gs(t),
2(t))
(1
gs(t))3 2(t) 2 gs(t)2 180
[5]
where gs(t) is the volume fraction of solid, and 2(t) is the secondary dendrite-arm spacing, which may be evaluated at any point of the mushy zone using a coarsening law:[19]
1/3 2 (t) 3 2,0
M(t)(t ln cl(t) cl(t)
tL ) ln cl,0 cl,0
with
[6]
M(t)
slDlTM
L(1
k)m
2,0 is the arm spacing before ripening (i.e., close to the dendrite tips), which is assumed to be equal to twice the tip radius. The M term is the coarsening factor (written here for a binary alloy, but similar law can be deduced for a multicomponent system[20]). The time, tL corresponds to the instant when the temperature reaches the liquidus (i.e., beginning of solidification), and thus, (t tL) is the time elapsed since the local temperature became lower than the liquidus. The parameters of this coarsening factor are the interfacial solid-liquid energy, sl; the diffusion coefficient of solute in the liquid, Dl; the melting point of the pure system, TM; the volumeric latent heat of fusion, L; the partition coefficient, k; the slope of the liquidus, m; and the current and the nominal concentration of solute in the liquid, cl(t) and cl,0, respectively. Consequently, the coarsening factor is also a function of time.
B. Mass Conservation Assuming that the solid phase is not moving and that there is no deformation[21] and, furthermore, neglecting the
specific mass of the bubbles, the average conservation equation of mass written for both phases gives t [ sgs
l gl ]
C. Hydrogen Mass Balance In most metallic alloys, there is a certain amount of gas dissolved in the liquid or of solute elements having a fairly high vapor pressure. Since the present contribution focuses on aluminum alloys, we will consider in the following the segregation of one gas species only, namely, hydrogen and neglect the influence on porosity formation of other elements, such as zinc*
*At 730 C, the vapor pressure of zinc is 13 kPa.
div ( lglvl)
[7]
where s is the specific mass of the solid, and vl is the effective velocity of the fluid between the solid skeleton (i.e., v glvl). If porosity formation has already occurred, the volume fraction of liquid, gl , is given by: gl 1 gs gp [8] where gp is the volume fraction of pores. Combining Eqs. [3], [7], and [8], one gets div ( K
l
(grad pl (1 gs
lg)
gp t gs
s
[9a]
l)
gs t
1. When no pore has formed. Yet at a given location (gp 0), there is only one variable, the pressure, appearing in Eq. [9], and the problem is well defined. Assuming leverrule and a locally closed system for the segregation of such a gas, the mass balance of hydrogen can be written, in this case, as [H]0
l
gp)
[H]s sgs
[H]l l(1
gs)
if gp
0 [11]
In other words, solidification shrinkage and specific mass variations, respectively, the first and two last terms on the right-hand side (RHS) of Eq. [9a], can be compensated for by feeding (the first term on the left-hand side (LHS)) and/ or by microporosity formation (the second term on the LHS). The RHS terms are supposed to be known from a heat-flow computation (including fluid flow in the liquid region or not) and from the knowledge of the specific mass of the solid and of the liquid phases. Since the product of the fraction of porosity and variation of the specific mass of the liquid is second-order, Eq. [9a] can also be written as: div K
l
where [H]0 is the nominal concentration of hydrogen in the melt, and [H]s and [H]l are the hydrogen concentrations in the solid and liquid, respectively. These last two concentrations are assumed to be related by the partition coefficient kH, i.e., [H]s kH[H]l , taking this value as equal to that of the saturated solutions (i.e., given by Sieverts law). Equation [11] then allows one to directly calculate the effective concentration in the liquid, [H]l(gs), as a function of the volume fraction of solid only. 2. As soon as porosity formation occurs. Hydrogen (or gas) conservation can be written as [H]0
l
(grad pl
0
lg)
gp t
[9b]
[H]s sgs gppp T
[H]l l(1 if gp
gs 0
gp)
[12]
d 0 T dT t
where 0 ( sgs gs)) is the average mass of the l(1 solid-liquid mixture without porosity. This value could be measured for example by densitometric measurements or calculated by using a solidification model for gs(T ) and knowing the specific mass (also possibly a function of T ) of each individual phase. In many alloys, the variations of specific mass in the liquid are much more pronounced than that in the solid[22] because of the fact that the partition coefficients, ki , are usually much smaller than unity. The term l can be written as
l l,0
is a gas conversion factor, pp is the pressure in the pores, and T is the temperature (in K). The gas concentrations in each phase are given by Sieverts law: [H]*(T, pp, cl) s [H]* (T, pp, cl) l Ss(T, cs) Sl(T, cl) pp p0 pp p0 and [13a]
[1
T (T
T0)
c(cl
cl,0)]
[10]
where l,0 is a reference specific mass of the liquid taken at the nominal temperature, T0, and liquid composition, cl,0. The terms T and c are the thermal and solutal expansion coefficients, respectively. The specific mass of the primary solid phase, s , can be considered as constant until a eutectic reaction occurs (specific mass of the eutectic, e s). As can be seen, two scalar fields appear in Eq. [9]: the pressure in the liquid, pl(x, t), and the volume fraction of microporosity, gp (x, t). This elliptic-type equation can be solved, provided appropriate boundary conditions are given for the pressure (Section IV), and these two fields are coupled via a supplementary equation of state. In the case of aluminum alloys, this additional equation is furnished by the segregation and precipitation of hydrogen.
where Ss and Sl are the temperature and solute-dependent equilibrium constants, p0 is the standard pressure, and cl is the solute concentration per unit mass. Sieverts constant for the liquid, Sl(T, cl), is correlated with temperature and composition of the liquid by the use of the following relationship:[7] Sl (T, cl) with ln K1 5872 T 1 K1 fH
c eHl cl c r Hlcl2
3.284 and log10 fH
[13b]
c c where eHl and r Hl are interaction solute coefficients on hydrogen of the first and second order.
The concentrations in the liquid and solid phases are given by the pressure in the existing pores and not in the liquid, since the reservoir of gas (i.e., the bubbles) with which these two phases are assumed to be in equilibrium is curved. The pressure in the pores is given by pp pl pr [14]
where pr is the overpressure due to the capillarity effect. The radius of curvature of the pore being r (refer to section IIID), pr is given by Laplaces law: pr pp pl 2 r
lg
[15]
where lg is the interfacial tension between the liquid and the pore (gas), which is not to be confused with the interfacial energy, sl , between the solid and the liquid (Eq. [6]). D. Nucleation Criterion and Growth Law of Pores As for the formation of a primary phase, pores must nucleate in a supersaturated liquid (cavitation) in order to overcome the curvature contribution. Three concomitant phenomena lead to a supersaturation in gas of the interdendritic liquid phase:[1] increase of [H]l due to segregation, pressure drop due to feeding, and temperature decrease. When [H]l exceeds, by a certain amount, the equilibrium value given by Sieverts law, [H]l* (T, pl , cl) (Figure 1), pores are assumed to nucleate with a given and fixed density, n0**.
**The present model of nucleation is, therefore, equivalent to setting a Dirac function of weight, n0, at the supersaturation, [ H]ln. As compared with more sophisticated nucleation models used in formation of solid phases,[23] this can be viewed as a Gaussian distribution of zero standard deviation, i.e., all nucleation sites becoming active at a unique supersaturation.
Fig. 3Various stages of a pore development: (a) just after nucleation on an oxide particle/inclusion, (b) during initial growth (constant radius), (c) later during growth (variable radius), and (d ) when it is constrained by the dendrite arms.
express the increasing difficulty for the pores to grow in between the dendritic network (i.e., pinching effect). The evolution of this radius of curvature is assumed to be given by r(gs,
2)
Defining the supersaturation for nucleation as [ H]l , pores will form if [H]l(gs) [H]* (T, pl, cl) l [ H]ln [16] In fact, [ H]ln can be converted into a critical radius of curvature of the pore nucleus, r0, by writing [H]l*(T, pl , cl) [ H]ln [H]l*(T, pp , cl), thus allowing to find pp via Eq. 15. Since pore nucleation is heterogeneous in nature (i.e., significantly large value of r0 and small value of [ H]ln), the presence of a nonwetting phase in the liquid, such as oxide particles,[1] is required (Figure 3). In gas porosity,[1] pores can be considered as spherical, since they appear at an early stage of solidification and are not too much constrained by the solid network. Consequently, Eq. 15, applied during the entire growth of the pores, allows the deduction of the pressure field, pl , and the fraction of porosity field, gp . However, initial nucleation and growth of pores in a constraining network of welldeveloped dendrites, require to describe an average radius of the curvature of the pores. The complete growth of a pore, from the instant of heterogeneous nucleation until the last growth stage when the pore morphology is constrained by the secondary dendrite arms, is shown schematically in Figure 3. In such cases, the radius of nucleation, r0, is kept constant until the volume fraction of pores corresponds to spherical pores. From that instant, the average radius of curvature is a function of the volume fraction of solid, gs , and of the secondary dendrite-arm spacing, 2, in order to
max r0; min until gs (1 ge)
3gp 4 n0
1/3
(1 gs
gs)
[17]
where ge is the final volume fraction of eutectic after solidification. IV. BOUNDARY CONDITIONS Since the coupled Darcy- and mass-conservation equations must normally be solved only in the mushy zone, it is necessary to introduce conditions at all the boundaries of this region. In previous articles dealing with microporosity formation, such computations were extended in the solid and liquid phases by using a penalty method, i.e., by setting a very small, respectively large, permeability. However, as shown by Ampuero et al.,[6] special care has to be taken for the boundary condition imposed on the eutectic front growing in between the dendrites in order to respect mass conservation and predict accurately the pressure field. In the present mushy-zone refinement technique, it is also essential to define appropriate boundary conditions at all the boundaries. As illustrated in Figure 2, it is, therefore, necessary to first distinguish the nature of the remaining liquid regions that can be present in a casting. A region of liquid is a continuous portion of the solidifying domain where only the liquid phase is present. It can be one of the following.
(1) Open if a portion of its boundary is in direct contact with another medium (usually a gas) for which a pressure can be defined (free surface). Such regions act as feeders of the mushy zone, and thus, their upper level can be lowered during solidification (i.e., formation of pipe shrinkage, Figure 2). (2) Partially closed if it is totally surrounded by a mushy zone, which is connected to an open-liquid region (i.e., this zone can still be fed by some liquid flowing from an open region of liquid). (3) Closed if it is totally surrounded by a mushy zone, which is surrounded by the solid or the mold. The boundary conditions applied on the various boundaries are as follows (Figure 4). (1) For an open-liquid region, a Dirichlet condition is imposed for the pressure at the liquidus front: pl pa pm pa
lgh
[18]
where pa and pm are the atmospheric and metallostatic pressures, respectively h being the height separating the free surface and the actual position of the liquidus front. (IIa) As long as the eutectic phase has not yet appeared at the surface of the mold (situation not represented in Figure 4), the condition to be applied at the roots of the dendrites is that of a zero velocity in the liquid. Using Darcys equation, this is equivalent to a Neumann condition for the pressure field: vl,n 0 or pl n
lgn
Fig. 4Boundary conditions applied at the boundary of the mushy zone for (I) the liquidus front of an open region of liquid, (II) the root of the mushy zone, (III) the liquidus front of a partially closed liquid pocket, (IV) the liquidus front of a closed liquid pocket, and (V) the free surface.
[19]
where n is the unit normal to the boundary, pointing outward of the mushy zone and gn g n.*
*This boundary condition is equivalent to assuming that the oxide skin at the surface prevents the interdendritic liquid from flowing in between the dendrite arms. Another possibility would have been to assume some (unknown) flow from the surface by setting a Dirichlet condition, p pa pmeniscus, where pmeniscus is a (negative) curvature contribution associated with the concave shape of the liquid surface in between dendrite arms. The calculated flow at the surface would then have been used to calculate surface porosity in a way similar to (IIIb).
(IIb) At the dendrite roots, the formation of a eutectic imposes the following mass balance,[6] vl,n
e l
enthalpy formulation of the heat-flow equation); (b) keeping the permeability constant during the eutectic reaction and equal to the value calculated at the onset of this reaction, i.e., K K(1 ge); and (c) setting the normal velocity of the fluid equal to zero at the end of the eutectic reaction (i.e., when gs 1). In the case of a 1-D geometry, Ampuero et al.[6] showed that this penalty method gives values very close to those obtained with a front-tracking technique. (IIIa) For a partially closed liquid pocket and as long as no macropore has formed in such region, an IBC must be prescribed along the liquidus front. The pressure, pQ , at some point, Q, of the boundary is set to an a priori unknown value, which must satisfy the following integral mass balance: div( lvl) dV
l l
1 ve,n
[20]
l vl
ndS
[22]
if no microporosity has formed. The term ve,n is the normal component of the eutectic front velocity, and e is the average specific mass of the eutectic. Using Darcys equation, this condition gives a Neumann condition again: pl n K gl
e l
1 ve,n
l gn
[21]
Condition [20] would normally imply the use of a front-tracking method at the eutectic/liquid interface (due to the presence of ve,n). Instead of this, a penalty method, developed by Ampuero et al.,[6] has been used to account for this condition on a fixed grid. It is equivalent to: (a) spreading the eutectic reaction over a few time-steps (this is the case anyhow with an
where l is the boundary of the partially closed liquid pocket, and n is its unit normal pointing outward of the mushy zone. The values at the other points of the boundary are then given by p pQ lgh, where h is the height separating this point and Q. In order to predict macropore formation in such a partially closed liquid pocket (IIIb), point Q is chosen as the highest location (with respect to gravity) of this region (lowest value of pQ). (IIIb) For a partially closed liquid pocket, when the pressure pQ falls below a cavitation pressure, pc , a macropore forms. If nucleation of a macropore is neglected, this pressure, pc , is directly given by Sieverts law Eq. [13], calculated with the corresponding solute concentration and temperature. In such a case, the pressure
is then prescribed (i.e., pQ p0 [H]02 Sl(T,cl,0)) 2) at point Q and the mass balance, Eq. [22], is no longer satisfied. The growth rate of the macropore volume, Vp , is directly given by dVp dt vl n dS
l
process, the volume of the void is also calculated at each time-step from (Eq. [9b]): Vp t
Nc
ad
i 1
1
l,i
o,i
[24]
[23]
Please note that the concentration of gas in the remaining liquid should normally be recalculated at each time-step according to a mass balance similar to Eq. [12] in order to account for the amount of gas in the macropore and to update the cavitation pressure with p0 [H]l2 Sl(T, cl) 2. However, considering all the other assumptions, the cavitation pressure is kept to the value given by Sieverts law for [H]0. (IV) Closed-liquid regions are treated exactly as partially closed ones in which a macropore has already formed (i.e., boundary condition (IIIb)). The only difference is that the algorithm automatically prevents any liquid to flow in such regions, i.e., the closed-liquid pocket and its surrounding mushy zone constitute a closed system. (V) At the boundary of the mushy region directly in contact with the ambient air (Figure 4), two situations might arise: (a) When the volume fraction of the solid, gs , at a point of the surface is smaller than a critical value, gs,c , mass feeding is supposed to occur. Therefore, the whole surface can move downward in order to compensate for shrinkage, and a Dirichlet condition, Eq. [18], ( p pa) is applied. The volume of pipe shrinkage appearing at this surface is obtained by evaluating the integral of the fluid flow over the open-region boundary (Eq. [23]). This volume is distributed evenly among all the cells of the surface for which gs gs,c . When all these cells are empty, a next row of cells is automatically considered, thus allowing to predict the shape of pipe shrinkage. (b) When gs gs,c at some point, the solid is supposed to remain fixed and a homogeneous Neumann condition, Eq. [19], is imposed. (The remark already made for liquid in contact with the mold (IIa) applies here.) Prescribing all the boundary conditions, as mentioned in (I) through (V), allows one to calculate the pressure in the liquid, pl, the volume fraction of microporosity, gp, and the volume of the macropores, Vp (refer to the numerical details in Section V). Furthermore, the volume of macropores is distributed among the highest cells of isolated liquid regions as solidification proceeds. Finally, at free boundaries where a Dirichlet condition is imposed (I) and (Va), the calculated pressure field allows the deduction of the velocity field at such boundaries and, thus, pipe shrinkage, i.e., lowering of free surfaces. Therefore, the whole model combines microporosity, macroporosity, and pipe shrinkage calculation in a consistent way. This is done only for the cells falling in mushy elements, therefore, with a sufficient accuracy and for the domain of interest. Finally, it should be noted that the calculation of the integral, Eq. [23], close to the liquidus has 1 to 2 pct inaccuracy because of the very wide range of permeability across the mushy zone. These errors cumulate over time. In order to respect the overall mass balance over the entire solidification
where the summation is carried out over all the mushy cells. This mass balance is then compared with the summation of microporosity, macroporosity, and pipe shrinkage volumes. The various surface integrals, Eq. [23], intervening in macroporosity and pipe shrinkage are corrected in proportion with the balance, Eq. [24], if it is necessary. V. NUMERICAL METHOD A macroscopic heat-transfer calculation is first performed in order to obtain the temperature and the solid fraction histories at all the nodes of the FE mesh falling in the solidifying domain. These results, interpolated for all the activated cells of the FV grid, are the input data for the porosity calculation. The FV formulation of Eq. [9] leads to the following set of equations: Kijplj (1 gs,i Mii gp,i)
l,i
dgp,i dt Gi
s,i
l,i)
dgs,i dt 1, Nc
[25]
d l,i dt
bi for i
where the indices i and j refer to cell locations; Kij and Mii are the rigidity and (diagonal) mass matrices, respectively; Gi is the contribution of gravity; and bi the term obtained from boundary conditions (implicit summation over repeated indices has been assumed). For the time discretization, the following implicit scheme has been used in order to avoid an instability-driven limitation of the time-steps:
n Kijplj 1
Mii (1 gs,i
n gp,i 1 l,i
gn p,i t
s,i
l,i)
gs,i t [26]
gp,i) for i
l,i
t 1, Nc
Gi
bi
The last two terms in the square brackets are obtained from the thermal calculation (shrinkage) and specific mass variation. Therefore, two unknown fields, { pl,i} and {gp,i}, have to be calculated from Eq. [26] and by applying the model of segregation/precipitation of gas (Section III). Because these two fields are related locally for each cell but through nonlinear functions, the system is nonlinear. In order to linearize the function gp,i (pl,i , Ti , gs,i), one iteration of Newtons method is performed for each node.[6]
n gp,i 1 n
gn p,i (T in
1
gp pl T n) i
n n ( pl,i i 1
pn ) l,i
n (gs,i 1
[27]
gp T
gp gs
gn ) s,i
The first derivatives of gp , with respect to pl , T, and gs are calculated analytically from Eq. [12] and using the related
equations, such as Sieverts and Laplaces laws (Eqs. [13] and [14]). The linear system obtained in this way can be rewritten as Aijpl,j fi for i 1, Nc [28]
where Aij is the matrix of the system, and fi is the RHS term. This system is solved using a preconditioned variant of the Bi-Conjugate Gradients method.[24] As soon as a partially closed liquid region appears during solidification, the linear system is modified because of the presence of a new variable, pQ , and of a new equation corresponding to the formulation of the IBC, Eq. [22]. For the cell located at Q, the formulation of the IBC over the whole boundary gives one equation in which pQ is the unknown. For any other cell, R, located at the boundary, where the unknown pressure pR is related to pQ , the corresponding equation is simply: pR pQ lghRQ , where hRQ is the height separating these two cells. VI. RESULTS AND DISCUSSION Several test cases of increasing complexity are presented in this section. First, a 1-D situation is considered in order to validate the implementation of the IBC for partially closed liquid pockets. Then, several 2-D calculations are performed in order to illustrate the effects of various parameters and to verify the accuracy of the model. Finally, a 3-D situation with a large number of cells clearly shows the efficiency of the algorithms. The 2-D and 3-D versions of microporosity formation have been implemented in the corresponding versions of the software CALCOSOFT,* this later being used
*The software CALCOSOFT, a joint development of the Ecole Polytech nique Federale de Lausanne and Calcom SA, Switzerland, is dedicated to the 2-D/3-D modeling of continuous processes, such as continuous casting, direct-chill casting, strip casting, as well as of advanced solidification processes. Fig. 51-D test of the IBC implemented for the pressure in partiallyclosed liquid regions.
to perform the thermal calculations. This FE method software was also modified to handle the cell definition, and visualization of porosity results. In order to test the IBC (IIIa), the 2-D model was first applied to a 1-D thermal situation, neglecting microporosity formation (Figure 5). For that purpose, a very narrow domain was selected with appropriate thermal boundary conditions at the lateral surface. Cooling conditions on the lateral side were chosen in order to obtain a partially closed liquid pocket located in between two mushy zones. The transverse Biot number, i.e., the product of the heat-transfer coefficient and transverse dimension of the domain, divided by the thermal conductivity of the alloy, was very small to ensure a nearly uniform temperature across the width. The thermal boundary conditions at both ends of this 1-D domain were adiabatic. The gray levels at the top of Figure 5 indicate the amount of solid according to the scale on the right. For the pressure drop calculation, it was assumed that no liquid could flow on the lateral sides of the domain and on the left (i.e., mold). On the right boundary, a Dirichlet boundary condition was applied, and gravity was assumed to be horizontal, toward the left. For the sake of simplicity, the specific masses of the solid and liquid were assumed to be constant but different. Calculations of the pressure drop in the domain were performed with and without introducing a cavitation pressure. When no cavitation pressure is introduced (dashed curve
of Figure 5), the pressure first increases from the atmospheric pressure in the open-liquid pocket because of the metallostatic contribution. The pressure drop in the first mushy zone is such that the liquid velocity entering the partially closed liquid pocket exactly compensates for shrinkage in the second mushy zone near the mold (left boundary). The pressure in this cavity, unknown at the beginning, is around 70 kPa at the highest point (with respect to gravity) and also increases with depth. Setting up a cavitation pressure of 80 kPa (continuous curve) shifts the pressure profile upwards to this value. As a consequence, the amount of liquid flowing from the open cavity will be lowered, while solidification shrinkage in the mushy zone is the same. Therefore, the mass balance given by Eq. [23] directly gives the volume of the macropore that appears in the partially closed liquid region. The amount of liquid flowing from the right boundary is also lowered, but one clearly sees the difficulty in predicting the amount of pipe shrinkage from the integral (Eq. [23]) applied to the liquidus position on the right. The pressure profiles of the dashed and continuous curves are nearly identical because the permeability in this region is very large. This is why it is necessary to correct such integral(s) with the help of the overall mass balance (Eq. [24]). In all the 2-D and 3-D examples that are shown hereafter, the volume fraction of microporosity is indicated with the help of a yellow-green palette, whereas macroporosity and pipe shrinkage are indicated by orange and red colors. The scale of the first palette is given for each figure, while orange and red mean 0 to 50 pct and 50 to 100 pct volume fraction of void, respectively. Moreover, the amount of hydrogen is set to 0.15 ccSTP/100 g* of metal unless it is explicitly given.
*For aluminum alloys, 0.1 cm3 of hydrogen at standard conditions of pressure and temperature in 100 g of metal corresponds roughly to 1 ppm.
The first 2-D example shown in Figure 6 has been inspired by the A1-4.5 wt pct Cu sand-mold casting calculated by Kubo and Pehlke.[4] For the heat-flow simulation, a uniform heat-transfer coefficient of 42 W m 2 K 1 was imposed at all the sand-metal boundaries, and an adiabatic condition
Table I. Material Properties for the Various Test Cases Shown in Figures 6 through 10 Al-4.5 wt pct Cu
sl
Al-7 wt pct Si 1 10 1 5 10 9 660 9.5 108 0.132 6.64 0.51 9 10 1 1 10 5 1 109 2380 2520 2520 7.0 5.36 10 4 1.65 104 3 10 2 8 10 4
1 2
Al-11 wt pct Si 1 10 1 5 10 9 660 9.5 108 0.132 6.64 0.85 9 10 1 1 10 5 1 109 2380 2520 2520 11.0 5.36 10 4 1.65 104 3 10 2 8 10 4 Jm 2 m2 s 1 C Jm 3 K (wt pct) J m 2 m m 3 kg m 3 kg m 3 kg m 3 C 1 (wt pct) 1 wt pct Pa s J mole 1
Dl TM L k m ge
lg
r0 n0
l,0 s e T c
cl,0
0
E c eHi c r Hi R
1 10 1 5 10 9 660 1 109 0.173 3.434 0.09 9 10 1 1 10 5 1 109 2440 2620 3400 1.24 10 4 1.09 10 2 4.5 5.36 10 4 1.65 104 3 10 2 4 10 4
8.3144 J mole 1 C ccSTPH2 K s2 m 269 100 g
was applied to the top free surface of this L-shaped casting. The other conditions and parameters are indicated in Table I. Figure 6 represents the final amount of porosity in the casting for two different nominal concentrations of hydrogen with and without pipe shrinkage. In Figure 6(a), the amount of hydrogen is 0.15 ccSTP/100 g of metal, and the top surface of the L-shaped casting is supposed to be connected to a riser (but the metallostatic head of the riser is neglected). Since the casting is well fed, the amount of microporosity is fairly limited (around 0.5 pct or lower). Under exactly the same conditions, a doubling of the initial concentration of hydrogen (Figure 6(b)) more than doubles the amount of microporosity (as much as 1.1 pct). As already noted by Rousset et al.,[22] microporosity must first offset the undersaturation of the melt and then overcome the nucleation barrier (Figure 1). Keeping the concentration of hydrogen at 0.15 ccSTP/100 g but assuming now that the top surface is not connected to a riser (Figure 6(c)) leads to a very large pipeshrinkage cavity: the shape of which is dictated by that of gs,c . The amount as well as the the isotherm when gs distribution of microporosity are also completely different. The largest amount of microporosity is no longer found near the end of the casting but close to the bottom of the pipe shrinkage. Such behavior seems in agreement with usual casting practices and results obtained by Kubo and Pehlke.[4] The second 2-D example, shown in Figure 7, illustrates the influence of the grain structure on the amount of microporosity, macroporosity, and pipe shrinkage for an A1-4.5 wt pct Cu Y-shaped casting. The mold is assumed to be around the whole component, except on the left-top surface of the Y shape (gravity is downward). The parameters being given in Table I, two calculations have been performed by just changing the value of the critical volume fraction of solid, gs,c , above which a free surface is assumed to remain fixed.
For columnar structure growing from the side of the mold (Figure 7(a)), the value of gs,c has been set to zero and feeding can only occur from the portion of the top surface that is fully liquid. As a consequence, feeding of the mushy zone becomes difficult early during solidification, and a macroshrinkage cavity forms within the casting. At the same time, pipe shrinkage is limited, while the amount of microporosity is high (between 8 and 10 pct). With the assumptions presently made, no pipe shrinkage is predicted near the right-top surface of the Y-shaped casting. It is equivalent to assuming that the mold is impermeable to air, which might not be true. However, such hypothesis could be relaxed by setting an appropriate pressure at this surface (Dirichlet condition) instead of a Neumann condition. Assuming now that the alloy was inoculated and equiaxed grains form, mass feeding can occur, i.e., the free surface can move downward even if the melt is already mushy. Setting arbitrarily the value of gs,c to 0.2 (Figure 7(b)), one can see that feeding is much more effective. The macroshrinkage cavity has disappeared, microporosity has been reduced, but pipe shrinkage is much more pronounced. In both cases, the total void formation (i.e., cumulated volume of microporosity, macroporosity, and pipe shrinkage) is equal to 8.33 10 4 m2, which corresponds to the overall solidification shrinkage of the alloy (10.25 pct, Table I, volume of the casting equal to 8.05 10 3 m2). In order to test the sensitivity of the model to solute concentration, the third 2-D example, shown in Figure 8, corresponds to the solidification of A1-7 wt pct Si (AS7) and A1-11 wt pct Si (AS11) alloys in a 2-D axisymmetric geometry. Again, this dummy blade is supposed to be surrounded by the mold everywhere, except at the top free surface where the atmospheric pressure is imposed. The two calculated results have been mounted in a symmetric fashion so as to clearly reveal the differences induced solely by the alloy concentration; all the other parameters having been kept identical. As can be expected from the extent of the mushy zones of these two alloys, the amount of microporosity in the AS7 casting is greater than in the AS11 one. Concomitantly, the cavitation pressure is reached at the midheight horizontal platform of the dummy blade during the solidification of the AS7 alloy, while some liquid remained. As a result, a macroporosity has formed near the upper surface of this platform, which is not the case for the AS11 alloy. This example also demonstrates that the model can account for macrocavity forming near (or at) an internal surface, which is still liquid. On the other hand, the critical solid fraction gs,c , being and equal to 0.2, pipe shrinkage in AS11 is larger than in AS7; the total volume of void in the two castings remaining nearly the same (2.6 10 5 m3). The last 2-D example (Figure 9) demonstrates the ability of the model to handle multiple liquid pockets and shows the influence of the metallostatic pressure and cooling conditions on the final microporosity and macroporosity. In this cross-shaped casting, only the top free surface is exposed to air. The upper square volume is, therefore, the only part that can feed three other volumes labeled A, B, and C. Gravity being downward, the metallostatic head of cavities A and B is the same, while it is larger for cavity C. Cooling conditions of cavities B and C are the same (heat-transfer coefficient equal to 5 Wm 2K 1), whereas cavity A is cooled faster (10 Wm 2K 1). Cooling conditions on the narrow
Fig. 6Influence of the nominal concentration of gas (a and b) and of the absence of a riser (c) connected to the top surface on the microporosity level and pipe shrinkage in a L-shape Al-4.5 wt pct Cu sand mold casting (conditions in Table I). Fig. 8Influence of the alloy concentration on the final microporosity, macroporosity, and pipe shrinkage in a dummy axisymmetric turbine blade of Al-Si alloys (conditions in Table I).
Fig. 7(a) and (b) Influence of the microstructure via the critical volume fraction of solid, gs,c , on the final microporosity, macroporosity, and pipe shrinkage in a Y-shape Al-4.5 wt pct Cu sand mold casting (conditions in Table I). Fig. 9Influence of the metallostatic pressure and cooling conditions on the final microporosity, macroporosity, and pipe shrinkage in a cross-shape Al-4.5 wt pct Cu alloy (conditions in Table I).
sections of the cross-shaped casting are dictated by a heattransfer coefficient of 150 Wm 2K 1. Although these cooling conditions are quite far from experiments, the result shown in Figure 9 clearly indicates that an increased metallostatic head simply delays the formation of macroporosity (compare cavities B and C) and reduces its final volume. Comparing now cavities A and B (same metallostatic head), the macroporosity developing in A is reduced, as the liquid in this region is cooled faster and feeding is more efficient. Please note the important volume of pipe shrinkage in the upper part of the cross-shaped casting. To conclude this section, Figure 10 shows the case of a real 3-D casting of 4.3 m3. Although this casting was originally made in steel, it was replaced in the present case by an A1-4.5 pct Cu alloy. The casting having a plane of
symmetry, only half was modeled. The two extreme risers were cut by the plane of symmetry, while the two central ones were not. Figures 10(a) and (b) show the temperature and solid fraction repartitions at a given time in the casting, with the corresponding scale on the top left corner, as obtained with CALCOSOFT. The porosity model was then run with more than half-a-million cells (total CPU time of about 31 hours on a 686 PC running under Linux). Figures 10(c) and (d) show the final porosity repartition and pipe shrinkage in the casting, as calculated with the parameters given in Table I. Figure 10(c) corresponds to the surface
Fig. 10(a) Temperature and (b) solid fraction repartitions at a certain time and (c) final microporosity level and; (c,d ) pipe shrinkage in a 3-D Al-4.5 wt pct Cu casting.
and symmetry plane section, whereas Figure 10(d) corresponds to selected transverse sections. As can be seen, pipe shrinkage is very important in the extreme right riser, which has the largest volume and remains liquid for a long period of time, whereas the three other risers have moderate pipe shrinkage. There is no macroshrinkage cavity, which indicates that rigging was correctly made, and the amount of microporosity is maximum near the pipe-shrinkage holes.
VII. CONCLUSIONS A FV model for the prediction of microporosity, macroporosity, and pipe shrinkage during the solidification of alloys has been developed for 2-D (cartesian and axisymmetric) and 3-D geometry. It includes all the basic physical phenomena, which are at the origin of microporosity. In particular, pressure drop in the mushy zone, segregation of gas, equilibrium
between gas bubbles and solid-liquid phases, laws of nucleation and growth of pores are taken into account. Pipe shrinkage and macroporosity are predicted by detecting automatically open, partially closed, and closed liquid regions and by applying appropriate boundary conditions. A mushyzone tracking procedure has been developed in order to have a good accuracy in the mushy zone, while limiting the number of nodes and CPU time. The model seems to give the correct trend for the influence on porosity and pipe shrinkage of various parameters, such as gas content, alloy concentration, nucleation parameters, cooling conditions, and gravity. In future work, numerical results will be compared with experimental measurements. ACKNOWLEDGMENTS This work was performed within the framework of the CTI Project, Contract No. 3666.1. This project has been officially certified as Eureka Project No. E!2091 gaining an internationally recognized quality label of technological innovation. The financial support of Assan (Turkey), Calcom (Switzerland), Doncasters (United Kingdom), Elkem (Norway), Hydro Aluminium (Norway), Pechiney (France), Rolls-Royce (United Kingdom), Sintef (Norway), Snecma (France), VAW (Germany), and The Swiss Commission for Technology and Innovation (CTI, Bern) are gratefully acknowledged. The authors also thank S. Pellerin, V. Maronnier, and J.-F. Joyeux, Calcom, for the development of the postprocessing, cell definition modules, and integration of the present model into CALCOSOFT, respectively; J.-L. Desbiolles, EPFL, for his valuable advice and various contributions; and B. Commet, Pechiney, for a remark regarding the hydrogen mass balance. REFERENCES
1. J. Campbell: Castings, Butterworth-Heinemann, 1991.
2. T.S. Piwonka and M.C. Flemings: Trans. AIME, 1966, vol. 236, pp. 1157-65. 3. S. Shivkumar, D. Apelian, and J. Zou: AFS Trans., 1989, vol. 97, pp. 989-1000. 4. K. Kubo and R.D. Pehlke: Metall. Trans. B, 1985, vol. 16B, pp. 359-66. 5. M. Rappaz: in Advanced Course in Solidification, Calcom SA, Lausanne, Switzerland, 1996. 6. J. Ampuero, C. Charbon, A.F.A. Hoadley, and M. Rappaz: in Materials Processing in The Computer Age, V.R. Voller, M.S. Stachowicz, and B.G. Thomas, eds., TMS, Warrendale, PA, 1991, pp. 377-88. 7. D. Carpentier: Ph.D. Thesis, Institut National Polytechnique de Lorraine, Nancy, France, 1994. 8. D.R. Poirier, K. Yeum, and A.L. Mapples: Metall. Trans. A, 1987, vol. 18A, pp. 1979-87. 9. J.D. Zhu and I. Ohnaka: in Modeling of Casting, Welding and Advanced Solidification Processes V, M. Rappaz, M.R. Ozgu, and K. Mahin, eds., TMS, Warrendale, PA, 1991, pp. 435-42. 10. J. Huang and J.G. Conley: Metall. Mater. Trans. B, 1998, vol. B29, pp. 1249-60. 11. S. Bounds, G. Moran, K. Pericleous, M. Cross, and T.N. Croft: Metall. Mater. Trans. B, 2000, vol. 31B, pp. 515-27. 12. P.D. Lee and J.D. Hunt: Acta Mater., 2001, vol. 49, pp. 1383-98. 13. P.D. Lee, A. Chirazi, and D. See: J. Light Met., 2001, vol. 1, pp. 15-30. 14. P. Vo, D. Maijer, C. Hermesmann, and S.L. Cockcroft: Light Met. 2001, J.L. Anjier, ed., TMS, Warrendale, PA, 2001, pp. 1115-21. 15. A.S. Sabau and S. Viswanathan: Light Metals, TMS, Warrendale, PA, 2000, pp. 597-602. 16. M. Gremaud and M. Rappaz: AFS Trans., 2002, in press. 17. N. Provatas, N. Goldenfeld, and J. Dantzig: J. Comp Phys., 1999, vol. 148, pp. 265-90. 18. C.-A. Gandin, J.-L. Desbiolles, M. Rappaz, and P. Thevoz: Metall. Trans. A, 1999, vol. 30A, pp. 3153-65. 19. W. Kurz and D.J. Fisher: Fundamentals of Solidification, 4th revised edition, Trans Tech Publications, Aedermannsdorf, Switzerland, 1989. 20. M. Rappaz and W.J. Boettinger: Acta Metall. Mater, 1999, vol. 47, pp. 3205-19. 21. T. Kajitani, J.-M. Drezet, and M. Rappaz: Metall. Mater. Trans. A, 2001, vol. 32A, pp. 1479-91. 22. P. Rousset, M. Rappaz, and B. Hannart: Metall. Mater. Trans. A, 1995, vol. 26A, pp. 2349-57. 23. M. Rappaz: Int. Mater. Rev., 1989, vol. 34, pp. 93-123. 24. S.-L. Zhang: Siam J. Sci. Comput., 1997, vol. 18 (2), pp. 537-51.
MCWASP 2006 Nice France Edited by TMS (The Minerals, Metals & Materials Society), Year
A POROSITY MODEL FOR MULTI-GAS SYSTEMS IN MULTI-COMPONENT ALLOYS

G. Couturier(1,2), J.-L. Desbiolles(1,2), M. Rappaz(1) Computational Materials Laboratory, School of Engineering, Ecole Polytechnique Fdrale de Lausanne, Station 12, CH-1015 Lausanne, Switzerland (2) Calcom ESI, PSE-A, CH-1015 Lausanne, Switzerland Keywords: porosity, gas microsegregation, pressure drop, gas thermodynamics. Abstract A general framework for the modelling of porosity formation in multi-component alloys with more than one gaseous element is considered in the present contribution. It offers several advantages and accounts for: i) the partial pressure of any gaseous element composed of one or two chemical elements (e.g., H2, N2, CO, etc.); ii) the influence of the alloy composition on partial pressures through chemical activities; iii) the account of both trace gaseous elements and volatile solute elements such as zinc through appropriate mass balances. The set of equations describing multi-gas equilibrium at a given location is first described, with the construction of appropriate databases for aluminium-, copper- and iron-base alloys. These local state equations are coupled to a macroscopic resolution of the Darcy-mass balance equations governing the pressure drop in the mushy zone. This solution is based on an evolving fine volume grid superimposed to a finite element mesh used for the heat flow computations [1]. A few applications illustrate the effects of process and alloy parameters on the final porosity fraction. Introduction Porosity in castings is a major defect since it affects the mechanical properties [2-6], in particular the initiation of fatigue cracks [2-4]. Therefore, the reduction of porosity fraction and size, the control of porosity distribution and morphology are crucial for the optimization of fatigue behaviour of as-cast components. Porosity is the result of two concomitant mechanisms: (i) solidification shrinkage induces a suction and thus a liquid pressure drop in the mushy zone (Darcys law [7]), (ii) trace gaseous elements in the liquid being generally less soluble in the solid phase, solidification induces gas microsegregation in the remaining liquid part. Cavitation may occur in the mushy zone when the effective gas concentration in the liquid, w lg , reaches the gas solubility, w l* . The solubility furthermore decreases with liquid pressure and temperature. g In aluminium-base alloys, hydrogen is the only diatomic gas enough soluble to lead to porosity formation [8-10]. Hydrogen in aluminium alloys is due to air moisture decomposition (H2O2H+O) and aluminium oxidation [11] (3H2O+2AlAl2O3+3H2) and is also produced by the decomposition of moisture or grease covering tools immersed into the melt. In copper-base alloys, hydrogen, moisture, and sulphur dioxide are responsible of porosity formation [12], whereas this defect is due to hydrogen, nitrogen and carbon monoxide in iron-base alloys [13]. The combined effects of hydrogen and nitrogen in steel has already been studied [14,15]. Carbon solubility in copper-base alloys seems to be too low to form carbon monoxide [10]. As the affinity of the reaction of formation of FeO is higher than that of SO2 [16], SO2 is never formed
(1)
in iron-base alloys. Some volatile solute elements (e.g., zinc) can also contribute to microporosity [17]. Several authors reported the effect of alloying elements on gas solubility in aluminium-, copper-, and iron-base alloys [11, 12, 14, 16]. The general framework of the present contribution is the formation of porosity in the presence of one or more gases, the solubility of which being dependent on several alloying elements. The basic conservation and thermodynamics equilibrium equations that govern porosity formation in the case of a multi-gas system in multi-component alloys are established in the first part. A realistic growth law for a pore constrained by the dendrite network is proposed in the second part. In the final part, this multi-gas approach is applied to the case of porosity formation in copper-base alloys. The impact on porosity fraction of the presence of H, S and O in solution in the alloy is studied for various solidification conditions. Multi-Gas and Multi-Component Approach The purpose of this part is not to describe again the equations that govern porosity formation in the case of one diatomic gas, g, (e.g., H2 in an aluminium-base alloy). These equations have been largely explained in previous papers [1,17]. In this part, a method for the extension of the approach to several gases composed of one or two chemical elements soluble in a multicomponent alloy is proposed. Gas Thermodynamics The gases responsible of porosity formation are generally composed of one or two chemical elements that can be solute elements of the alloy or not (see introduction). While Sieverts law applies to diatomic gases, the more general case of a gas composed of one or several elements is handled through solubility products. The involved reactions for gases made of maximum two elements are:
Formation of gas ( = 1, ng): n A A or n A A + n B B with = A n or A n Bn (1) A A B
There are ng gases contained in the gaseous phase and ns chemical elements involved in the formation of these gases (represented by capital letters). n indicates the stoechiometry of the A element A in gas (e.g., 2 for H in H2). On the left hand side of these reactions, the elements are dissolved in the liquid phase. The solubility product for gas formed from A and B elements is recalled here:
p p0
(f
l A
X l* A
) (f
n A
l B
X l* B
n B
-G 0 = exp = K (T) RT
(2)
p is the partial pressure of gas , G 0 ( T ) the variation of the standard Gibbs free energy for the
l reaction of formation of gas , K ( T ) is the constant of the gas formation reaction, f A and l f B are the activity coefficients of elements A and B, and X l* and X l* the molar fractions in the A B
liquid phase of elements A and B, respectively, in equilibrium with the gas phase. The Gibbs free energy variation, G 0 ( T ) , can be expressed as a function of the standard enthalpy, H 0 ( T ) , and 0 the standard entropy S ( T ) (usually assumed constant in the temperature range):
G 0 ( T ) = H 0 ( T ) - TS0 ( T )
(3)
Eq. (2) can be transformed into the following relation:
(f
'
p p0
l A
X l* A
) (f
n A
* l A
' l B
n A
X l* B
n B
0 H ' = A exp = K (T ) RT
n B
(4)
with A =
(f ) (f)
* l B
S0 exp R
and
'
l fA =
l fA * l . fA
l l The activity coefficient of A, * f A , corresponding to the pure liquid metal, it follows that ' f A is equal to 1 in the absence of solute elements. The parameter A has no physical meaning but will be called the gas formation coefficient.
Similar expressions to relation (4) are obtained for other gases, and these expressions are the first l* ng dependency relations proposed between the (ng + ns) unknown variables X A (A = 1, ns), p l ( = 1, ng). The activity coefficient ratio ' f A is given by the following relationship [11]:
' l f A = 10 S 2 eS cS +rA cS A S
(5)
S where eS and rA are the first- and second-order interaction coefficients of the solute element S A on the gaseous element A, respectively. In this relation, the solute element concentrations, cs, are expressed in wt%. Interaction coefficients for several solute elements in various alloys (Al, Cu, 0 Fe,), as well as H and A can be found in the literature [11,12,14,16,18].
A simple thermodynamic analysis [17] has shown that solute elements with a high vapour pressure (e.g., zinc) can contribute to the increase of the pore fraction in high melting point alloys (e.g., copper-base alloys). Relation (4) is also perfectly adapted to describe the transformation of such solute elements into a vapour phase. Gaseous Element Conservations The gas element conservation equation established in reference [17] for a diatomic gas in the presence of porosity can be easily extended to a multi-gas system. For each element A, one has:
X lAo = l (1- g s ) X l* + s g s k A X l* + M o A A alloy
gp
RT g A
g A
pg
(6)
where X lAo and X l* are the nominal molar concentration and the solubility limit in the liquid A phase, respectively, and kA is the partition coefficient of gaseous element A. l and s are the specific masses of the liquid and solid, respectively, while is the average specific mass of the solid-liquid mixture in the absence of porosity. gs and gp are the volume fractions of solid and porosity, respectively, M o alloy is the molar mass of the alloy in the initial state (before
solidification), T is the temperature and R the perfect gas constant. In this equation, the summation of the gas partial pressures is carried out for all the gases containing element A, with a weight given by the stoechiometry of element A in each gas. In the presence of porosity, Eqs (6) provide ns additional relationships between the unknown l* l* l* variables X A , X B , ..., X N , p , ..., p and gp. Therefore, Eqs (4) and (6) provide ng+ns relations in which ng+ns+1 unknown variables are present, i.e., one equation is still missing. In the absence of porosity, relation (6) becomes:
X =
l A
l (1- g s ) + s g s k A
X lAo
(7)
Mechanical Equilibrium of a Pore An additional equation is provided by the mechanical equilibrium condition of a pore:
p p = p + ... + p = p l + p r
(8)
where pp and pl are the pressures in the pore and in the surrounding liquid , respectively, while pr is the Laplace contribution associated with the curvature of the pore:
p r =
2 gl r
(9)
where gl is the surface energy of the pore/liquid interface and r the radius of curvature of the pore. Relation (8) introduces two unknown variables, pl and r (i.e., ng+ns+3 unknowns with ng+ns+1 equations), but two more equations are provided by: The mass conservation equation coupled with Darcys law that relates the liquid pressure with the porosity fraction. This relation has been largely detailed in previous papers [1,17] and will not be repeated here. A relationship between gp and r. For a spherical pore, this relationship is straightforward, while a simple model for a pore constrained to grow within a dendritic network is developed in the next section.
In order for a pore to nucleate in the liquid, its initial radius of curvature, ro, must satisfy Eq. (9), i.e., the supersaturation, p + + p - pl, must be equal to the Laplace contribution. Solution The mass conservation equation coupled with Darcys law is solved using an evolving fine volume grid superimposed to a finite element mesh used for the heat flow computations [1]. The set of ng+ns+2 equations governing at a local scale the relationship gp(pl) in a multi-gas system is strongly non linear. In order to limit the computation time, these equations are solved for each grid of the mesh by one step of Newton-Raphtons method. This is equivalent to deriving Eqs l* l* (4), (6) and (8) in order to obtain a linear system with respect to dX A , dX B , ..., dp , dp ..., dp l and dgp.
Growth Law
As the interface energy, gl, is on the order of 1 Jm-2 and the pore curvature radius is equal to a few tens of micrometers, the curvature contribution (Laplaces overpressure, Eq. (9)) cannot be neglected in Eq. (8) and strongly influences the pore fraction. While the relationship between gp and r is straightforward for spherical pores (i.e., gas porosity), a simple model for the curvature of a pore constrained to grow in a well developed dendritic network (i.e, shrinkage porosity) is derived in this section.
Figure 1. Regular stacking of dendrite arms, showing the space available for pores (rmax): hexagonal arms without impingement (left), and cylindrical arms with impingement (right). The left scheme of Fig. 1 is an illustration of a too simple geometrical solution that does not take into account secondary dendrite arms impingement. The present model considers a simplified 3-dimensional network of cylindrical secondary dendrite arms and takes into account their impingement (right scheme in Fig. 1). Assuming that pores can grow in between the cylindrical arms, assumed to be infinite in length, the maximum radius of the pore is simply given by:
12 g (i.e. before impingement), rmax = 2 2 - s 2 4 2 12 1 r r 1 g s = 2 - 2 max + max + 4 - arcos rmax 4 2 2 4 2 -2 2
if g s
(10)
else,
1 r 2 - max . 2 2
As the liquid phase is assumed to completely wet the solid, the contact angle at the triple point (pore-solid-liquid) is zero. The relationship given by Eq. (10) is shown in Fig. 2, together with another relation, rmax = 0.5 2(1 - gs0.5), that corresponds to the solution of the left geometry in Fig. 1. It appears that the new relation (14) seems more adapted to the modelling of shrinkage porosity. Indeed, with rmax = 0.5 2(1 - gs0.5), rmax 2 does not exceed 0.025 for g s 0.9 . Taking 2 = 40 m, rmax will be smaller than 1 m, and pr will be greater than 1.8 MPa. Therefore, pores will have almost no chance to grow if they do not nucleate before gs = 0.9. Doing the same calculation with Eq. (10) ( rmax 2 < 0.15 for gs > 0.9), the curvature contribution is on the order of 300 kPa during the last stage solidification, thus allowing shrinkage porosity formation. Eq. (10) will be retained for the simulations presented in the last part.
Results
The present porosity model has been applied to brass alloys, more specifically to a Cu-10%wtZn alloy. For the solidification path, the Scheil-Gulliver microsegregation model was used. The solidification range was assumed equal to 43 C (no peritectic reaction was considered). The resolution of Darcy+mass balance equations is possible if solidification shrinkage is known [1]: the liquid and solid specific masses were assumed constant and equal to 7940 and 8960 kg/m3, respectively. In order to speed up the computations, an ideal one-dimensional directional casting was chosen. The casting velocity, v, and the thermal gradient, G, were equal to 0.01 m/s and
500 C/m, respectively. The simulation results deal with the porosity fraction for the stationary regime, showing the impact of hydrogen, moisture and sulphur dioxide on porosity fraction.
0.8 0.7
1.50 H, S and O
relations (10)
1.25
H and O H
0.6 0.5 rmax/2 (-)
0.5*(1-gs1/2)
1.00 gp(%)
S and O
0.4 0.3 0.2
0.75
0.50
0.25
0.1 0.0 0.0
0.00
0.1 0.2 0.3 0.4 0.5 0.6 gs (-) 0.7 0.8 0.9 1.0
2= 50 m
shrinkage
shrinkage no shrinkage no shrinkage 2= 25 m 2= 50 m 2= 25 m
Figure 2. Representation of the maximum radius of a pore (normalized by the secondary dendrite arm spacing) growing in a mushy zone for the two geometrical models shown in Fig. 1.
Figure 3. Porosity fraction in Cu-10 wt%Zn for different gaseous element systems l ( X lHo = 30 ppm (at) , X So = 0.8 % (at) and X lOo = 20 ppm (at) if present). Two secondary dendrite arm spacings were tested, without and with account of solidification shrinkage.
In Fig. 3, the contribution to porosity formation of various gaseous elements is represented for two secondary dendrite arm spacings, with and without shrinkage. It can be seen that dissolved H2, H2O and SO2 can have a concomitant effect on the microporosity level. The selected oxygen nominal concentration is low (20 ppm (at)) in order to avoid Cu2O precipitation in the liquid phase (see Cu-O phase diagram), a situation that the model is unable to handle if X lO path is unknown. The chosen sulphur nominal concentration is high (0.8 % (at)), i.e., at least twice the concentration usually added to improve machinability. It is observed that the presence of hydrogen creates a significant amount of porosity (diamonds), but this amount is drastically increased with the presence of oxygen (formation of water vapour). This figure also shows that the presence of sulphur can produce porosity if no deoxidation step was employed before pouring. In Fig. 4, the porosity fraction is represented as a function of hydrogen and oxygen nominal concentrations for different secondary dendrite arm spacings, without and with account of solidification shrinkage, using the thermal gradient and isotherm speed mentioned before. Several thousand computations were made, but the presence of sulphur was not considered. In Figs 4(a) and (b), it is observed that, at high O and H nominal concentrations, the porosity fraction slightly decreases with the secondary dendrite arm spacing. At low gaseous element concentrations, the porosity level is larger for smaller 2 values because the permeability is reduced [19] (i.e., the pressure drop is larger). This effect is apparently larger than the opposite one associated with the curvature contribution (i.e., higher pr with smaller 2). On the other hand, solidification shrinkage has a strong influence on the porosity level. This is confirmed in Fig. 4: (i) for high nominal concentrations of gaseous elements, the porosity fraction is significantly lower when the shrinkage is not account for, and (ii) the porosity fraction is zero at low gaseous element concentrations, whereas it is not zero when shrinkage is accounted for. These maps would allow the determination of the gaseous element nominal concentrations below
which the porosity fraction will be lower than a given value. Of course, these concentrations strongly depend on the process and alloy parameters.
3.0
3.0
2.5
2.5
2.0
2.0
(% )
1.5
1.5
1.0 0.5 40 35 30 25 40 35 30 25 20
o
1.0 0.5 40 35 30 25 40 35 30 25 20
o
20 15 10 5 5 0
(p pm
15 10 5 5 0
Ho
10 0
10 0
(a) 2 = 50 m, with shrinkage.

3.0
(b) 2 = 25 m, with shrinkage.

3.0
2.5
15 (p p m a t)
15 (p p m a t)
Ho
(p pm
at)
20
at)
2.5
2.0
(% )
2.0
1.5
1.5
1.0 0.5 0.0 40 35 30 25

at)
1.0 0.5 0.0 40 35 30 25 35 30 25 20

o
35 30 25
X
O o
20 15 20 15 (p p m a t) 5 5 0 10
20 15 10 5 5 0
(p pm
Ho
10 0
15 (p p m a t)
10 0
(c) 2 = 50 m, without shrinkage .
(d) 2 = 25 m, without shrinkage.
Figure 4. Maps of the porosity fraction in Cu-10wt%Zn as a function of H and O nominal concentration and for different secondary dendrite arm spacings, without and with account of solidification shrinkage.
Conclusion
In this paper a general approach to model porosity in multi-component alloys for multi-gas systems has been detailed. An application to Cu-Zn has shown the influence of various gas elements, of the secondary dendrite arm spacing and of solidification shrinkage. The contribution of the vapour pressure of volatile solute elements such as zinc is detailed elsewhere [17]. This model is being validated on several aluminium alloys.
Ho
(p pm
at)
40
40
(% )
(% )
Acknowledgements
The authors would like to thank the financial support of the Commission for Innovation and Technology, CTI, Bern (grant 6167.1 KTS), and of the industries Alcan (CH), Alcan (FR), Calcom-ESI (CH), HydroAluminium (DE), General Motors (USA) and Union Minire (BE).
References
1. Ch. Pequet, M. Gremaud, and M. Rappaz, Modelling of microporosity, macroporosity, and pipe-phrinkage formation during the solidification of alloys using a mushy-zone refinement method: applications to aluminium alloys, Met. Mater. Trans., 33A (2002) 2095. 2. M.J. Couper, A.E. Neeson, and J.R. Griffiths, Casting defects and the fatigue life of an aluminum casting alloy. Fatigue Fract. Eng. Mater. Struct., 13 (1990) 213. 3. B. Skallerud, T. Iveland, and G. Hrkegard, Fatigue life assessment of aluminum alloys with casting defects, Eng. Fracture Mech., 44 (1993) 857. 4. Q.G. Wang, D. Apelian, and D.A. Lados, Fatigue behaviour of A356-T6 aluminum cast alloys. Part I. Effect of casting defects, Journal of Light Metals, 1 (2001) 73. 5. M. Garat, Effets respectifs de la finesse de structure et de la compacit sur les caractristiques mcaniques statiques et dynamiques de lA-S7G06, Fonderie Fondeurs daujourdhui, nov. (1989) 21. 6. M. Morishita, K. Nakayama, and M.G. Chu, Effect of Hydrogen Porosity and As-Cast Grain Structure on the Mechanical Properties of Cast and Forged Al-1.0Mg-0.6Si Alloy, Solidification Of Aluminum Alloys, Eds Men G. Chu et al (TMS Pub., Warrendale, USA, 2004) p. 283. 7. H. Darcy, (1856), Les Fontaines Publiques (Dalmont Publ., France). 8. J. Campbell, Castings (Elsevier, 2003). 9. Smithells Metals Reference Book, 7th edition, Eds E.A. Brandes and G.B. Brook (Butterworth-Heinemann Publ., Woburn, MA, USA, 1992). 10. J. Charbonnier, Gaz dans les alliages daluminium de fonderie, Trait Matriaux Mtalliques, (Techniques de lIngnieur, 1991) article M218. 11. G. K. Sigworth, and T. A. Engh, Chemical and kinetic factors related to hydrogen removal from aluminium. Met. Mater. Trans., 13B (1982) 447. 12. Series on the metallurgy of copper, (The International Copper Research Association, Library of Congress, USA, 1977). 13. ASM Casting Handbook, Vol. 15, 4th edition, Ed. D.M. Stefanescu et al (ASM Intern., Metals Park, Ohio, USA, 1998). 14. P.K. Sung, D.R. Poirier, and S.D. Felicelli, Continuum model for predicting microporosity in steel castings, Modelling Simul. Mater. Sci. Eng., 10 (2002) 551. 15. D.R. Poirier, M.M. Andrews, and A.L. Maples, Modeling macrosegregation and porosity in steel castings, 1st International Steel Foundry Congress (Steels Founders' Society of America, Des plaines, IL, USA, 1985), 307. 16. J. Philibert, A. Vignes, Y. Brchet, and P. Combrade, Mtallurgie du minerai au matriau (Masson, Paris, 1998). 17. G. Couturier, and M. Rappaz, Effect of volatile elements on porosity formation in solidifying alloys, Modelling Simul. Mater. Sci. Eng., (2005) to appear. 18. Landolt-Bornstein, Phase equilibria, crytallographic and thermodynamic data of binary alloys, New series, group IV, macroscopic properties of matter, (Springer Verlag, 1994). 19. D. R. Poirier, Permeability for flow of interdendritic liquid in columnar-dendritic alloys, Met. Mater. Trans. 18B (1987) 245.
Symposium on Simulation of Aluminum Shape Casting Processing Edited by Qigui Wang

TMS (The Minerals, Metals & Materials Society), 2006
MODELING OF POROSITY FORMATION IN MULTICOMPONENT ALLOYS IN THE PRESENCE OF SEVERAL DISSOLVED GASES AND VOLATILE SOLUTE ELEMENTS
G. Couturier1,2 and M. Rappaz1
1
Computational Materials Laboratory, Ecole Polytechnique Fdrale de Lausanne MXG, Station No 12, CH-1015 Lausanne, Switzerland 2 Calcom-ESI SA, Parc Scientifique, CH-1015 Lausanne, Switzerland
Keywords: Porosity, Gas microsegregation, Pressure drop, Volatile elements, Zinc Abstract Although microporosity is a critical issue for the mechanical properties of cast parts, modeling of such defects is still not yet fully satisfactory. The present contribution addresses the problem of multicomponent alloys and multigas systems including volatile solute elements. The effects of all solute elements on the activity of gas elements (hydrogen in the case of aluminum alloys) are first taken into account, together with the partial vapor pressure of volatile elements such as zinc. After writing an overall mass balance for each element contributing to porosity formation, a linearized system of equations can be solved to relate the increment of porosity to the local pressure. This general formalism has been implemented in a program that calculates the pressure drop in the mushy zone using Darcys equation and an adaptive refined mesh. Examples related to aluminum alloys will be shown in order to emphasize the various contributions to porosity formation. Introduction Mechanical properties of cast aluminum alloys, in particular fatigue resistance and toughness, are affected by the presence of microporosity.[1] As shown in Fig. 1, porosity is the result of two concomitant mechanisms: solidification shrinkage and segregation of gases. The higher density of the solid induces a suction of the viscous liquid towards the permeable mushy zone, thus decreasing the pressure in the liquid. On the other hand, dissolved gases such as hydrogen in aluminum alloys are usually less soluble in the solid phase. Being segregated in the remaining l liquid part, the gas in the liquid can reach a concentration, wg , which exceeds the solubility l limit, wg*(T,pl), especially since this limit decreases with the temperature, T, and pressure in the liquid, pl. At this stage, nucleation and then growth of pores has to be considered. A previous paper has outlined a general framework for the calculation of microporosity formation in binary alloys in the presence of one gas element.[2] It is briefly summarized in the next section. Subsequent improvements of this model for multi-component and multi-gas systems have been given in [3]. The contribution to porosity formation of volatile solute elements such as zinc has been considered in [4]. A summary of these extensions follows. Finally, the last section presents a few applications of this model implemented in the software CalcoSOFT and ProCAST.[5]
(a)
(b)
Figure 1: Schematic representation of (a) the liquid pressure drop in the mushy zone due to solidification shrinkage and (b) the increase of gas concentration, w lg , in the mushy zone due to gas microsegregation (thick line) and the decrease of the solubility limit, w l* (T,pl) (thin line). At equilibrium, i.e., without the Laplace overpressure, g cavitation occurs at the crossing point of these two curves.
General Theoretical Framework Figure 2 shows a schematics of a simplified casting. Solidification induces three types of voids: (i) at a free surface (e.g., of risers), the level of liquid decreases as solidification proceeds (piping); (ii) within closed liquid pockets (hot spots), a macropore surrounded by microporosity will be present; (iii) microporosity finally dispersed within the mushy zone might finally appear at an early stage of solidification (gas porosity) or deep in the mushy zone, especially when a dense interdendritic phase forms (shrinkage porosity). An open region of liquid has at least one surface in contact with a gas of known pressure. A closed liquid pocket is totally surrounded by the solid or mold. Before it becomes closed, it can be partially closed, i.e., it is still connected to an open region of liquid through the mushy zone.
Piping Open liquid pocket TL TS
Microporosity
Finite Element Mesh
Active elements
Macroporosity
Closed liquid pocket
Casting Fine Volume Grid Figure 2 : Schematics of a casting showing piping at a free surface and macroporosity formation in a closed liquid pocket. A zoom of the unstructured FEM mesh and structured fine volume grid are shown on the right (see text).
Since microporosity develops within the mushy zone, calculations must be performed for the solid-liquid two-phase region only. For that purpose, a dynamic mushy-zone tracking procedure has been developed. Using a fixed finite element (FEM) unstructured mesh, heat and fluid flow computations are performed at the scale of the whole casting + mold. At each time step, the position of the mushy zone and thus the mushy elements are known. For such elements, volume elements of a structured grid are activated for the calculation of the pressure drop. Pressure In The Liquid The mushy zone is considered to be a fixed porous medium within which liquid flows according to Darcys equation: K (1) v= [grad pl - lg] where v is the average velocity of the liquid, l its density and its dynamic viscosity. g is the gravitation vector and K is the permeability of the mushy zone, assumed to be given by CarmanKozenys relationship: (1 - gs(t))3 2(t) K(gs(t), 2(t)) = g (t)2 180 s
2
(2)
gs is the local volume fraction of solid and 2 is the secondary dendrite arm spacing given by a coarsening law. Combining Eq. (1) with the mass conservation equation of the alloy gives a second-order partial differential equation in pressure: gp K div l [grad pl - lg] + l = (3) t t gp is the volume fraction of porosity and is the average density of the solid-liquid mixture in the absence of porosity. Such a density can be calculated from the density, (T), of the various phases () and their volume fraction, g(T). In the absence of porosity (i.e., gp = 0), Eq. (3) can be solved on the fine structured grid shown in Fig. 2 providing appropriate boundary conditions are given on the liquidus and solidus surfaces of all the regions of the mushy zone. These conditions are detailed in [2]. However, in the presence of porosity, a local equation describing the relationship between pl(r,t) and gp(r,t) must be found. It is given by thermodynamic considerations. Thermodynamic Cconsiderations The gases responsible of porosity formation are generally composed of one or two chemical elements that can be solute elements of the alloy or not. In the case of aluminum alloys, H2 is the main gas responsible of pore formation. In the 7xxx series alloys, the influence of the high vapor pressure of Zn on porosity formation has never been assessed. Once supersaturation is locally exceeded, a pore of double average curvature, , forms ( = 2R-1 for a spherical portion of a liquid-gas surface, R being the radius of curvature, see Fig. 3). Mechanical equilibrium imposes that: pp =
p
i
= pl + sl
(4)
pp = pi R pl
Figure 3: Schematics of a pore constrained in a dendritic network. The internal pressure within the pore, pp, is given by the sum of the partial pressures of the various gases. Mechanically, it is also equal to the pressure in the liquid, pl, plus the surface tension contribution associated with its radius of curvature, R.
liquid
solid
where the pis are the partial pressures of the gases present in the pore (internal pressure pp) and sl is the interfacial gas-liquid energy. For gases made of one or two chemical species (case usually encountered in solidification processes: H2, N2, CO, CO2, SO2, etc.), the partial pressures are given by solubility products [3,6]: pi l l* = Ki(T) 'fA XA po
nA
i
l l* 'fB XB
nB
for a gas (i) of composition AniABniB
(5)
po is a reference pressure (e.g., atmospheric pressure) and Ki(T) a constant that can be derived l from the enthalpy and entropy of formation of the gas (i). The coefficients, 'fA and 'flB, are the activity coefficients of elements A and B in the liquid alloy, normalized by the corresponding activity coefficient for the pure liquid. (Note that the concentrations of the elements A and B in solution in the liquid are given here in atomic fraction). In the case of a diatomic gas such as H2 i i (nA = 2, nB = 0), Eq. (5) retrieves the well known Sieverts law. The coefficients 'flA are given by [7] :
l 'fA
= 10
esAcs + rsAc2 s s
(6)
The summation is carried out over all the solute elements (s) composing the melt with the concentration cs (wt%). The coefficients entering into this development are given for hydrogen and the most common solute elements of aluminum alloys in Table I.
Table I: 1st and 2nd-order coefficients for various solute elements in Al alloys (after [7]). Ce Cu Cr Fe Mg Mn eA [wt%-1] -0.08 0.03 0. 0. -0.01 0.06
s
rA [wt%-2] -0.0004 . -
Ni Th Ti Si Sn
eA [wt%-1] 0.04 -0.006 -0.1 -0.03 0.004
rA [wt%-2] -0.0008 -
Equations (5) provide as many equations as there are gases in the pore. These equations must be complemented by element conservation equations. For each chemical element (A) entering into the pore formation, one has: XAo = l(1-gs) XA + sgs kAXA + Malloy
l* l* o
gp RT
iA
np
i A i
(7)
XAo is the nominal molar concentration in the liquid phase and kA the partition coefficient of gaseous element A between the solid and liquid phase. M o is the molar mass of the alloy in alloy
the initial state (before solidification) and R is the perfect gas constant. In this equation, the summation of the gas partial pressures is carried out for all the gases containing element A, with a weight given by the stoichiometry of element A in each gas. Equations (7) allow to locally close the problem: if the local pressure, p, in the liquid is known at a given point of the fine grid, Equations (4), (5) and (7) provide the right number of equations to find gp, providing a relationship between gp and (or R) is known. For spherical pores, this is fairly straightforward. For pores pinched in a dendritic network (Fig. 3), various empirical relationships R(gs) can be deduced from simple geometrical considerations. As the formation of pores relaxes the pressure drop in Eq. (3), the solution of Darcy and mass conservation equations must be coupled. The set of equations governing at a local scale the relationship gp(pl) in a multigas system being strongly non linear, they are solved for each grid of the mesh by one step of Newton-Raphtons method. Finally, this general approach is also perfectly appropriate to describe the contribution of volatile solute elements such as zinc to pore formation. In this case, however, the mass balance (7) is not needed as the liquid can be considered as an infinite source of vapor.
Results
One of the first input to consider in a microporosity model is the pore-free density, , of the alloy. It requires two separate inputs: the density of each phase that can form during solidification, and the evolution of the volume fraction of each phase. While this latter entity can be calculated from a phase diagram computation such as ThermoCalc, the densities must be obtained from a database. Using the database of the software ProCAST, the evolution of the density as a function of temperature has been calculated for a multi-component alloy Al0.6wt%Mg-0.7wt%Si-0.1wt%Fe-2.3wt%Cu (Fig. 4).
1 0.9 0.8 0.7 Solid fraction (-) 0.6 0.5 0.4 0.3 0.2 0.1 0 510 2400 650 2450 Solid fraction Mixture specific mass 2500 2550 2600 Mixture specific mass (kg/m3) 2650
530
550
570
590
610
630
Temperature (C)
Figure 4: Representation of the solid fraction and of the specific mass of the mixture of phases as a function of temperature for an Al-0.6wt%Mg-0.7wt%Si-0.1wt%Fe-2.3wt%Cu alloy (Scheil assumption).
Once (T) is known, the microporosity calculation can be performed but several parameters are not straightforward: if the level of hydrogen in the melt can be measured, the major unknowns are those related to nucleation (supersaturation at which pores are nucleated, density of nuclei) as well as the evolution of the radius of curvature of the pores during growth. Assuming that pores form on oxide films [1], their density is given by that of oxides while supersaturation at the onset of nucleation can be fairly low (nuclei could even pre-exist in folded films). In the present calculations, the supersaturation was set as pr = 400 or 200 kPa (i.e., initial radius of curvature the nucleus equal to 5 or 10 m, respectively). But the growth of the pores is governed by the space available in between the dendrites arms. Considering the impingement of cylindrical dendrite arms arranged in a square lattice (Fig. 5), the evolution of the maximum radius of curvature of a perfectly non-wetting pore, Rmax, can be easily computed. The continuous curve, Rmax(gs), shown on the right of Fig. 5 is compared with a simpler model of hexagonal arms arranged in a hexagonal network for which Rmax = 0.52(1 gs0.5).[8,9] With this last model, Rmax does not exceed 0.0252 for gs > 0.9. Taking 2 = 40 m, Rmax will be smaller than 1 m, i.e., pr will be greater than 1.8 MPa, which is unrealistic. Pores will have almost no chance to grow if they do not nucleate before gs = 0.9. Doing the same calculation with the model shown in Fig. 5, Rmax < 0.152 for gs > 0.9. (about 6 m) and thus pr > 300 kPa during the last stage of solidification, thus allowing shrinkage porosity formation. Considering the spherical ends of secondary arms, i.e., cubic arrangement of spheres, the conditions are even more favorable for pore formation, since Rmax < 0.22 and pr > 225 kPa beyond 90% fraction of solid.
1 normalised maximum pore radius
0.8
0.6
0.4 3D spheres 0.2 2D cylinders 0.5(1-gs ) 0 0 0.2 0.4 0.6 0.8 1 solid fraction
0.5
Figure 5: Impingement of cylindrical dendrite arms (above) and maximum radius of curvature of a nonwetting pore growing in between and normalized by 2 (right). The case of spheres (end of cylindrical arms) arranged on a cubic network is also represented. The dashed line corresponds to a simple impingement of hexagonal arms arranged in a hexagonal network.
Once all the parameters are defined, the model can show the influence of various process or material parameter. In Fig. 6, the influence of the chemical activity of the solute elements Mg, Si, Fe and Cu on the pore fraction is shown for a 1-dimensional Al-0.6wt%Mg-0.7wt%Si0.1wt%Fe-2.3wt%Cu casting. The top three figures correspond to the final porosity level, while those at the bottom are maps of the last pressure in the liquid. The top black part in these maps corresponds to piping, while the porosity level or pressure are indicated with various grey levels. Microporosity varies between 0 and 2.1%, while the last pressure scale is from 100 kPa (atmospheric pressure) to -200 kPa. Please note that these negative pressures in the remaining liquid are not unrealistic if the added curvature contribution finally produces a positive pressure, pp, in the pore.
According to Table I, copper and silicon decrease the solubility limit of hydrogen in liquid aluminium, whereas magnesium increases it. As the magnesium concentration is lower than that of copper, it is expected that for a given nominal hydrogen concentration, the final porosity fraction will be greater if the activity of solute elements is taken into account. In Fig. 6b, it is observed that this effect is weak effect but positive when only the nominal concentrations are taken into account. However, when microsegregation of the solute elements is also included in the calculation of the activities, this effect is reinforced (Fig. 6c). As porosity formation occurs earlier in this case, this also limits the pressure drop. As the pressure drop is decreased, this also limits the amount of porosity, reason why such effects are attenuated. Please note that these calculations were done at fixed nominal hydrogen level (0.15 ccSTP/100 g of metal). The effect of Mg on the tendency to oxidise the melt in the presence of water, thus resulting in an increase of hydrogen in the melt, was not considered.
(a)
(b)
(c)
Figure 6: Representation of the final porosity fraction (top) and last pressure in the liquid (bottom) for a cylindrical casting of Al-0.6wt%Mg-0.7wt%Si-0.1wt%Fe-2.3wt%Cu: (a) the solute elements effect on the hydrogen activity is not accounted for; (b) the solute element effect is taken into account, but not solute segregation; (c) both the solute element effect and the solute microsegregation are taken into account. Increasing microporosity levels are indicated with lighter greys, decreasing pressure in the liquid (except the atmospheric pressure) with darker greys.
The effect of zinc solute element is considered next.[4] Pure zinc, which is an alloying element in Al 7xxx series alloy and in brass, has a high vapour pressure (it melts at 419 C and boils at 907 C). Its partial vapour pressure will of course be influenced by the activity coefficient when it is dissolved in aluminium- or copper-base alloys. ThermoCalc calculations show that the activity coefficient in Al (Henrys law) is about 2.4, while it is only 0.14 in Cu. Using this information and accounting at the same time for the segregation of zinc during solidification, the vapour pressure of zinc in Al-5wt%Zn and in brasses of various zinc concentration (Cu10wt%Zn, Cu-20wt%Zn and Cu-30wt%Zn) is shown in Fig. 7 as a function of the solid fraction, gs.
600
20000 18000
500 wZn,0 = 5 wt% 400 pZn (Pa)
16000 14000 12000
300
pZn (Pa)
10000 8000
200
6000 4000 2000

0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 gs (-) 0.7 0.8 0.9 1.0
wZn,0 = 30wt% wZn,0 = 20wt% wZn,0 = 10wt%
100
0 0.0
0.1
0.2
0.3
0.4
0.5 0.6 gs (-)
0.7
0.8
0.9
1.0
Figure 7: Vapor pressure of zinc in Al-5wt%Zn (left) and in brasses of various concentrations (right).
In a first step, one can see that the vapour pressure in aluminium first increases as a result of segregation of zinc, but then decreases during the last stage of solidification as the temperature decrease dominates. However, the pressure never exceeds 1 kPa, thus indicating that zinc has a negligible contribution to pore formation in Al alloys. Such is not the case in brasses. Indeed, at the melting point of copper, the vapour pressure of pure zinc is 460 kPa (i.e., 4.6 times the atmospheric pressure). However, the activity coefficient being only 0.14, the effect is decreased when zinc is alloyed with copper. As can be seen in Fig. 7, the pressure can reach 10 kPa in such alloys, i.e., 10% of the atmospheric pressure. It can therefore contribute to porosity formation in such alloys. Further investigations on this effect can be found in [4]. A calculation of porosity performed for an industrial high pressure die casting is shown in Fig. 8. The geometry is shown on the left. Applying a pressure of only 1 bar (central figure), the amount of porosity which is shown with various grey levels is fairly high. Macropores are even present at some locations (e.g., dark zone at the centre of the zoomed region). The application of a pressure of 100 bars during solidification (figure on the right) allows to completely eliminate these macropores as well as the microporosity generated by gas segregation and pressure drop. However, with the given cooling conditions, the previous hot spots remain now as isolated mushy pockets for which the application of a high pressure has no effect. There are no longer holes in these regions, but a high level of microporosity due to uncompensated solidification shrinkage.
Finally, the porosity module presented here has been also coupled with a stress calculation in order to predict hot tearing.[10] Using a semi-coupled approach, deformation of the partially coherent solid skeleton is first calculated using a porous continuous media theory. In such approach, the compressible solid can deform plastically while neglecting the much lower liquid pressure in the mushy zone. Once the strains in the solid, , are known, an additional 0, where v is the deformation speed of the solid, is added to contribution, div (vs) = Tr s Eq. (3). If the solid experiences tensile strains, this induces more liquid suction, whereas compressive zones can compensate partially solidification shrinkage. The result of such computations for a DC cast AA5182 alloy is shown in Fig. 9.[10]
Figure 8: Example of the type of porosity calculation that can be carried out for a complex shape A383 casting (Courtesy of Calcom-ESI). The geometry is shown on the left, while the calculated porosity fraction is displayed with various grey levels for a few sections when the applied pressure is 1 bar (center) and 100 bars (right).
Figure 9: Results at 500s in a DC cast AA5182 alloy when both shrinkage and volumetric strain are taken into account (a) temperature distribution (b) solid fraction (c) Mises stress (d) liquid pressure (After [10]).
Conclusion
The porosity model initially developed for binary alloys and one gas element has been extended to multi-component multi-gas systems. Based on a thermodynamic approach, it can predict piping, macro- and microporosity, providing the gas content and nucleation parameters are given. While an evolving fine mesh allows to follow the mushy zone and to give a good accuracy for the liquid pressure calculation, the curvature contribution remains a key parameter in such computations. A pinching model based on cylindrical or spherical arms has been introduced, but further work is certainly needed in this area, e.g., using phase field simulations. Finally, this porosity model has been coupled with stress-strain calculations in order to also predict hot tearing.
Acknowledgements
The authors would like to thank the financial support of the Commission for Innovation and Technology, CTI, Bern (grant 6167.1 KTS), and of the industries Alcan (CH), Alcan (FR), HydroAluminium (DE), General Motors (USA) and Union Minire (BE). They also thank JeanLuc Desbiolles for his help in the implementation of the model into CalcoSoft. They also thank V. Mathier, EPFL, and Dr. Ph. Thvoz, Calcom-ESI, for providing the hot tearing result and the simulation of an actual shape casting.
References
1. John Campbell, Casting (Elsevier, 2001). 2. Ch. Pquet, M. Gremaud, and M. Rappaz, Modeling of Microporosity, Macroporosity and Pipe Shrinkage Formation during the Solidification of Alloys using a Mushy-Zone Refinement Method, Met. Mater. Trans., 33A (2002) 2095. 3. G. Couturier, J.-L. Desbiolles, and M. Rappaz, A Porosity Model for Multi-Gas Systems in Multi-Component Alloys, Modeling of Casting, Welding and Advanced Solidification Processes 10, Eds C.-A. Gandin et al (TMS Publ., Warrendale, USA, 2006), to appear. 4. G. Couturier, and M. Rappaz, Effect of volatile elements on porosity formation in solidifying alloys, Mod. Sim. Mater. Sc. Engng., (2005) to appear. 5. CalcoSOFT and ProCAST are two commercial software distributed by Calcom-ESI, Lausanne, Switzerland. 6. L.S. Darken, R.W. Gurry, and M.B. Bever, Physical Chemistry of Metals (Mc Graw Hill, New York, 1953). 7. G. K. Sigworth, and T. A. Engh, Chemical and Kinetic Factors Related to Hydrogen Removal from Aluminum, Met. Trans., 13B (1982) 447. 8. D.R. Poirier, K. Yeum, and A.L. Maples, A Thermodynamic Prediction for microporosity Formation in Aluminium-Rich Al-Cu Alloys, Met. Trans., 18A (1987) 1979. 9. J.D. Zhu, and I. Ohnaka, Computer Simulation of Interdendritic Porosity in Aluminium Alloy Ingots and Casting, in Modeling of Casting, Welding and Advanced Solidification Processes 5, Eds M. Rappaz et al (TMS Publ., Warrendale, USA, 1991), p. 435. 10. V. Mathier, J.-M. Drezet, and M. Rappaz, Two-Phase Modeling of Hot Tearing in Aluminum Alloys using a Semi-Coupled Method, Modeling of Casting, Welding and Advanced Solidification Processes 10, Eds C.-A. Gandin et al (TMS Publ., Warrendale, USA, 2006), to appear.
INSTITUTE OF PHYSICS PUBLISHING
MODELLING AND SIMULATION IN MATERIALS SCIENCE AND ENGINEERING doi:10.1088/0965-0393/14/2/009
Modelling Simul. Mater. Sci. Eng. 14 (2006) 253271
Effect of volatile elements on porosity formation in solidifying alloys

G Couturier1,2 and M Rappaz1
1
Computational Materials Laboratory, School of Engineering, Ecole Polytechnique F d rale de e e Lausanne, Station 12, CH-1015 Lausanne, Switzerland 2 Calcom-ESI SA, Parc Scientique de lEPFL, CH-1015 Lausanne, Switzerland E-mail: gael.couturier@ep.ch and michel.rappaz@ep.ch
Received 23 August 2005, in nal form 10 November 2005 Published 15 February 2006 Online at stacks.iop.org/MSMSE/14/253 Abstract Besides dissolved gases such as hydrogen, the partial vapour pressure of volatile solute elements, such as zinc, has been accounted for in equations governing the nucleation and growth of pores in solidifying alloys. In particular, a simple analytical solution giving the porosity fraction as a function of the fraction of solid is proposed. This solution is then used to study the inuence of zinc on porosity formation in aluminium- and copper-base alloys. It is shown that the zinc vapour pressure is too low to increase the porosity of aluminiumbase alloys, whereas it has a signicant effect in brasses. Implementing this contribution into an existing software which calculates the pressure drop in the mushy zone and the segregation of dissolved gases, simulations have been performed in order to assess in more realistic situations the inuence of volatile solute elements on porosity.
1. Introduction Porosity that is formed and trapped in the mushy zone of solidifying alloys is the result of two concomitant mechanisms (gure 1): (i) solidication shrinkage induces a suction of the viscous liquid towards the fully solid region, thus creating a liquid pressure drop in the mushy zone [1]; (ii) as molten alloys usually contain traces of gaseous elements that are less soluble in the solid phase, solidication induces an enrichment of gas in the interdendritic liquid. Cavitation may occur in the mushy zone when the effective gas concentration in the liquid, wg , reaches the gas solubility, wg . This solubility limit is decreasing when both the pressure and temperature are decreased, thus favouring the formation of porosity. In aluminium-base alloys, hydrogen is the only diatomic gas soluble enough to lead to porosity formation [24] (see table 1). Hydrogen in aluminium alloys is due to air moisture decomposition (2H2 O 2H2 + O2 ) and aluminium oxidation (3H2 O + 2Al Al2 O3 + 3H2 ). It is also produced by the decomposition of moisture or grease introduced by tools immersed into the melt or deposited on the mould
0965-0393/06/020253+19$30.00 2006 IOP Publishing Ltd Printed in the UK 253
254
G Couturier and M Rappaz
(a)
(b)
Figure 1. Schematic representation of (a) the liquid pressure drop in the mushy zone due to solidication shrinkage and (b) the increase of gas concentration, wg , in the mushy zone due to gas microsegregation () and the solubility limit, wg (-). At equilibrium, i.e. without pr , cavitation occurs at the crossing point of these two curves.
surface. In copper-base alloys, hydrogen, moisture and sulphur dioxide are responsible for porosity formation, whereas in iron-base alloys it is associated with hydrogen, nitrogen and carbon monoxide [5, 6]. Several factors are used to limit or reduce gas concentration in alloys [5, 6]: working with cleaned and pre-heated tools and moulds, decreasing the melt superheat in order to limit gas absorption into the melt, maintaining the melt at the liquidus temperature for some time in order to reach its solubility limit at this temperature and uxing the melt with inert or reactive gases (e.g. in aluminium alloys, N2 , Ar or AlCl3 small bubbles provide efcient hydrogen removal before solidication). In addition, oxidationdeoxidation steps are used to reduce the hydrogen content in copper-base alloys (addition into the melt of oxygen-enriched copper to facilitate the reaction 2H + O H2 O, followed by copper deoxidation just before casting to avoid steam porosity obtained by the reaction of hydrogen at the mould surface with cuprooxides (Cu2 O + 2H 2Cu + H2 O)) [4]. This deoxidation step is essential, as Cu2 O cannot be extracted by gravity; phosphorus is used because cupro-phosphorus oxides oat at the melt surface. Deoxidation can also be accomplished with carbon (a graphite crucible) combined with a vacuum treatment to facilitate the removal of CO. To avoid re-oxidation from air, a charcoal cover is added on the top of the melt. As zinc also contributes to the deoxidation of copper and forms a strongly stable oxide, brasses generally have no problem of steam and sulphur dioxide porosity. In addition, zinc can limit hydrogen pick-up by the melt at high temperature since its vapour pressure dramatically
Effect of volatile elements on porosity formation
255
Table 1. Solubility limit and partition coefcients of H2 , N2 and O2 in aluminium, copper and iron [1, 2, 3]. Values at atmospheric pressure and melting point H solubility ( ) (ccSTP /100 g) H solubility (s) (ccSTP /100 g) H partition coefcient Equilibrium compound N solubility ( ) (ccSTP /100 g) N solubility (s) (ccSTP /100 g) N partition coefcient Equilibrium compound O solubility ( ) (wt%) O solubility (s) (wt%) O partition coefcient Equilibrium compound Al 0.7 0.05 0.07 H2 0 0 \ \ 0 0 \ Al2 O3 Cu 5.2 1.9 0.36 H2 0 0 \ \ ? 0.0077 ? Cu2 O Fe 24.5 6.9 0.28 H2 35.2 10.3 0.29 N2 0.18 0.009 0.05 FeO
increases. But if hydrogen remains in the melt, the high zinc vapour pressure could also contribute to gas porosity formation. Modelling of porosity formation has been undertaken by several authors [813] but in most cases for only one diatomic gas (often H2 ). Sung et al have considered the case of two gaseous elements, namely H2 and N2 [13]. In this paper, the contribution of volatile solute elements is taken into account in equations governing the nucleation and growth of pores in solidifying alloys. An analytical solution for the porosity fraction evolution as a function of the solid fraction is proposed. On the basis of this analytical solution, the effect of zinc on porosity formation in aluminium- and copper-base alloys is then demonstrated. Finally, the contribution of volatile solute elements has been implemented in a microporosity software module [14] which calculates the pressure drop in the mushy zone and microsegregation of dissolved gases. This module is based on an evolving ne volume grid for the solution of Darcys and mass conservation equations in the mushy zone, superimposed to a nite element mesh used for the heat ow computations at the scale of the casting. The simulation results are presented and discussed.
2. Theory Besides the effect that solute elements can have on the equilibrium of gas in metallic alloys, they can also contribute to microporosity formation in two ways if their vapour pressure is large enough. They can rst lower the concentration of gas in the liquid which is required to induce cavitation (i.e. nucleation conditions) and they can increase the pore fraction (i.e. growth conditions). Both points are claried in this part through a description of the equations that govern the mechanical equilibrium of a pore, the gas formation constant, the gas conservation and the vapour pressure of volatile elements.
2.1. Basic equations As explained below, cavitation begins earlier in the presence of volatile elements. In the presence of gases and volatile solute elements, the mechanical equilibrium of a pore obeys the
256
following relationship: pg +
g i
pi = p + p r ,
(1)
where pg and pi are the partial pressures of gas (g) and volatile alloying elements (i), respectively. p is the local pressure in the liquid surrounding the pore and pr is the Laplace contribution associated with the curvature of the pore: 2g pr = . (2) r g is the surface energy at the pore/liquid interface and r is the radius of curvature of the pore. In order for a pore to nucleate in the liquid, most likely on foreign particles such as oxides or intermetallics [2], its initial radius of curvature, ro , must satisfy equation (2), i.e. the supersaturation, g pg + i pi p , must be equal to the Laplace contribution. Providing solute elements or gases do not modify the interfacial energy, the supersaturation required for nucleation will therefore be reached earlier in the presence of volatile elements. The liquid pressure in a solidifying alloy can be calculated from the mass conservation equation coupled with Darcys law describing the ow in a porous medium, thus giving a relationship between the liquid pressure, p , and the pore fraction, gp . As derived in [8], one gets div gp K [gradp g] + = . t t (3)
In this equation, is the specic mass of the liquid, K is the permeability of the mushy zone, is the dynamic viscosity of the liquid, g is the gravity and is the average specic mass of the solidliquid mixture in the absence of porosity. 2.2. Dissolved gas The formation of a gaseous phase in a liquid is governed by the reaction constant, Kg [15]: Kg (T ) = ag ag
g 1/n
1 ag
pg po
1/n
(4)
n is the number of moles of the gaseous element in solution in the liquid that is necessary to form one mole of this element in the gaseous phase. For a diatomic gas such as H2 , n is equal to 2 and relation (4) is referred to as Sieverts law. The activity of the element in the gaseous g phase, ag , is equal to the gas partial pressure, pg , which has been normalized in this case with a reference pressure, po . The activity of the gas in the liquid phase, ag , increases with its g concentration, w , and is given by the following relation: ag = g (wi )wg . (5)
The activity coefcient, g , does not depend on wg at very low concentration (Henrys law) but can depend on the solute element concentrations, wi (and any other gaseous element if present, as well as the pressure and temperature). Gaseous elements being usually less soluble in the solid phase, they are segregated in the remaining liquid during solidication (i.e. the gas partition coefcient, kg , is smaller than unity). Locally at the scale of the dendritic network, gaseous elements can be considered to have uniform (but different) concentrations in the liquid and solid phases, i.e. lever rule applies. This is at least the case for hydrogen in aluminium alloys if one considers the very large Fourier numbers associated with typical solidication conditions. The Fourier number
257
in a given phase is the ratio of the time available for diffusion (e.g. solidication time, ts ) and the characteristic time of diffusion over a certain distance (e.g. secondary dendrite arm spacing, 2 ). Taking 2 = 50 m and ts = 100 s, the Fourier numbers in the liquid and solid phases are given by 4D ts 3 107 100 4Ds ts 6 108 100 4 = 48 000 and Fs = 4 = 9600. (2 )2 (50 106 )2 (2 )2 (50 106 )2 The coefcients of diffusion in the liquid and solid phases, D and Ds , are taken from [7] at the melting point of pure aluminium. Two additional assumptions are made: rst, the representative volume element within which the mass balance of gas is made is a closed system and, second, the density of the solid, s , is equal to that of the liquid, . Under such assumptions, the local mass conservation equation for a gaseous element is then given by (lever rule): F = wg = wgo 1 gs (1 kg ) . (6)
gs is the volume fraction of solid and wgo is the initial gas concentration by weight1 in the liquid. In the appendix, this mass balance is extended to the case when s = . Equations (4)(6) give the following local relationship between the gas partial vapour pressure and the solid fraction: n n wgo pg = g wg Kg = g Kg . (7) po 1 gs (1 kg ) In the presence of porosity, the gas conservation equation under the assumption of a closed system (i.e. with s = and no gas escape) becomes: g p pg = wgo . (8) (kg gs + 1 gs )wg (pg ) + B T BoyleMarriots law has been used to convert the amount of gas contained in the pores at pressure, pg , and temperature, T , to standard conditions (po = 760 mmHg or 101.325 kPa, TSTP = 273.15 K). In this conversion, the factor B appears as 22 400 ccSTP 0.1 = 269.4 when wg is given in ccSTP per 100 g of metal. B= 8.314 Jmole1 K 1 Please note that unlike equation (6) in which the concentration in the liquid, wg , is given by the mass balance, this is now the equilibrium value, wg (pg ) given by Sieverts law for a diatomic gas (or by equations (4) and (5) in the more general case) which appears in this equation. Indeed, since a pore has formed with an internal pressure, pg , the concentration in the liquid must satisfy such a relationship. The extension to s = of this gas balance is also detailed in the appendix, together with the treatment of another problem that arises when the pore fraction is fairly important. Indeed, even under the assumption of s = , the formation of a pore introduces a third phase which changes the overall volume. If a Lagrangian approach is used (i.e. evolving volume encompassing the three phasesliquid, solid and pore), there is no problem. However, as most solutions of equation (3) are obtained with a xed grid (i.e. Eulerian approach), some precaution must be taken when writing equation (8) for a xed volume element (see appendix).
It is common to give weight concentration of dissolved gases in the units of ccSTP per 100 g of metal. The conversion into atomic concentration is given by 1 ccSTP /100 g = n/22 400 M /100, where n is the number of moles of the gaseous element in solution in the liquid that is necessary to form one mole of this element in the gaseous phase (e.g. n = 2 for a diatomic gas) and M is the molar mass of the metal. The conversion into weight concentration is given by 1 ccSTP /100 g = n/22 400 Mg /100, where Mg is the molar mass of the gaseous phase (e.g. Mg = 2 g for H2 ).
1
258
2.3. Analytical solution As will be seen in section 4, the mass conservation and Darcys equations (equation (3)) can be solved in a solidifying alloy only through the use of numerical methods (nite element or nite volume methods). However, a local analytical solution can be found if the liquid pressure is assumed constant and equal to the atmospheric pressure. In the case of an alloy containing only one diatomic gas (n = 2) and one volatile solute element, the pore fraction can be deduced from equations (1), (4), (5) and (8): gp = 1 gs (1 kg ) T wgo B(p + pr pi ) g K g p + pr pi po
1/2
(9)
Assuming that the pressure in the liquid and the radius of curvature of the pore are xed, this expression reveals that the partial vapour pressure, pi , of a volatile solute element increases the porosity fraction for two reasons. The square root term comes from Sieverts law, i.e. equilibrium condition between pg and wg . The introduction of pi in this term simply decreases the amount of gas stored in the solid and liquid phases when a pore forms, because the pore is partially lled with the volatile element (i.e. pg is reduced for p + pr xed). The denominator of equation (9) comes from BoyleMariots law: introducing pi reduces pg and thus increases the volume of porosity. (Of course, the radius of curvature, and thus pr , will change when the volume is increased but the trend remains.)
2.4. Volatile solute elements The vapour pressure, pi , of a solute element (i) in a solvent is given by the relationship pi = i Xi pi
pure
(T ).
(10)
i and Xi are the activity coefcient and the molar fraction of the solute element in the liquid, respectively. The mole fractions appearing in equation (10) can be converted easily into pure weight fraction, wi , while the vapour pressure of the pure element, pi (T ), is given by an Arrhenius law: Hi pure pi (T ) = Ai exp , (11) RT where Hi is the latent heat of vaporization per mole and Ai will be called the vapour pressure coefcient of the solute element. As the concentration of the solute element in the liquid, Xi , changes during solidication, equation (10) has another implicit temperature dependence (besides i which is a function of all the variables except Xi ). Assuming a ScheilGulliver law for the microsegregation of the solute element, the concentration in the liquid as a function of the volume fraction of solid is given by: Xi = Xio (1 gs )ki 1 . (12)
In the case of a binary alloy in which the volatile element is the only solute, the volume fraction of solid can be easily expressed as a function of temperature since Xi and T are related by the liquidus line (e.g. for a linearized phase diagram T = Tm + mi Xi , where Tm is the melting point of the pure solvent and mi is the slope of the liquidus line). For a multicomponent alloy, equation (12) can be written for each solute element (including the volatile one) and gs (T ) is implicitly given by T = Tm + mi Xi (gs ). Using equations (10)(12), the vapour pressure

0.11
259
101325 Pa
0.10 0.09 0.08 0.07 (MPa) 0.06 0.05 0.04 0.03 0.02 0.01 0.00
1 3 2 5 4 6
Zn Al Cu
1 : Tm,Zn = 419 C 2 : Tb,Zn = 907 C 3 : Tm,Al = 660 C 4 : Tb,Al = 2519 C 5 : Tm,Cu = 1084 C 6 : Tb,Cu = 2562 C
pi
pure
500
1000
1500 T (C)
2000
2500
3000
Figure 2. Vapour pressure of the pure elements Zn, Al and Cu as a function of temperature between melting and boiling points.
of the volatile solute element, pi , can be expressed as a function of T or gs in the whole mushy region. It is to be noted that, unlike gaseous elements which are in trace amounts in alloys, the concentration Xi (or wi ) is almost unaffected by the formation of porosity since a negligible amount of this volatile element will be found in the gaseous phase. Therefore, it is not necessary to write a solute balance for volatile elements such as that given by equation (8) for gaseous elements. In the next section, the inuence of the zinc vapour pressure on porosity formation is assessed in the case of aluminium- and copper-base alloys. Zn has a high vapour pressure and is an important solute element in the A7xxx series alloys for aluminium and in brasses.
3. Analytical results As illustrated in gure 2, the vapour pressure of pure zinc is fairly high at the solidication temperatures of aluminium alloys (4655 Pa at 660 C) and copper alloys (465 kPa at 1084.8 C). Figure 3 gives the zinc activity, aZn = Zn XZn , in aluminium and copper as a function of the zinc concentration, as calculated with ThermoCalc and the thermodynamic database SGTE Solution v.2 (also called SSOL2). For Al alloys, the activity coefcient, Zn , is equal to 2.4 and thus the corresponding vapour pressure of Zn in this alloy (equation (10)) is not sufcient to induce gas porosity at atmospheric pressure if no gas element is in solution in the alloy. In Cu-base alloys, the vapour pressure of Zn at the melting point of copper is 4.6 times the atmospheric pressure, but the activity coefcient is only 0.14. Thus the same conclusion is reached for this alloy. As hydrogen is the only gas responsible for porosity formation in both aluminium- and copper-base alloys, the next two sections deal with the coupled effects of zinc and hydrogen.
260
1.0 0.9 0.8 0.7 0.6 aZn (-) 0.5 0.4 1 0.3 0.2 0.1 0.14 0.0 0.0 0.1 0.2 0.3 1 in Al in Cu 2.4
(a)
0.5 0.6 xZn (-) 0.7 0.8 0.9 1.0
0.4
1.0 0.9 0.8 0.7 0.6 aZn (-) 0.5 0.4 0.3 0.2 0.1 0.0 0.0 in Al in Cu
(b)
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Figure 3. Zinc activity in aluminium and copper as a function of (a) zinc molar fraction and (b) zinc weight fraction, as calculated with ThermoCalc .
3.1. Aluminium alloys The rst alloy tested is an Al5 wt% Zn containing traces of hydrogen. Table 2 lists the values of the main physical properties of this alloy used in the calculations. In gure 4, the
Effect of volatile elements on porosity formation Table 2. Physical properties of Al5 wt% Zn. Phase diagram (linearized) wZno Tm m k wZn,eut geut Other data Zn AZn HZn Al5 wt% Zn 5 660 1.78 0.44 94 0(Scheil) 2.4 1.148 1010 114 200 Units wt% C K (wt%)1 wt% Pa J mole1
261
600
500 wZn,0 = 5 wt% 400 pZn (Pa)
300
200
100
0 0.0
0.1
0.2
0.3
0.4
0.5 0.6 gs (-)
0.7
0.8
0.9
1.0
Figure 4. Zinc vapour pressure in Al5 wt% Zn as a function of the volume fraction of solid (Scheil approximation).
vapour pressure of zinc calculated with the analytical solution is represented as a function of the solid fraction. It is observed that pZn rst increases with gs because of segregation of Zn in the liquid phase. Then pZn decreases during the last stage of solidication because the temperature decrease becomes preponderant (see equation (11)), while gs does not vary much. This gure shows that pZn does not exceed 550 Pa, a negligible value compared with the atmospheric pressure and the Laplace overpressure (typically, pr = 105 Pa for r = 20 m). It is concluded that zinc volatility is too low to increase microporosity fraction in aluminium alloys. Zinc could nevertheless inuence the formation of porosity in aluminium alloys containing hydrogen through the activity coefcient H (wZn ), but this was not tested in the present contribution.
262
0.020 0.018 0.016 0.014 pZn (MPa) 0.012 0.010 0.008 0.006 0.004 0.002 0.000 0.0 0.1 0.2 0.3 0.4 0.5 gs (-) 0.6
wZn,0 = 30wt% wZn,0 = 20wt% wZn,0 = 10wt%
0.7
0.8
0.90
0.95
1.00
Figure 5. Zinc vapour pressure in brasses of various Zn concentration as a function of the volume fraction of solid (Scheil approximation).
3.2. Copper alloys Whereas the zinc vapour pressure is too low to contribute appreciably to porosity formation in aluminium alloys, such is not the case for brasses as shown below. It is rst recalled that steam porosity is a major problem in copper-base alloys [4, 5]. Oxygen alone cannot give rise to porosity because of the formation of a stable cuprous oxide. But steam porosity can form in the presence of hydrogen if no oxidationdeoxidation process has been performed before solidication (see introduction). It is assumed here that such a process has been employed in order to prevent steam porosity formation. Zinc is sufciently volatile to contribute to an increase in the porosity fraction. In gure 5, the vapour pressure of zinc is represented as a function of the solid fraction for three brass compositions (10, 20 and 30 wt% Zn). It is indeed observed that the zinc vapour pressure is much higher in this case as compared with aluminium alloys. For the same reasons given before, pure the vapour pressure might rst increase (due to wi (T )) and then decrease (due to pZn (T )) near the end of solidication. But zinc also inuences the equilibrium with hydrogen through H (wZn ). The presence of 1 wt% of zinc in copper alloys decreases the hydrogen solubility by about 0.2 ccSTP /100 g under one atmosphere of hydrogen [4] (see table 3). As the hydrogen solubility, wH , is around 5.2 ccSTP /100 g in pure liquid copper at the melting point, wH is roughly 3.2 ccSTP /100 g at the liquidus point of Cu10 wt% Zn. As the coefcient KH (T ) is also unknown, it was decided to keep the factor (H KH )1 appearing in equations (4) and (5) constant and equal to 3.2 ccSTP /100 g, thus eliminating the inuences of the temperature and Zn concentration on the hydrogen solubility limit (i.e. only the pressure dependence is accounted for). This also offers the advantage of decoupling the effects of zinc and hydrogen on porosity formation. The analytical solution (equation (9)) was used for the estimation of the evolution of the porosity fraction in the Cu10 wt% Zn alloy, with and without account of the zinc vapour pressure (gure 6). As the hydrogen solubility and the gas partition coefcient
263
Table 3. Change in hydrogen solubility in copper (ccSTP /100g 1 atm H2 ) with the addition of 1 wt% of alloying element [4]. No addition Zn Sn Pb Al P Liquid 5.2 (at Tm ) 0.20 0.20 0.54 0.46 3.70 Solid 1.9 (at Tm ) 0.01 0.01 0.05
4.0 3.6 3.2 2.8 2.4 gp (%) 2.0 1.6 1.2 0.8 0.4 0.0 0.5 0.6 0.7 gs (-) 0.8 0. .0 with pZn without pZn
Figure 6. Evolution of porosity fraction in Cu10 wt% Zn as a function of the solid fraction, without and without account of the zinc vapour pressure (analytical solution (9)).
(kH = 1.9/3.2 = 0.59) are high, a fairly large initial concentration of hydrogen was required to give rise to microporosity (wHo = 3.8 ccSTP /100 g). The other data used for the results shown in gure 6 are listed in table 4. The liquid pressure was assumed constant and equal to the atmospheric pressure, while the curvature radius of the pore was assumed constant and equal to the nucleus size (r0 = 10 m)2 . It is seen in gure 6 that the vapour pressure of zinc does increase in this case the nal porosity fraction from 2.7% to 3.6%, i.e. a relative increase of 30%. As a conclusion, zinc seems to be sufciently volatile to increase noticeably the porosity fraction in brasses. The contribution of the vapour pressure of volatile solute elements has been implemented in the porosity module of the solidication software Calcosoft [14]. This module uses a
2 Assuming a constant radius of curvature while the fraction of the pores is changing is like assuming the pore is growing in between a fairly well-developed network of dendrite arms and not like a free sphere. Therefore, this case is more relevant for shrinkage porosity occurring at a late stage of solidication than for gas porosity (although the names in this case are confusing).
264 Table 4. Physical properties of Cu10 wt% Zn system. Phase diagram (linearized) wZn,o Tm m k wZn,peri gperi Other data s g kH wH AZn HZn Zn Cu10 wt% Zn 10 1 084.87 4.85 0.87 37.5 0(Scheil) 7 940 8 960 1.3 0.59 3.2 1.148 1010 114 200 0.14 Units wt% C K (wt%)1 wt% kg m3 kg m3 J m2 ccST P /100 g Pa J mole1
dynamic nite volume grid for the calculation of the liquid pressure drop in the mushy zone. The evolutions of the temperature and volume fraction of solid are calculated on a xed, coarser nite element mesh. For each volume element, the pressure, interpolated temperature and volume fraction of solid are coupled with the microsegregation of gas and the nucleationgrowth of pores in order to predict the pore fraction. All the details concerning this module can be found in [14], while the next section shows a few results dealing with the inuence of zinc on porosity formation during the continuous casting of brass alloys. 4. Simulations The present case deals with the continuous casting of a copper wire on a wheel. The rectangular wire is assumed straight and vertical with a cross section of 4 3 cm2 (see gure 7). The Cu10 wt% Zn alloy is cast at 2 cm s1 and the boundary conditions are given in gure 7. The volume fraction of solid, gs (T ), and the average density without porosity, (T ) , have been obtained under the following assumptions: microsegregation given by ScheilGulliver, linearized phase diagram and constant (but different) specic mass for the solid and liquid phases (see table 4). As shown in gure 7, the mushy zone is not exactly centred; as one of the lateral faces (#4) has a lower heat transfer coefcient, the bottom of the mushy zone is moved towards this face. Furthermore, the mushy zone is thick at the centre of the wire (i.e. low thermal gradient, G) and thin at the periphery (i.e. high G). At the same time, the solidication rate (speed of the liquidus isotherm), vT , is equal to the casting speed, vc , at the centre under stationary conditions, while near the surface it is given by the relationship, vT = vc cos , where is the angle between the casting direction and the normal to the liquidus isotherm. As is close to 90 at the periphery, vT is low. Consequently, the ratio, G/vT , which is a good indicator of the pressure drop in the mushy zone3 [15], is low at the centre (i.e. large pressure drop) and high at the surface of the casting (i.e. small pressure drop). As shown below, the predicted amount of porosity follows this trend (i.e. higher amount of porosity at the centre).
3
As demonstrated by Niyama [16], a simple analytical solution for the pressure drop, p, can be obtained from equation (3) in a one-dimensional stationary conguration with the following assumptions: no gravity, no porosity,
265
Figure 7. Initial transient until a stationary regime is reached for a continuously cast Cu10 wt% Zn small slab. Boundary conditions: 1 adiabatic, 2 = 3 convective heat transfer h, 4 convective heat transfer h/2 (h = 1000 W m2 K1 ). (a) Solid fraction; (b) liquid pressure in the absence of dissolved hydrogen; (c) porosity fraction calculated for wH,0 = 4.8 ccSTP /100 g Cu without (c1) and with (c2) account of the zinc vapour pressure.
The evolution of the zinc concentration in the liquid phase is given by the phase diagram (table 4) and equation (12). Although the concentration varies in the mushy zone, the activity coefcient, Zn , is assumed constant and given by the value at wZn 0 (Henrys law, see gure 3). The calculations of pressure drop, including the formation of pores, have been
constant viscosity and specic masses. This gives after integration p= VT G
TE
g (T )dT . K(g )
TL
The pressure drop is proportional to the ratio vT over G, the viscosity, the shrinkage factor, = (s )/ and the solidication interval, TL TE , where TL and TE are the liquidus and eutectic temperatures, respectively.
266
125000 120000

1.0
0.8 115000 110000 pl (Pa) 105000 100000 95000 0.2 90000 85000 23 0.0 28 0.4
0.6 gp (%) 24 25 time (s) 26 27
Figure 8. Evolution of the liquid pressure and porosity fraction at point A localized in gure 7 (Cu10 wt% Zn, wH,o = 4.8 ccSTP /100 g Cu) with account of the vapour pressure of Zn.
performed for the starting phase of continuous casting, until a stationary situation is reached. Three situations have been modelled. (1) Liquid pressure calculation in the absence of dissolved hydrogen. (2) Liquid pressureporosity calculation with wH,o = 4.8 ccSTP /100 g Cu and without account of the zinc vapour pressure. (3) Liquid pressureporosity fraction calculation with wH,o = 4.8 ccSTP /100 g Cu and with account of the zinc vapour pressure. The rst case is shown in gure 7(b). For the fully solid phase, the colour in this map corresponds to the last pressure in the liquid before solidication was complete. As expected, in the absence of porosity, the pressure drop in the mushy zone is the largest at the central part of the wire. It reaches nearly 100 kPa in the stationary regime. In gure 7(c), the nal porosity fraction without (c1) and with (c2) account of the zinc vapour pressure is shown. The amount of porosity is localized at the core of the wire in both cases and the maxima are about the same. However, this defect is slightly more spread across the width of the bloom when the zinc vapour pressure is accounted for. Indeed, in the presence of zinc, a lower pressure drop is required to initiate porosity when pZn = 0. Figure 8 gives the evolution of the liquid pressure and porosity fraction at point A in gure 7(c). It is observed that, before cavitation (i.e. t < 27.5 s), the liquid pressure rst increases due to the metallostatic head (gravity is parallel to the casting direction) and then rapidly decreases near the bottom of the mushy zone. When cavitation occurs, the liquid pressure increases again. Nevertheless the pore fraction increases monotonically because of the gas microsegregation. This last pressure increase might be particular to alloys in which gas solubility is very high: when pores are formed, the concentration of gas in the liquid is very high. As the solid also rejects a large amount of gas, this gas is transformed into a gaseous phase. If the pore growth rate, gp , is higher than the solidication shrinkage rate, / , some liquid is expulsed from the mushy zone. According to Darcys law, the liquid pressure is thus constrained to increase.

12 11 10 9 8 7 (%) 6 5 4 3 2 1 0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 gs (-) 0.7 0.8 0.9 1.0
267
12 11
(a)
shrinkage
Al
(b)
Cu
10 9 8 7 (%) 6 shrinkage 5 4 3 2 1 0 0.5 0.6 0.7 gs (-)

80
porosity fraction (wH,0=0.62 ccSTP/100g)
porosity fraction (wH,0=4.16 ccSTP/100g) porosity fraction (wH,0=3.85 ccSTP/100g)
0.8
0.
.0
40
(c)
35 shrinkage 30 25 (%)
Fe
70 60 50 (%) 40 30 20 10 0 0.5
(d)
shrinkage
Fe
porosity fraction (wH,0=18.8 ccSTP/100g)
porosity fraction (wN,0=27.7 ccSTP/100g)
20 15 10 5 0 0.5
0.6
0.7 gs (-)
0.8
0.9
1.0
0.6
0.7 gs (-)
0.8
0.9
1.0
Figure 9. Solidication shrinkage and porosity fraction as a function of the fraction of solid for various metal/gas systems (analytical solution (9) at atmospheric pressure).
This pressure increase might also happen in the case of nitrogen or hydrogen in ironbase alloys, as shown in gure 9. In this gure, the evolution of shrinkage, (gs ) = ( (gs ) )/ , and of the pore fraction, gp , as obtained with the analytical solution (equation (9)), are represented as a function of the volume fraction of solid. As can be seen, the slope of the pore evolution, gp /gs , is larger than that of the relative density, /gs , in all cases, except for Al alloys. From equation (9), high values of , T and gas solubility, wg = (g Kg )1 , give high values of the gp (gs ) slope. As a consequence, whatever the initial gas concentration, liquid will always be rejected from the mushy zone in copper- and iron-base alloys when pores start to form, while this never occurs in Al alloys. This phenomenon seems to be correctly reproduced by the software calculating porosity fraction, once the equations were modied to account for this phenomenon (see appendix). This pressure increase had some impact on the computation time required to calculate the results shown in gures 7 and 8. As gp /gs is large, so is the derivative of gp with respect to time. Therefore, according to the mass conservation equation (equation (3)), the
268
time derivative of the liquid pressure is also large as soon as porosity forms, and small time steps are required to capture these variations (typically 0.01 s in this case). Furthermore, the solid fraction and the liquid pressure strongly vary with position in a relatively thin mushy zone: ne nite element and nite volume meshes have been employed to correctly solve the gs - and p -elds for this simple geometry (180000 FEM tetrahedrons, 1000000 cubic FVM elements). These calculations took 10 h CPU time on a biprocessor Intel Xeon@2.8 Ghz with 2 GB RAM. As the solubility of hydrogen and nitrogen is even higher in iron-base alloys (see table 1), the gp (gs ) curve is steeper and thus more abrupt will be the variations of the liquid pressure. It is interesting to note in this case that nearly 40% H2 and 55% N2 porosity form during solidication of the last 10% of liquid! This might look surprising. However, it should be kept in mind that the analytical solution (equation (9)) is based on an Eulerian representation. Assuming s = for the sake of simplicity, the formation of gp Vo porosity during the solidication of 0.1 Vo remaining liquid simply means that the nal Lagrangian volume Vt = (1 + gp )Vo . In other words, liquid is expulsed out of the initial volume during the last stage of solidication. These very large fractions of porosity are of course associated with the large initial amounts of gas necessary to saturate the melt near the end of solidication. Decreasing this initial amount (or increasing the supersaturation necessary to reach nucleation) does not change the slope of the curves shown in gure 9. For example in gure 9(b), decreasing wH,o to 3.85 ccSTP /100 g Cu results in a porosity fraction of 3.5%: this is less than the overall shrinkage of the metal (about 11%), but during pore formation, the slope gp is larger than that of the specic mass, i.e. ow is still rejected out of the volume element. This makes the Eulerian-base calculation of porosity very sensitive to the time step, and some time stepping renement might be necessary to obtain solutions under reasonable times. At the same time, as the slope gp (gs ) is unchanged, the formation of only 1 or 2% porosity in Cu- or Fe-base alloys must occur during the very last stage of solidication, i.e. when (1 gs ) < 0.01. At this stage, the interdendritic liquid is in the form of liquid droplets rather than continuous lms, in which case Darcys equation no longer holds.
5. Conclusion The vapour pressure of volatile solute elements has been incorporated in equations governing the nucleation and growth of pores. A simple thermodynamic analysis has shown that the partial vapour pressure of zinc in solution in aluminium alloys is too low at the melting point to increase signicantly the pore volume fraction in these alloys. But this vapour pressure is sufciently high to increase the porosity fraction in brasses. An analytical solution in which the liquid pressure drop in the mushy zone is not taken into account reveals that this increase is signicant for a copper-base alloy containing 10 wt% of zinc. Finally, the contribution of volatile solute elements has been implemented in a software predicting porosity formation. It has been shown that the accounting of the vapour pressure of zinc in the continuous casting of brass expands the region where porosity forms. Besides the effect of Zn, the case of brass has revealed that the pressure in the interdendritic liquid increases again once pores start to form. This increase has been analysed with the analytical solution: it has been shown that a high density and melting point of the metal, as well as a high solubility of the gas, favour such situations. When the pore fraction increases faster than the shrinkage rate, liquid is expulsed and the pressure increases. Such situations require a modication in the mass balance when an Eulerian framework is adopted as shown in the appendix.
269
Figure A1. Representation of an elementary volume of liquid V0 at t = 0, containing no moles of alloy, and its repartition between solid, liquid and gas at t. This volume does not exchange matter with the exterior (its mass mo , its number of moles of alloy no , and its number of moles of gas are constant).
Acknowledgments The authors would like to thank the Commission for Innovation and Technology, CTI, Bern (grant 6167.1 KTS), and the industries Alcan (CH), Alcan (FR), HydroAluminium (DE), General Motors (USA) and Union Mini` re (BE) for their nancial support. They also thank e Jean-Luc Desbiolles for his help in the implementation of the model into CalcoSoft . Appendix In order to model porosity, a conservation equation has to be written for the gas present in the solid, liquid and gas phases. This is done below for dihydrogen, considering an elementary volume and assuming that this volume does not exchange matter with the exterior (gure A1). One has in this case gpL Vt no XHo = (no ns )XH + ns kH XH + 2 pH 2 . (A1) RT no is the nominal number of moles of alloy in a given volume, V0 , which is entirely liquid initially. XHo and XH are the nominal molar concentration and the solubility of hydrogen in the liquid phase, respectively. ns is the number of moles of the solid phase and kH is the partition coefcient of hydrogen between the liquid and solid phases. gpL is the volume fraction of the gaseous phase in a Lagrange representation. Vt is the sum of the volumes of the liquid, the solid and the gaseous phases at the current time. T is the temperature, and R is the perfect gas constant. The factor 2 in the last term comes from the fact that the gas in the dense phases is monoatomic, while it is diatomic in the pore. Equation (A1) has to be transformed in order for the volume fraction of solid to appear. The three relationships below give the dependence between no , ns , (no ns ) and the nominal specic mass of the liquid alloy, o , and the solid and liquid specic masses, s and , respectively:
o
o no Malloy
V0
s =
s ns Malloy
Vs
(no ns )Malloy (Vt Vs Vp )
(A2)
o s Malloy , Malloy and Malloy are the molar masses of the alloy in the initial state, and of the solid and the liquid part at the current time, respectively. The mass conservation relation can be
270
written (neglecting the specic mass of the pore) as: o V0 = ( (1 gsL gpL ) + s gsL )Vt , (A3)
where gsL and gpL are the volume fractions of the solid and gas phases, respectively, in a Lagrangian representation. Combining equations (A1), (A2) and (A3) gives the following conservation relation: ( (1 gsL gpL ) + s gsL ) (1 gsL gpL ) s gsL gpL XHo = X H + s kH X H + 2 pH . o Malloy Malloy RT 2 Malloy (A4) The calculation of the solid fraction evolution is often based on an Eulerian approach with a xed mesh. In equation (A4), one has to replace the Lagrangian fractions of solid or porosity by the Eulerian ones. Assuming that the volume variation of a quantity is restricted to the volume variation of the porosity part, Vt is equal to V0 + Vp . If the solid and the gaseous parts are conned in the xed volume, V0 , gsE = Vs /V0 and gpE = Vp /V0 . As gsL = Vs /Vt and gpL = Vp /Vt one can easily demonstrate that: gsL = gsE 1 + gpE and gpL = gpE . 1 + gpE (A5)
In such a relationship, the liquid and solid specic masses are considered equal and constant. Finally, relation (A4) becomes: gpE ( (1 gsE ) + s gsE ) (1 gsE ) s gsE pH , XHo = XH + s kH XH + 2 o Malloy Malloy RT 2 Malloy (A6)
Multiplying this equation by the molar mass of the (monoatomic) gas, one obtains the mass balance of gas for the weight concentrations: gpE (A7) pH , wHo = (1 gsE )wH + s gsE kH wH + Mg RT 2 where Mg is the molar mass of the diatomic gas and is the average specic mass of the mixture of solid and liquid phases for a pore-free material. This equation is identical to equation (8) under the assumption s = = . In the absence of porosity, equation (A7) can be transformed into the following relationship: wH = wHo , (1 gsE ) + s gsE kH (A8)
which is again identical to equation (6) under the hypothesis of constant and equal densities of the dense phases. Making a lever rule, adapted to a three-phase problem, for the mass conservation of gases is appropriate if one considers the Fourier numbers involved (see section 2.2). However, the assumption of a closed system when gp = 0 is more questionable. Indeed, when the shrinkage term is not exactly balanced by the formation of porosity, there is a ow of interdendritic liquid (see equation (3)). If there is a ow of liquid, v , the solute conservation in an Eulerian framework is given by: gpE (A9) g E wH + s gsE kH wH + B p H2 = 0 t T provided the solid phase and the pores are xed and diffusion of gas at the macroscopic scale is negligible. div( g E v wH ) +
271
References
` ` [1] Darcy H 1856 Les fontaines Publiques: exposition et applications des principes a suivre et des formules a employer dans les questions de distribution deauouvrage termin par un appendice relatif aux fournitures e ` deau de plusieurs villes, au ltrage des eaux et a la fabrication des tuyaux en fonte, de plomb, de t le et de o bitume Dalmont [2] Campbell J 2003 Castings ed O Butterworth-Heinemann (Oxford: Elsevier) [3] 1992 Smithells Metals Reference Book 7th edn Naional Bureau of Standards, Gaithersburg, USA, Card number 7771709 [4] INCRA: The International Copper Research Association 1977 Series on the Metallurgy of Copper Library of Congress [5] 1998 ASM Handbook: Casting (ASM Int. Metals Park, Ohio, USA) 4th edn, ed D Stefanescu et al, vol 15 Library of Congress [6] Charbonnier J 1991 Gaz dans les alliages daluminium de fonderie, Techniques de lIng nieur, trait Mat riaux e e e m talliques, M218 e [7] Eichenauer W and Markopoulos J Z 1974 Z. Metallk. 65 649 [8] Kubo K and Pehlke R D 1985 Met. Trans. B 16 35966 [9] Poirier D R, Yeum K and Maples A L 1987 Met. Trans. A 18 197987 [10] Zhu J D and Ohnaka I 1991 Modeling of casting, welding and advanced solidication processes V ed M Rappaz et al (Warrendale: TMS) [11] Lee P D, Chirazi A and See D 2001 J. Light Met. 1 1530 [12] Sabau A S and Viswanathan S 2002 Met. Mater. Trans. B 33 24355 [13] Sung P K, Poirier D R and Felicelli S D 2002 Modelling Simul. Mater. Sci. Eng. 10 55168 [14] Pequet Ch, Gremaud M and Rappaz M 2002 Modelling of microporosity, macroporosity and pipe-phrinkage formation during the solidication of alloys using a mushy-zone renement method: applications to aluminium alloys Met. Trans. A 33 2095106 [15] Darken L S, Gurry R W and Bever M B 1953 Physical Chemistry of Metals, (New York: Mc Graw Hill) [16] Niyama E, Uchida T, Morikawa M and Saito S 1982 A method of shrinkage prediction and its application to steel casting practice AFS Int. Cast Metals J. 7 5263

APM Papers

Transféré par

Informations du document

Description originale:

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

APM Papers

Transféré par

Droits d'auteur :

Formats disponibles

Modeling of Microporosity, Macroporosity, and PipeShrinkage Formation during the Solidification of Alloys Using a Mushy-Zone Refinement Method: Applications

gs(t))3 2(t) 2 gs(t)2 180

M(t)(t ln cl(t) cl(t)

[H]s sgs gppp T

3.284 and log10 fH

METALLURGICAL AND MATERIALS TRANSACTIONS A

max r0; min until gs (1 ge)

8.3144 J mole 1 C ccSTPH2 K s2 m 269 100 g

2106VOLUME 33A, JULY 2002

METALLURGICAL AND MATERIALS TRANSACTIONS A

A POROSITY MODEL FOR MULTI-GAS SYSTEMS IN MULTI-COMPONENT ALLOYS

Eq. (2) can be transformed into the following relation:

X lAo = l (1- g s ) X l* + s g s k A X l* + M o A A alloy

0.6 0.5 rmax/2 (-)

0.4 0.3 0.2

shrinkage no shrinkage no shrinkage 2= 25 m 2= 50 m 2= 25 m

(a) 2 = 50 m, with shrinkage.

(b) 2 = 25 m, with shrinkage.

1.0 0.5 0.0 40 35 30 25

1.0 0.5 0.0 40 35 30 25 35 30 25 20

(c) 2 = 50 m, without shrinkage .

(d) 2 = 25 m, without shrinkage.

Symposium on Simulation of Aluminum Shape Casting Processing Edited by Qigui Wang

Finite Element Mesh

Closed liquid pocket

for a gas (i) of composition AniABniB

eA [wt%-1] 0.04 -0.006 -0.1 -0.03 0.004

500 wZn,0 = 5 wt% 400 pZn (Pa)

16000 14000 12000

6000 4000 2000

0.5 0.6 gs (-)

INSTITUTE OF PHYSICS PUBLISHING

MODELLING AND SIMULATION IN MATERIALS SCIENCE AND ENGINEERING doi:10.1088/0965-0393/14/2/009

Modelling Simul. Mater. Sci. Eng. 14 (2006) 253271

Effect of volatile elements on porosity formation in solidifying alloys

G Couturier and M Rappaz

Effect of volatile elements on porosity formation

G Couturier and M Rappaz

Effect of volatile elements on porosity formation

G Couturier and M Rappaz

Effect of volatile elements on porosity formation

G Couturier and M Rappaz

500 wZn,0 = 5 wt% 400 pZn (Pa)

0.5 0.6 gs (-)

G Couturier and M Rappaz

Effect of volatile elements on porosity formation

G Couturier and M Rappaz

Effect of volatile elements on porosity formation

G Couturier and M Rappaz

0.6 gp (%) 24 25 time (s) 26 27

Effect of volatile elements on porosity formation

10 9 8 7 (%) 6 shrinkage 5 4 3 2 1 0 0.5 0.6 0.7 gs (-)

porosity fraction (wH,0=0.62 ccSTP/100g)

porosity fraction (wH,0=4.16 ccSTP/100g) porosity fraction (wH,0=3.85 ccSTP/100g)

porosity fraction (wH,0=18.8 ccSTP/100g)

porosity fraction (wN,0=27.7 ccSTP/100g)

G Couturier and M Rappaz

Effect of volatile elements on porosity formation

(no ns )Malloy (Vt Vs Vp )

G Couturier and M Rappaz

Effect of volatile elements on porosity formation

Vous aimerez peut-être aussi