METHOD FOR MODIFYING THE GRAIN SIZE OF CAST PRODUCTS OBTAINED FROM COMMERCIAL MELTS

FIELD OF THE INVENTION

The invention relates to casting. The invention deals with a method of grain size regulation for ensuring a possibility to obtain castings of any configuration and from any commercial melts characterized by unified specified structure in any cross-section which is equal to obtaining of castings without anisotropy of service properties.

BACKGROUND OF THE INVENTION

Actually in production of castings from homogeneous melts and regulation of their structure whether methods of size reduction, for example by ultrasonic traveling fields or methods of growing of large structures or monostructures are used (at large).

If castings are to be produced from heterogeneous melts various methods of modifying and processing with ultrasonic, electromagnetic fields are applied. The methods listed above are the only ones for ensuring forming of castings of desired structure and, therefore, of desired service properties, all other conditions being equal.

A well-known method of monostructure generation is based on creation of supercoolings in the melt corresponding (approximately) to the maximal linear rate of crystal growth (Csochralsky J.Z., Physik. Chem. 1917, Bd. 92, S. 219; Chalmers B. Principles of Solidification, 1968, p.20).

This well-known method was taken as a prototype of the claimed solution.

Efficiency of the abovementioned methods largely depends on the melt type, casting volume, conditions of heat takeoff (temperature decrease rate, direction of cooling). Thereat due to practical impossibility to identify conditions of heat takeoff in the melt circumference and from its central zone the casting, naturally, is forming with anisotropy of the grain size.

SUMMARY OF THE INVENTION

This invention is intended for solution of the engineering problem of modifying of the grain size of castings due to directed melt crystallization in the gravitational field of a centrifuge under the conditions of uniform volume cooling. Obtained technical result consists in a possibility to produce castings of any configuration from any commercial melts characterized by unified specified structure in any cross-section which is equal to obtaining of castings without anisotropy of service properties.

The said technical result is obtained owing to the fact that the method of modifying the grain size of castings from commercial melts consisting in forming of directed crystallization in the course of melt cooling implies carrying out of uniform volume cooling of melt with a rate of (2-10)°C/sec in a gravitational field created by a centrifuge with the gravitation coefficient growth from 10 to 1000 during crystallization.

These features are essential and interconnected forming a fixed aggregate of features sufficient for obtaining desired technical result.

DESCRIPTION OF FIGURES

This invention is explained with the following figures:

• Fig. 1. Scheme of normalized Tamman dependences
• Fig. 2. Plot of casting grain size against created supercooling
• Fig. 3. Pattern of initial potential profile
• Fig. 4. Pattern of disturbed potential profile
• Fig. 5. Plot of casting grain size against gravitation coefficient
• Fig. 6. Ceramic shell mould
• Fig. 7. Centrifuge for production of castings
• Fig. 8. Crystallizer pan KP-1000 intended for industrial application
• Fig. 9. The same as Fig. 8, viewed from the other side
• Fig. 10. Ingot produced with crystallizer pan KP-1000
• Fig. 11. Annular casting form
• Fig. 12. Plot of grain size modification in the samples from A99
• Fig. 13. plot of grain size modification in the samples from AL4

DESCRIPTION OF AN EMBODIMENT OF THE INVENTION

A possibility to obtain castings characterized by structures in a range from monometallic to ultradispersed from any commercial melt by centrifugal method with varying intensity of gravitational field of the centrifuge was theoretically predicted and experimentally proved.

Scientific significance of this phenomenon consists in obtaining of crucially new information on the metal melt behavior in created gravitational fields during crystallization. Of special interest is the process of nucleating and growth of crystal phase under conditions of uniform cooling. In contradistinction from accumulated information on monotonous reduction of grain size against growth of gravitation coefficient data on apparently non-linear character of this dependence are obtained and substantiated.

Practical value of the invention consists in obtaining of universal method of regulation of casting structure in a range from monometallic to ultradispersed without use of modifying processes and vibrating machines.

Theoretical proof is as follows.

Analyzing normalized Tamman dependences (Fig. 1) it can be supposed that forming of supercooling equal to $\mathrm{\Delta }{T}_{\mathit{opt}}^n$ in the melt would inevitably result in obtaining of castings characterized by maximally refined structure owing to the forced growth of nucleus number (even in a homogeneous system) in a limited melt volume without introduction of additional modifiers.

Generating of supercoolings $\mathrm{\Delta }{T}_{\mathit{opt}}^v$ corresponding to would lead to forced growth of the crystal linear dimensions which is the case for the systems of both homogeneous and heterogeneous types.

Generalizing the last conclusions one can synthesize an algorithm of regulation of the grain size of castings with supercooling being an operational parameter. In this case dependence of the casting grain size (produced by any method) against supercooling value is as shown on Fig. 2.

The desired suipercooling in the melt can be obtained (apart from conventional cooling methods) can be obtained by creating pressure field of any profile in the melt. Taking into account that melt crystallization temperature is defined as follows: $${\mathit{T}}_{\mathit{cr}}^{\mathit{Px}}={\mathit{T}}_{\mathit{cr}}^{\mathit{Po}}+\mathrm{\alpha }\mathit{Px}$$

where

${\mathit{T}}_{\mathit{cr}}^{\mathit{Px}},{\mathit{T}}_{\mathit{cr}}^{\mathit{Po}}$ - crystallization temperatures for the pressure Px and normal pressure Po;

α - factor of the adopted model

the dependence of supercooling value ΔT against Px can be obtained supposing that the melt is thermostabilized at the level of crystallization temperature ${\mathit{T}}_{\mathit{cr}}^{\mathit{Po}}:$ $$\mathrm{\Delta }T={\mathit{T}}_{\mathit{cr}}^{\mathit{Px}}-{\mathit{T}}_{\mathit{cr}}^{\mathit{Po}}=\mathrm{\alpha }\mathit{Px}$$

Values of α for some metals:

aluminum:

6.4·10^-2oCm²/Mn

iron:

3.0·10^-2oCm²/Mn

copper:

4.2·10^-2oCm²/Mn

nickel:

3.7·10^-2oCm²/Mn

This means that for creation of considerable supercooling (adequate values of metastability intervals) a pressure of some tens of Nm/m² must be created in the melt, which in some cases, for example in production of shaped castings in ceramic moulds, is impossible.

Another, a somewhat differing method of the use of pressure to force the growth of solid phase whether due to increase of nucleation rate n*(ΔT) or due to increase of linear growth rate V*(ΔT) of the crystals. Indeed, using dependence of the melt supercooling degree against cooling rate the entire melt can be supercooled to the level of, for example, $\mathrm{\Delta }{T}_M^n$ and than, having created in the melt a pressure $$\mathit{Px}1=\frac{1}{\mathrm{\alpha }}\left(\mathrm{\Delta }{T}_{\mathit{opt}}^n-\mathrm{\Delta }{T}_M^n\right)$$

desired result can be obtained as far as forming of the casting structure is concerned.

A similar result can be obtained when structure coarsening is necessary: $$\mathit{Px}2=\frac{1}{\mathrm{\alpha }}\left(\mathrm{\Delta }{T}_{\mathit{opt}}^v-\mathrm{\Delta }{T}_M^v\right)$$

In this case for stable regulation of casting structure much less pressure Px1, Px2 are necessary as: $$\mathrm{\Delta }{T}_M^v>\mathrm{\Delta }{T}_{\mathit{opt}}^v-T_M^v$$ $$\mathrm{\Delta }{T}_M^n>\mathrm{\Delta }{T}_{\mathit{opt}}^n-\mathrm{\Delta }{T}_M^n$$

Taking into account that the value, for example, of the rate of change dn*/d(ΔT) in the range $\mathrm{\Delta }{T}_M^n÷\mathrm{\Delta }{T}_{\mathit{opt}}^n$ exceeds 10¹⁰, it isn't difficult to calculate required pressure necessary to increase nucleation rate by specified times Z: $$P=\frac{\left(Z-1\right)n\left(\mathrm{\Delta }{T}_M^n\right)}{\mathit{dn}*/d\left(\mathrm{\Delta }T\right)}$$

which eventually will define the casting grain size.

Under the conditions of directed melt crystallization and their volume unified cooling in non-linear gravitation fields created by centrifuges, to the aforesaid mechanism of influence on n* and V*of the pressure created due to centrifugal forces Fc: $$\mathrm{Px}\mathrm{\approx }\mathrm{0.05}\mathrm{\gamma Xkg}$$

where

γ - melt specific weight

X - distance from the melt to roll axis

Kg - gravitation coefficient distortion of initial potential profile by Fc and as a consequence - creation of substantial conditions for forced growth of solid phase.

Fig. 3 shows initial potential profile.

An atom located in a potential hole due to symmetric potential profile can move whether in +X or in -X direction. Analysis of Tamman dependencies $$n\left(\mathrm{\Delta }T\right)=K\exp \left\{\frac{1}{\mathrm{\Delta }T-\mathrm{To}}\left[\frac{\mathit{Uo}}{R}+\frac{B{\mathrm{\zeta }}^3}{{\left(\mathrm{\Delta }T\right)}^2}\right]\right\}$$

where

Uo - heigth of potential profile

R - gas constant

B - substance constant

σ - surface strain

To - melting temperature leads to a conclusion that identical increase of n(ΔT) can be obtained whether by increase of ΔT or by decrease of Uo.

The centrifugal force Fc equal to: $$\mathrm{Fc}\mathrm{=}\mathrm{m}\mathrm{1}\mathrm{\omega }\mathrm{2}\mathrm{x}\mathrm{=}\mathrm{m}\mathrm{1}\mathrm{gKg}$$

where

m1 - atomic weight

ω - angular rate of the melt rotation

with potential energy $$\mathrm{E}\left(\mathrm{x}\right)\mathrm{=}\mathrm{-}\mathrm{XFc}\mathrm{=}\mathrm{-}\mathrm{m}\mathrm{1}\mathrm{\omega }\mathrm{2}\mathrm{x}\mathrm{2}$$

distorts potential profile (Fig. 4). Indeed, for movement of an atom from the hole Xi to the hole Xi+1 activation energy is necessary: $$U_{+}^0=U^0-\frac{\mathrm{\delta }}{2}\mathit{Fc}=U^0-\frac{\mathrm{\delta }}{2}m1\mathrm{\omega }2x1$$

For the alternative movement in the -X direction this energy is equal to: $$U_{-}^0=U^0+\frac{\mathrm{\delta }}{2}\mathit{Fc}=U^0+\frac{\mathrm{\delta }}{2}m1\mathrm{\omega }2x1$$

Resulting energy change is equal to: $$\mathrm{\Delta }{U}_{\mathrm{\Sigma }}^0=\mathrm{\delta }\mathit{Fc}=\mathrm{\delta }m1\mathrm{\omega }2x1$$

In case of absence of Fc (Fig. 3) the flow of the melt atoms to solid phase can be defined as follows: $$\mathrm{\Pi }1=K1\exp \left(-\frac{U^0}{\mathit{kT}}\right)$$

where K1=nkνρ

nk - number of atoms contacting with a nucleus;

ν - frequency of atom oscillations;

ρ - probability of the atom movement towards attaching.

Evidently, the corresponding flow of holes moves in the -X direction (here and henceforth not mass transfer but single atom movements to the phase boundary is implied).

The flow of atoms in the +X direction (without derivation) is equal to: $$\mathrm{\Pi }2=\frac{\mathrm{\delta }\mathit{nSm}1\mathrm{\omega }2x}{6T0\mathit{KT}}\exp \left(-\frac{U^0}{\mathit{kT}}\right)$$

where S - surface of atom attaching; $$\mathrm{T}\mathrm{=}\frac{\mathrm{1}}{\mathrm{\nu }}$$

Joint analysis of the expressions (14) and (15) demonstrates, that their ratio is as follows: $$Z*=\frac{\mathrm{\delta }\mathit{nSm}1\mathrm{\omega }2x}{6\mathit{KTnkP}}=0.7\cdot {10}^{45}\frac{\mathrm{\delta }\mathit{Sm}1}{T}\mathit{kg}$$

In real condition of casting production at Kg>>1 Z>>1 even without consideration of the corresponding supercoolings ΔT. This circumstance accounts for a possibility to form solid phase in non-linear gravitation force fields.

In general, adequate for intensity influence of Kg reduced (converted) to ΔT can be estimated as follows: $$\mathrm{\Delta }T=\sqrt{\frac{B{\mathrm{\sigma }}^3}{\mathit{Ten}\mathrm{\mu }}}$$ $$\mathrm{\mu }=\frac{K}{c\frac{m1}{T}\mathit{kg}+K\exp \left\{\frac{{B\left[\mathrm{\sigma }\left(1.11\cdot {10}^{-7}\mathrm{\gamma }\mathit{kg}-\right)\right]}^3}{T{\left(\mathit{Tc}+\mathrm{\epsilon }\mathit{kg}\right)}^2}\right\}}$$ $$c=\frac{\mathrm{\delta }\mathit{nSm}1g}{6T0\mathit{KT}}\epsilon =1.11\cdot {\mathrm{10}}^{\mathrm{-}\mathrm{4}}\mathrm{\gamma \alpha }$$

where Tc - supercooling value obtained naturally.

Thus some adequacy of influence on the process of forming of solid phase from the melt in external irregular gravitational field created, for example, by a centrifuge.

In this connection a dependence is evident (similar to that shown on Fig. 2) of the casting grain size against gravitational coefficient starting from 10 (established minimal threshold value below which the nucleation pattern repeats actually existing traditional crystallization processes) (Fig. 5). Grain size d (obtained statically) slightly decreases with Kg growth which is explained by fractures of dendritic crystals and modifying of the melt with fragments. Further growth of Kg up to 1000 leads to crystallization in the zone of maximal dependence of n*(ΔT) or V*(Kg), that is in the zone of drastic increase of the grain size up to the monostructure. After the gravitational coefficient exceeds 1000, grain modification pattern during crystallization doesn't reveal substantial changes.

Further on, with growth of Kg crystallization is occurring in the zone of maximal n*(ΔT) or n*(Kg), which results in drastic increase of the crystallization center number and hence to reduction of the grain size at $\mathrm{Kg}=\mathrm{Kc}\left(\mathrm{Kc}\sim \mathrm{\Delta }{T}_{\mathit{opt}}^n\right)\mathrm{.}$

The grain size after reaching minimal value dc slightly increases which, apparently, can be explained by concomitant pressure rise in the melt, reduction of diffusion and growth of viscosity.

Evidently, the discussed curve (Fig. 5) is possible only when the gravity force field of required intensity and quasithermostatting of the melt are the driving forces conditioning the start of crystallization.

Researches were carried out with the following metals and alloys: A99, VAL5, VAL8, AL4, R9, ZHS6K, RS-A10Mg.

In the course of the experiments the range from 10 to 1000 was studied in detail at the rates of concomitant cooling of the working melt of (2-10)°C/s.

Thereat both ultradisperced and monostructures were obtained. The monostructures were obtained without nucleators with the monocrystal at the growth rate of (0.5-1) mm/s having dislocation density no more than 5·10⁶ sm^-2.

The scheme and photographs o the crystallizer pan used for the experiments are shown on Fig. 8 and Fig. 9. The Popov crystallizer pan (KP-1000) intended for industrial application with maximal ingot diameter of 1000 mm is shown on Fig. 10. Fig. 12 and Fig 13 show technological curves of the grain size for A99 and AL4.

Thus theoretical suppositions were proved in practice.

On a basis of this invention scientific problem of dynamic stimulation of nucleation and crystal phase growth in non-linear steady force fields can be completely solved. In addition, a possibility is presented to develop analytic determined algorithms of reliably regulation of the casting structure forming.

Practical significance of the invention in a possibility to obtain castings of any configuration and from any commercial melts characterized by unified specified structure in any cross-section which is equal to obtaining of castings without anisotropy of service properties.

Furthermore some second order effects are to be noted:

• elimination of pores;
• elimination of cast seams;
• considerable reduction of gas content (by more than10 times);
• increase of casting density.