US3416977A - Cryogenic cooling - Google Patents

Description

Dec. 17, 1968 R. H. REIN 3,416,977

CRYOGENIC COOLING Filed April 1, 1966 3 Sheets-Sheet 1 l-L/QU/D ,w moss/v 2-5 /c //v LIQUID NITROGEN 3-10 w: //v LIQUID NITROGEN +100 4-30 ICE m L/QU/D NITROGEN 5-507. ICE //v 1.10010 NITROGEN. 6-60% ICE m LIQUID NITROGEN 7-55 ICE nv uowo NITROGEN 8-WA7'ER +500 LI- 0 L; I a: E 300 DJ 0. 2 Lu 0 20 -40 so so I00 I20 I40 TIME, sscouos INVENTOR RICHARD H. REIN ATTORNEY 3,416,977 CRYOGENIC COOLING tichard H. Rein, North Tonawanda, N.Y.', assignor to Union Carbide Corporation, a corporation of New York Filed Apr. 1, 1966, Ser. No. 539,429 11 Clalms. (Cl. 148-206) ABSTRACT OF THE DISCLOSURE A method for controllably extracting heat from a ma terial over a predetermined temperature range by contacting said material with a fluid dispersion comprising at least one cryogenic fluid which is at its boiling point, and at least one finely divided solid additive, charac- Patented Dec. 17, 1968 It is known that in heat treating various metal and alloys rapid cooling is critical within a particular temperature range, depending on the particular metallic.

composition, if the metal is to achieve its optimum properties. In other words, a cooling rate particular for each alloy is required, which in turn requires a tailor-made terized by having its melting point and at least a portion of its stable liquid phase within the temperature range formed by the boiling point of the cryogenic fluid and the temperature of the material from which'heat is to be extracted. t i

Thisapplication relatesto a method for improving the heat transfer characteristics of cryogenic fluids, to the fluids thus improved. and to a method for utilizing such fluids toextract heat rapidly and controllably from materials.

Although th methods, compositions and conceptsof this invention are applicable to heat transfer in general, rapid and controllable cooling is of particular significance in the quenching of metals. For this reason the invention will be described and illustrated as it relates to the quenching of metals. "Itis to be understood, however, that the invention is not limited to this particular, preferred embodiment. r

A large variety of quenching media have been developed over theyears in an attempt to obtain an optimum balance of properties for various metals. Ordinarily, heat treating and quenching processes are carried out in such manner as to achieve optimum physical properties, 'i.e.,

yield strength, ultimate strength, ductility and resistance to corrosion, with a minimum of distortion and cracking.

it has long been known that certain steels, for example, can be hardened by first heating them to a high temperature and then cooling them with blastsof air or Iquenching them with oil or water. it is also known that quenching of heat treated aluminum parts in cold water causes them to warp. but that this problem can, to some extent at least, be alleviated by quenching the hot parts in a heavy mist of water droplets suspended in air.'Such a process is handicapped'considerably by the fact that the internal sections, particularly of complex fabricated parts, may not be contacted by the mist, and consequently warpage occurs because heat has not been removed from the part in a sufficiently rapid and uniform manner. It is known also that residual stress in fabricated parts of aluminum and magnesium alloys may be relieved by solution heat treating them and then quenching-first in a conventional quenching medium, such as water, then in liquid nitrogen to cool the part to -l00' F. or lower, and thereafter rapidly reheating it to room temperature with steam.

More recently, Dullberg in US. Patent No. 3,185,600 discloses a process for quenching hot sheet metal parts, at their solution heat treatment temperature, directly into a cryogenic fluid, such as liquid nitrogen, so as to reduce the temperature of the part as rapidly as possible below -50 F. The Dullberg process is severely limited in its area of application because the cooling rate of the heat treated part is not adjustable, and therefore proper only for a limited number of alloys, and then only for thin stock, i.e., those characterized as sheet metal parts.

quenching medium having a controllable cooling rate over a predetermined temperature range.

The discoveries mentioned above and others have solved some of the problems relating to quenching of metals. Certain problems however, still remain. Although quenching with water causes very rapid removal of heat, it also tends to causetwarpage or cracking of quenched objects because of uneven removal of heat from thinner and thicker sections of an object. Quenching of fabricated parts in liq'uid'nitrogen, though useful in eliminating warpage, is limited in its application to thin or sheet' metal parts, and then only to specified alloys for which the quenching rate produced by liquid nitrogen happens to be proper. In other words, water quenches too fast, liquid nitrogen quenches too slowly, and there is'no quenching oil known which maximizes strength in a material without distorting it. v g

It is an object of this invention to improve the heat transfer characteristics of cryogenic fluids.

It is another object of this invention to provide a tailormade cryogenic quenching medium which can be adjusted to provide a predetermined cooling rate over a prede termined temperature range for any particular material to be cooled.

It is yet, anotherv object of this invention to provide 7 a method for extracting heat rapidly andcontrollably from a material in such manner that the material will not be distorted, nor substantially reduced in strength. I These and other objects, which will become apparent from the accompanying description and claims, are achieved by a process which removes heat from a mav terial according to a predetermined cooling rate over a predetermined temperature rangeby means of a novel cryogenic fluid mixture.

'One aspect of this invention comprises a methodfor improving the heat transfer characteristics of a cryogenic 1 fluid at its boiling point, particularly its ability to extract heat from a material or surface. This method consists of adding a finely divided solid to such cryogenic fluid. In

i order to accomplish the desired results the finely divided solid must have its melting point, and at least a portion of its stable liquid phase, within the temperature range formed by the boiling point'of the cryogenic fluid and the temperature of the material from which heat is to be extracted. g

A second aspect of the present invention consists of the'cryogenic fluid mixture itself which is a dispersion of theabove defined solid in the cryogenic fluid, said mixture having a greater ability to extract heat from a material than the pure cryogenic fluid.

A third aspect of the present invention consists of a method't'or controllably extracting heat from a material over a predetremined temperature range comprising contacting said material with a fluid dispersion comprising: (i) a cryogenic fluid, which is at its boiling point and (2) a finely divided solid additive characterized by having its melting point, and at least a portion of its stable liquid phase, within the temperature range formed by the boiling point of the cryogenic fluid and the temperature of the material from which heat is to be extracted,

and wherein the temperature of said material is above the when viewed in connection with the accompanying drawings in which:

FIGURE 1 is a graphshowing the cooling curves obtained on quenching copper test specimens into respectively: pure liquid nitrogen, finely divided ice particles in liquid nitrogen (dispersions varying in solids content from to 65%) and water.

FIGURE 2 is a graph showing the cooling curves obtained on quenching copper test specimens into respectively: pure liquid nitrogen, and mixtures of various finely divided solid additives dispersed in a liquid notrogen.

FIGURE 3 is a graph showing the cooling curves obtained on quenching test specimens, initially at a temperature of 330 F., into. respectively: pure liquid nitrogen, and a 50% solids dispersion of finely divided ice in liquid nitrogen.

FIGURE 4 is a graph showing the cooling curves obtained on quenching test specimens, initially at a temperature of 225' F., into respectively: pure liquid nitrogen, and a 50% solids dispersion of finely divided ice in liquid nitrogen.

FIGURE 5 is a graph showing the cooling curves obtained on quenching test specimens, initially at a temperature of 145' F., into respectively: pure nitrogen, and a 50% solids dispersion of finely divided ice in liquid nitrogen. Y Y

The process of this invention is useful for cooling any material which requires rapid and controlled cooling from a relatively high temperature. Such a material is most frequently a metal, particularly an alloy requiring heat treatment, but it may be a nonmetallic material as for example, a cermet, a ceramic, a cementitious material, or a natural or synthetic rubbery or resinous material. The material may be in any physical formtthus, it may be particulate or be an object such as'a sheet, a rod, a slab. a fiber, or a complex fabricated part.

the heat transfer characteristics of any cryogenic fluid or mixture of cryogenic fluids can be improved by the process of the present invention..'ihe term "cryogenic fluid," as used throughout this disclosure, is intended to mean a substance having its normal boiling point below the freezing point of water, i.e., 32' F. Illustrative cryogcnic fluidsinclude liquid air, as well as, the fluids listed in Table i.

TABLE 1 Cryogenic Fluid Fonnula Boilgng g'oint Chlomtrifluoroethane.. Methylchloride CHsCl. Dimethylother. CH;OCH

The finely divided solid useful as an additive'in the present invention to aiterthe heat transfer characteristics of cryogenic fluid must have at least a portiono! its stable liquid phase within the temperature range formed by the boiling point of the cryogenic fluid and the temperature of the material from which heat is to be extracted. All materials which fall within this generie'deiinltion are suitable forcarrying out the processes of this invention.

A preferred class of flnely divided solids is character- 1 the boiling point of the cyrogenic fluid and the temperature of the material from which heat is to be extracted. I it is to be understood, of course, that reference to a a single fluid or a mixture of several such'fluids, and likewise the terms "a finely divided solid or a finely divided solid additive" are intended to include a single such solid or mixture of several such solids. Table 2 below lists a considerable number of illustrative solid additives including metals, salts, acids and organic materials which are suitable. This list, however, is not all inclusive and is given by way of illustration only.

TABLE 2.8UXTABLE BOLIDADDITIVES' Boiling Framing Finely Divided Solid Formula Point Point F-) Manganeous chloride (MnCh)...."... 2,172 1.202 Cadmium chlorlde.. (CdCis) l. 796 l, 142 Z! (Zn)... 1.692 788 (Cd).. 1.412 609 (Cdi'). 1.314 713 (SnChg. 1,162 475 (BiBrs set 424 (S)...'.. 833 23a (BiCis). R25 446 ercury (lig)... 627 39 Ferric chloride (FeCh) 698 540 Kerosene 380-400 Nonane........ (only). 304 ---08 Stanlc chloride. (BnCh). 237 26 Water (ii|0).. 212 32 Methyl alcohol (Cih0ii)..... 149 -l43 'iriehlorofluoromethane.. (CClsF) 74 i68 Chloroditluorornethane (CHClF|)..... '266 'Ietrafluoromethane....... (Ch) -l98 +299 United States Standard series screen sizes. The data in Table 3 demonstrates that the optimum cooling rate is obtained when the ice crystals have a particlesize of be. tween 20 and 40 mesh. As the particle sizes vary in either direction, i.e., grow larger or smaller from the optimum* range, the time required for cooling the object increases. The plus (-i-) Sign in front of the mesh indicates that the material remained on the screen size indicated, while the minus sign indicates that the material went through the indicated screen size. Thus, 5-3 +6 mesh" in Table3 means thatthe material passed through a 3 mesh screen I but was retained on a 6 mesh screen.

ized by having its liquid phase and at least a portion of its vapor phase within the temperature range formed by Particle size (mesh):

TABLE 3 Eflect of particle size on cooling rate Time, seconds pure liquid nitrogen requires 78 seconds. The effect of solids content upon a dispersion of ice in liquid nitrogen is demonstrated in FIGURE 1 which shows cooling curves, i.e., a plot of time versus temperature, for a 5%, Va a a a and a mixtureof ice in liquid nitrogen. For purposes of comparison curves for liquid water and pure liquid nitrogen are also shown. All percentages are by weight. FIGURE 1 demonstrates that the cooling ability of liquid nitrog'enice mixtures are far greater than that of pure liquid nitrogen. As the percentage of solids decreases, it tends to approach that of the pure cryogenic fluid. However, it can, be seen that as little as 10 percent ice has a great effect upon the cooling rate of the liquid nitrogen.

The useful upper limit of the solids content is dependent upon the fluidity of the mixture. In order to be useful in the process of this invention the mixture must be fluid. The solids concentration at which fluidity ceases depends upon several factors, including the relative densities of the cryogenic fluid and the finely divided solids, the average permitting adequate fluidity.

Cryogenic fluids are generally regarded as poor media'- for heat transfer. The reason for this is believed to be that heat transfer by the slow film-boiling mechanism takes place on smooth surfaces having temperatures more than about 40' F. higher than thenormal boiling point of the fluid. While not wishing to be limited to any theory of opcration. it is believed that the addition of finely divided solids having the'specified characteristics improves heat transfer to cryogenic fluids by causing the heat transfer, in part at least. to occur by melting of the solid and by nucleate-boiling of the resulting liquid on the surface of the material from which heat is being extractedrlt is visualized that this process takes place in three distinct zones. The first zone consists of the material being cooled. The second zone consists of a gas space adjacent to the surface of the material being cooled. and contains solid as well as melted additive particles. The third zone consists of the main body of the cryogenic fluid containing the solid additive particles. In other words'. zone two is located in between zone one 'andzonethree. This arrangement is thought to be representative of the situation during the film-boiling regime of the cryogenic fluid. As cryogenic fluid boils away, the additive particles in zone three are 6 without having reachedtheheat source. This theory is consistent with the data in Table 3 showing that very fine particles (below mesh) and very large particles (above 6 mesh) have little or no effect upon the cooling rate of the cryogenic mixture, and that optimum cooling depends on having an optimum particlev size. Particles of an intermediate size which reach the surface of the heat source will melt, thus causing cooling of the material by' absorbing the latent heat of fusion of the solid. This theory is also consistent with the data in FIGURE 2 showing that a finely divided additive such as Dry Ice will not be effective. since it does not have a stable liquid phase at normal pressures between the temperature of the boiling point of the cryogenic fluid and that of the material being cooled.

The quenching mixtures were prepared as follows. Where the additive was a liquid at room temperature and pressure, the cryogenic quenching mixtures were prepared by atomizing the liquid and spraying the atomized droplets 1 into cryogenic fluid which was contained in a dewar. Atomization was caused-by forcing the liquid through an atomizing nozzle at a pressure of about 5 p.s.i.g. The dispersion was prepared by holding the top of the atomizer about one inch above the level of the cryogenic fluid. The atomized droplets f-roze upon contacting the cryogenic fluid. The mixture was continuously stirred with a conventional two blade laboratory mixer to avoid formation of a frozen crust on the surface of the mixture. The dispersions were prepared in a dewar which held about five pounds of mixture.

In those cases where the additive was solid at inch in diameter. 'A chromel-alumel thermocouple was fused in the center of each specimen for recording its temperature. The thermocouple cold junction was maintained at -320 F... and temperatures were recorded with an automatic fast response millivolt recorder having a chart travel speed of one inch per 10 seconds. The experimental procedure consisted of heating the test specimen in a salt bath un il the entire specimen reached thermal equilibrium. .900 F. in case of the data plotted in FIGURES l and 2. and 330 F.. 225' F., and F. respectively in case of FIGURES 3. 4 and 5. The specimen was then removed from the molten salt bath and immediately plunged into the bath of stirred quenching mixture. The specimen was allowed to remain in the cryogenic dispersion until it came to thermal equilibrium with the bath. Where desirable. the hot'obiect may be contacted with the quenching propelled toward the heat source (zone one) and their traiectory within zone two is determined by ,the balance of forces acting upon the particles. These particles are accelerated towards the heat sources because of the large valume changes that occur when the cryogenic fluid evaporates. After a particle has entered zone two, itis acted upon by at least four separate forces. Vetrically downwardis a gravational force and in the opposite direction there is a viscous drag force due to the gases rushingup through.

media by spraying insteadof immersion.

The data from the cooling curves obtained as described I above has been plotted and is shown in FIGURES l and 2. FIGURE 1. which compares the effect of the solids content of a dispersion of ice in liquid nitrogen has already been discussed. FIGURE 2 shows the cooling curves obtained using a copper test specimen quenched in the followin media: pure liquid nitrogen. 10 and 50 percent bv weight dispersions, respectively. of finely divided crystals of methanol and kerosene. 10 percent brine crystalstcontnining 10% NaCl in water). 10 percent sulfur. and 10 ercent ferric chloride. all dispersed in liquid nitrogen These curves demonstrate how the cooling rate of an immersed object can be varied by the quantity and ltind of the solid additive. t

FIGURES 3. 4 ands demonstrate the effect of varying the initial temperature of the material being cooled. It can be seen that'in each case the use ofa dispersion of 50% ice in liquid nitrogen results in a shorter cooling time than the use of pure liquid nitrogen. However. as the temperature of the hot material is lowered, .the differences between the cooling curves become less. 'Thus, while the improvements resulting from the practice of this invention are properties for the specimen as when quenched in pure liquid nitrogen. Comparison of the cooling curves for achieved even when the temperature of the material, being. I

cooled is relatively low--it may be just above the melting point of the solid additive-considerably greater improve- The use of cryogenic mixtures for quenching aluminum alloy 7075 was also investigated. This alloy which has a nominal composition of 1.5 percent copper, 2.5 percent magnesium, 2.5 percent zinc and minor amounts of silicon, iron and manganese was selectedfor testing because it is one of the most difiieult to heat. treat since its physical properties show a high sensitivity to the quenching rate. Aluminum alloy 7075 is one of the strongest of the aluminum alloys and capable of achieving a tensile strength as high as 83,000 p.s.i. in the wrought condition. Maximum strength is obtained by heat treating it to the T-6 condition. This condition is obtained' by solution heat treating the alloy in the temperature range of 910-- 930' F. and thereafter quenching it in water. The quench is followed by tempering for 24 hours at 240-260 F. The water quenching treatment frequently results in warpage of the aluminum. r

The mechanical properties of test specimens of aluminum alloy 7075 were measured after being quenched in various media from the solution testtreating temperature (9i0-930' F.) and temperature to the T-6 condition as described above. Tensile specimens having a one inch gage length were cut from Ms inch thick sheets in the transverse direction. Two tensile specimens were prepared for each quenching condition. Table 4 below contains a tabulation of the mechanical properties obtained.

TA liLE 4.COMl'ARiS()N 0i" THE MECHANICAL lilitll'i-Ilt- 'llES OF HEAT TREATED ALUMINUM IALLOY 7075 QUENCHED IN DIFFERENT MEDIA Yield Ultimate 7 Quenching Medium Liquid N1 plus Liquid Ns plus The specimens quenched in water show a yield strength- 00 half way between that obtained with liquid nitrogen alone.

and that obtained with water. Liquid nitrogen plus $0 percent kerosene gave a yield strength of 61,000 p.s.i. while liquid nitrogen and'50 percent methanol gave a yield strength of 56,800 p.s.i. An explanation of this difference is obtained from a comparison of the cooling curves shown in FIGURE 2. This comparison reveals that the kerosene mixture gives a faster cooling rate in the temperature interval from S50-750' F. than does the methanol mixture. Since the cooling rate in this temperature interval is known to be critical in determining the physical properties of alloy 7075 it demonstrates that in order to optimize the physical properties of a particular alloy it is necessary to have a tailor-made cooling mixture which will produce the proper cooling rate in the temperature interval critical for that particular alloy. it is the'concept and means for accomplishing this tailor-made-cooling mixture which constitutes the crux of this invention.

Quenching the specimens in a mixture of liquid nitrogen and 50 percent C0; resulted in the same physical these two solutions (see FIGURE 2) shows identical cooling rates. This data also supports the theoretical explanation of the mechanism which requires the solid'additive to have at least a portion of its stable liquid phase within the temperaturerange formed by the boiling temperature of the cryogenic liquid and the temperature of the material being cooled, so that the solid additive will melt and cool the surface of the material during the cooling process. The mixture consisting of 40 percent graphite flakes in liquid nitrogen, and pure liquid argon both gave weaker physical properties than pure liquid nitrogen. The reason for this is that these mixtures give slower cooling rates than pure liquid nitrogen. This demonstrates again that the yield strength of alloy 7075 depends upon the cooling which gives a fast cooling rate in this temperature interval,

but not so fast as'to cause'warpage as is the case with water. I

Good strength properties, however, are not sufficient;

it is also important that thematerial not be undesirably :warped during the quenching operation. Aluminum parts are frequently highly complex, fabricated pieces which have been machined to close mechanical tolerances. These parts must not only be strong but must be substantially free of distortion. in the past, it has frequently been neces-. sary to go through costly straightening operations to cure the warpage caused by heat treating and quenching operations. The tendency of various quenching media to causewarpage or distortion wasexperimentally determined with a modified NavyC test'specimen fabricated from it inch thick aluminum sheet. This test is described-more fully.

in "The Amerizan Society for Testing Metals Handbook,"

8th Ed.. vol. 2, page 24. Thetest was modified by eliminating the notches at the base, changing the gap to 0.50 inch and the diameter to 3.75 inches. Experiments were performed by solution heat treating the C shaped test specimens in a salt bath at 900' F. and then quenching them in various media. The gap opening was care- 7 fully measured before the heat treatment and after the quenching. The amount of change is a measure of the tendency of the quenching medium to cause distortiom Theresults reported in Table 5 below have of :0.002 inch.

TABLE 6.OAP DISTORTION AFTER QUENCHINU 1N VARIOUS SOLUTIONS FROM 000 F.

an accuracy Quenching Medium Change (inches) Percent Change Water 0.0028 0.56 Liquid N itrogeu 0.0003 g 0.06 Liquid Nitrogen 50% [as 0.00% 0. 06

Table 5 shows the distortion that occurred using water, liquid nitrogen, and a dispersion of 50% ice in liquid nitrogen. Each of theresults reported is the averages of five testspecimens. it can be seen that by far thegreatest warpage occurred on quenching in water. it can also be seen that the distortion is minimal with liquid nitrogen .material over" a predetermined temperature range, comprising contacting said material with a fluid dispersion comprising: (i) at least one cryogenic fluid which is at its boiling point, and (2) at least onefinely divided solid additive characterized byhaving its melting point and at least a portion of its stable liquid phase within the temperature range formed by the boiling point of the cryogenic fluid and the temperature of said material, and

wherein the temperature of said material is above the I melting point of said finely divided solid additive.

2. The method of claim 1 wherein the finely divided solid additive is characterized by having its liquid and vapor phases within the temperature range formed by the boiling point of the cryogenic fluid and the temperature of said material. 3 V

3. The method of claim 2 whereiii the material from which heat is to be extracted is a metallic material.

4. The method of claim 2 wherein the material from solid additive has a particle size of from 3 to 250 mesh..

8. The method of claim 2 wherein the finely divided solid additive has aparticle size of from 2010 60 mesh.

9. The method of claim 2 wherein the'finely divided solid additive is ice. v r i 10. The method of claim 2 wherein the cryogenic fluid is liquid nitrogen.

11. A method'for controllably extracting heat from a fabricated metallic article over a predetermined temperature range comprising: quenching said article in a fluid dispersion of ice in liquid nitrogen which is at its boiling point, said dispersion containing 10 to 70 percent by weight ice having a particle size of from'about 3 to 250 mesh.

References Cited 7 UNITED STATES PATENTS 2,772,540 12/1956 Vierkotter 62-64 2,919,862 1/1960 Beike ci a]. 62-64 2,949,392 8/1960 Willey 143-125 3,228,838 1/1966 Rinfret et al 62.-74 x OTHER REFERENCES 0RNL-3415, Special Report, Mar. 4, 1963, relied on 20 vpages 60-68.

CHARLES N. LOVELL, Primary Examiner.

US. Cl. X.R.

Images