Apparatus For Heating A Sample With Radiation Energy Concentrated By A Reflector Essentially Composed Of Spheroidal Surface Portions

Abstract

A reflector for use in concentrating radiation energy onto a crystalline sample, comprises two pairs of congruent and coaxial prolate spheroidal surface portions, each of which is disposed outwardly of the other. A predetermined number of confocal prolate spheroidal surface portions may be disposed between the adjacent ends of the coaxial spheroid portions of each pair of spheroidal surface portions, and a pair of hemispherical surfaces may be disposed outwardly of the coaxial spheroidal surface portions. The coaxial spheroids of each pair have a common focus, and the adjacent coaxial spheroids of the respective pairs have a common focus. The foci of the confocal spheroids are coincident with those two foci of the coaxial spheroids of each pair which are conjugate to the common focus of the pair, and the centers of the hemispherical surfaces are coincident with the respective axially outer ones of the conjugate foci. Sources of radiation energy are placed at the respective axially outer ones of the conjugate foci. The sample is placed at the common focus of the spheroidal surface portions.

Description

This invention relates generally to radiation heating apparatus, and more particularly to an apparatus for heating a sample with radiation energy concentrated onto the sample by a pair of reflecting, spheroidal surface portions. The invention is especially suited for use in an apparatus for growing a single crystal by means of a floating zone method.

It is known to heat a sample with radiation energy that is concentrated onto the sample by either an ellipsoidal or a paraboloidal reflector. The prolate spheroidal reflector is superior to the paraboloidal and the general ellipsoidal reflectors but is still defective in that the resulting distribution of radiation incident on the surface of the sample is not uniform. Even when the sample is rotated, it is still heated by a considerably nonuniform radiation distribution which is objectionable in some applications, as, for example, in the floating-zone method for growing a single crystal because of the resulting poor quality of the grown crystal.

An object of this invention is to provide an apparatus having a radiation distribution that is sufficiently uniform for use in various applications in which strictly uniform distribution is required.

A further object of the invention is to provide an apparatus of the type described that is especially well suited for growing a single crystal by the floating zone method.

Another object of the invention is to provide an apparatus of the type described that enables the radiation energy to be concentrated onto an image that is substantially congruent with at least one source of radiation energy.

Still another object of the invention is to provide an apparatus of the type described that provides an excellent uniform radiation distribution around the periphery of a cylindrical sample.

According to this invention, there is provided an apparatus for heating a sample by concentrated radiation energy which comprises a pair of inwardly reflecting, substantially prolate spheroidal surface portions disposed outwardly of each other, said spheroids having aligned major axes and a common focus. A source of radiation energy is placed at one of the two foci of these spheroids which are conjugate to the common focus, and the sample is placed at the other of the two conjugate foci. The spheroids are preferably congruent, and the spheroidal surface portions enclose the aligned major axes of the spheroids at a portion of these axes lying between the radiation source and the sample being heated.

According to one aspect of this invention, the apparatus includes a second pair of spheroidal surface portions and a second source of radiation energy in symmetry with respect to the first-mentioned pair and the first-mentioned source of radiation energy, respectively.

According to another aspect of the invention, a predetermined number of inwardly reflecting, substantially prolate confocal spheroidal surface portions is provided between the adjacent ends of the coaxial spheroid portions of each pair of surface portions, the confocal spheroids having their foci at the two conjugate foci of the pair.

According to a further aspect of the invention, the apparatus may further comprise an inwardly reflecting, substantially hemispherical surface disposed axially outwardly of that coaxial spheroidal surface portion at one focus of which the source of radiation energy is placed. The hemispherical surface has its center at the last-mentioned one focus and a diameter greater than the dimensions of the radiation energy sources.

To the accomplishment of the above and to such further objects as may hereinafter appear, the present invention relates to an apparatus for heating a sample with radiation energy concentrated by a reflector essentially composed of spheroidal surface portions substantially as defined in the appended claims and as described in the following specification taken together with the accompanying drawings in which:

FIG. 1 is a schematic axial sectional view of a conventional prolate spheroidal reflector;

FIG. 2 is a schematic axial sectional view of a reflector according to an embodiment of this invention;

FIG. 3 is a graph showing numerical limits for several preferred modifications of this invention; and

FIG. 4 is a schematic diagram showing a preferred arrangement of the confocal spheroidal reflectors used in the embodiment of this invention shown in FIG. 2.

In a conventional prolate spheroidal reflector, as shown in FIG. 1, the radiation from a point source placed at a first focus F.sub.1 is reflected by the reflector to a second focus F.sub.2. In practice, the source of radiation has finite dimensions. The radiation energy incident on various points of the surface of a sample placed at the second focus F.sub.2 is determined essentially by a magnification factor m of the image and the solid angle within which the radiation incident thereon is emitted from the source. The radiation 1 of the source reflected by a reflector portion 2 produces a real image with a magnification factor m given by f.sub.2 /f.sub.1, where f.sub.1 and f.sub.2 are the distances from the reflector portion 2 to the first and the second foci F.sub.1 and F.sub.2, respectively. The left and the right halves of the image are primarily formed of the radiation reflected by a left reflector portion 3 placed to the left of the plane passing through the second focus F.sub.2 perpendicular to the major axis, and by the remaining, right reflector portion 4, respectively. It follows, therefore, that the images of the respective halves are greater and smaller than the source. The energy concentrated to the images of the left and the right halves is dependent on the solid angles A.sub.1 and A.sub.2 subtended by the left and the right reflector portions 3 and 4, respectively. These solid angles A.sub.1 and A.sub.2 are given by

A.sub.1 = 2 .pi.[1 + 2 .sqroot.1 - b.sup.2 /a.sup.2 /(2 - b.sup.2 /a.sup.2)]

and

A.sub.2 = 2 .pi.[1 - 2 .sqroot.1 - b.sup.2 /a.sup.2 /(2 - b.sup.2 /a.sup.2)],

where a and b are the major and the minor axes, respectively. The energy incident on the left half is obviously greater than that on the right half. Although the quantitative calculation of the radiation distribution is very difficult, the combined effect of the magnification factor and the solid angle is that the radiation energy distribution is of a larger amount and relatively flat on the left side of the image and of a lesser amount on the right side, having a sharp peak at the right-side surface portion opposite to the source. As a result of this unequal radiation energy distribution, the prior art reflector of FIG. 1 is unacceptable for use in many crystal growing applications.

Referring to FIG. 2, an embodiment of this invention is shown which avoids the drawbacks of the prior art reflectors. As shown, the reflector comprises a pair of inwardly reflecting substantially prolate spheroidal surface portions 11 and 12 disposed outwardly of each other. The spheroids have aligned major axes and a common focus F.sub.10 and are preferably substantially congruent. A source of radiation energy 13 is placed at one focus F.sub.11 of those foci F.sub.11 and F.sub.0 of the coaxial spheroids which are conjugate to the common focus F.sub.10. A holder 14 is provided for placing a sample 15 at the other focus F.sub.0 of the two conjugate foci F.sub.11 and F.sub.0. The sample 15 may be typically a polycrystalline rod that is subjected to a floating zone method to be grown into a single crystal. The radiation energy source 13 may be helically wound filament of a halogen lamp that is disposed parallel to the growing single crystal. Preferably, the spheroidal surface portions 11 and 12 are continuous at the junction (not shown) and have closed axial ends (not shown).

The radiation emitted from source 13 passes through the common focus F.sub.10 and is focussed at sample 15. With congruent spheroidal surface portions 11 and 12, it is easily understood from the relations of the magnification factor m and the solid angles A.sub.1 and A.sub.2 that the image of radiation source 13 produced at sample 15 is substantially of the same dimensions and energy distribution as source 13. It is, therefore, possible to provide a heating zone of excellent uniform azimuthal energy distribution at sample 15, to thereby concentrate the radiation on sample 15 at a desired heating area by selecting the dimensions of source 13, and to provide a heating area of the desired energy distribution by accordingly modifying the energy distribution of source 13. It is, however, inevitable in this embodiment, that the radiation directed from source 13 to sample 15 is impeded by the sample and the adjacent portions of the sample holder 14, to thereby reduce the energy density on the right side of sample 15. This reduction increases when sample 15 is enclosed within a quartz tube (not shown) for providing the desired atmosphere.

Referring further to FIG. 2, the reflector apparatus of the invention further comprises a second pair of inwardly reflecting, substantially prolate spheroidal surface portions 21 and 22 disposed outwardly of each other and of the pair of spheroidal surface portions 11 and 12. The second pair of spheroidal surface portions 21 and 22 have a common focus F.sub.20 and major axes aligned with the first-mentioned major axes and are preferably congruent with the first pair of spheroidal surface portions 11 and 12. One focus F.sub.0 of those foci F.sub.0 and F.sub.22 of the coaxial spheroids of the second pair which are conjugate to the common focus F.sub.20 of the second pair is coincident with focus F.sub.0 of the two conjugate foci F.sub.11 and F.sub.0. A second source 23 of radiation energy is placed at the other focus F.sub.22 of the two conjugate foci F.sub.0 and F.sub.22. Preferably, the second pair of prolate spheroidal surface portions 21 and 22 are continuous at the junction (not shown) and have closed axial ends (not shown). Further improvement may be achieved in the uniformity of the azimuthal energy distribution at sample 15 so that, when the device is used to grow a single crystal by the floating zone method, it is possible to subject the crystal to rotation of an extremely low frequency, to thereby attain the desired conditions for optimum crystal growth.

As thus far described, the radiation directed from each of radiation sources 13 and 23 to sample 15 within the solid angle subtended by the junction of the spheroidal surface portions 11 and 12 or 21 and 22 of each pair of prolate spheroidal portions, is divergent at sample 15 to be converted to heat loss and thereby to adversely affect the azimuthal energy distribution at the sample. With congruent spheroidal surface portions 11, 12, 21, and 22, the last-mentioned solid angle is the solid angle A.sub.2 mentioned with reference to FIG. 1. The heat loss therefore decreases as the ratio b/a is reduced. The minor axis b cannot be much reduced because of the finite dimensions of radiation sources 13 and 23 and sample 15. An excessive dimension a of a major axis is objectionable in view of the resulting bulkiness of the apparatus.

Referring to FIG 3, a curve 26 shows the relation between the ratio b/a and the percentage of the solid angle A.sub.2 in question to 4.pi.. It is appreciated from curve 26 that the heat loss suddenly increases as the ratio b/a approaches unity and that the ratio b/a need not be much reduced to achieve a practically negligible heat loss, with the result that a ratio b/a not greater than 0.8 gives a loss of about 5 percent at most. The heat loss results in the maximum reduction in energy at the surface portion of sample 15 facing radiation source 13 or 23. The maximum heat loss reduction is given by

[(b.sup.2 /a.sup.2)/(2 - b.sup.2 /a.sup.2)] .times. 100 percent

and, as shown by a curve 27 in FIG. 3, about 22 percent and 5 percent for the ratios b/a of 0.8 and 0.6, respectively.

Referring now to FIGS. 2 and 4, the reflector apparatus of the invention may still further comprise a predetermined number of inwardly reflecting substantially prolate confocal spheroidal surface portions 31 and 32, between the adjacent ends of the coaxial spheroid portions of each pair. Although only three confocal spheroidal surface portions are shown in FIGS. 2 and 4, a greater number may be included when desired. The foci of the confocal spheroids are placed at the conjugate foci F.sub.0 and F.sub.11 or F.sub.22 of each pair of spheroidal surfaces. The outermost confocal spheroidal surface portion 31 is continuous to the ajdacent coaxial spheroidal surface portions 11 and 12 or 21 and 22 of each pair of prolate spheroidal portions, with its end peripheries lying on the respective intersections of the adjacent coaxial spheroidal surfaces of each pair, with a first circular conical surface having its axis coincident with the major axes of the spheroids, the vertex at the common focus F.sub.10 or F.sub.20 of each pair, and a predetermined vertical angle. The end peripheries of each of the inner confocal spheroidal surface portions 32 lie along the respective intersections of the first conical surface with the second circular conical surfaces having their respective axes coincident with the major axes and the respective vertices at the two conjugate foci F.sub.0 and F.sub.11 or F.sub.22 of each pair and passing through the respective end peripheries of the outwardly adjacent confocal spheroidal surface portions 31 and 32, the last-mentioned end peripheries being in staggered relation to the end peripheries of the confocal spheroidal surfaces 31, 32.

The inner confocal spheroidal surface portions 32 are supported by the coaxial spheroidal surface portions 11, 12, 21, and 22 with suitable means (not shown). With a relatively large vertical angle of the first conical surface, the images of sources 13 and 23 produced at sample 15 have a magnification factor m of unity. Too large a vertical angle, however, results in an unnecessarily large number of the confocal spheroidal surface portions 31 and 32. In practice, the value of the vertical angle is not critical. The number of the confocal spheroidal surface portions is predetermined so that no similar surface portions are disposed in the proximity of the common foci F.sub.10 and F.sub.20 where the radiation is concentrated to a considerable extent to damage such surface portions, if any. It should be recalled here that the ratio b/a need not be very small so that the number of the necessary confocal spheroidal surface portions is reduced and it is possible to omit some intervening confocal spheroidal surface portions that are to be placed in the proximity of the innermost one.

As will be readily seen, Fresnel lenses or similar means (not shown) may be used in place of each set of the confocal spheroidal surface portions for concentrating on sample 15 the divergent radiation emitted from sources 13 and 23 and not towards the adjacent ones of the coaxial spheroidal surface portions 11 and 22. On account of the availability of a considerably large value of the ratio b/a, the Fresnel lenses or the similar means need not concentrate that portion of the divergent radiation which is emitted from sources 13 and 23 within a predetermined, relatively small solid angle around the major axes.

Referring still further to FIG. 2, the reflector apparatus of the invention may further comprise an inwardly reflecting substantially hemispherical surface 36 disposed axially outwardly of that coaxial spheroidal surface portion 11 or 22 of each pair at one focus F.sub.11 or F.sub.12 at which radiation source 13 or 23 is placed. Each hemispherical surface 36 has a diameter greater than the axial or longitudinal dimensions of sources 13 and 23 and is supported by means (not shown) with its center in coincidence with focus F.sub.11 or F.sub.22. The radiation emitted from sources 13 and 23 axially outwardly is reflected by the hemispherical surfaces 36 back to the respective sources 13 and 23 to raise the temperature of those sources. This enables radiation sources 13 and 23 to be run at a power less than the nominal power.

While the invention has been herein specifically disclosed with respect to several preferred embodiments, it will be apparent that modifications may be made therein all without departing from the spirit and scope of the invention.

Claims

What is claimed is:

1. An apparatus for heating a sample with concentrated radiation energy comprising a first pair of inwardly reflecting substantially prolate spheroidal surface portions disposed outwardly of each other, said spheroidal surface portions having aligned major axes and a first common focus, a first source of radiation energy placed at one of the first two foci of the spheroids providing said spheroidal surface portions that are conjugate to said common focus, a second pair of inwardly reflecting substantially prolate spheroidal surface portions disposed outwardly of each other and of said first pair of surface portions, said second pair of spheroidal surface portions having a second common focus and major axes aligned with said first-mentioned major axes, one of the second two foci of the spheroids defining said second pair of spheroidal surface portions that are conjugate to said second common focus being placed at the other of the first two conjugate foci, a second source of radiation energy placed at the other of the second two foci, and means for holding a sample at said other of said first two conjugate foci, said first and second pairs of spheroidal surface portions substantially enclosing said major axes at a portion thereof lying between said first and second sources of radiation energy.

2. The apparatus as claimed in claim 1, wherein said first and second pairs of spheroidal surface portions are substantially congruent.

3. The apparatus as claimed in claim 2, wherein the adjacent ends of said first and second pairs of spheroidal surface portions are continuous.

4. The apparatus as claimed in claim 2, wherein the ratio of the minor axis to the major axis of each of said congruent spheroidal surface portions is not greater than 0.8.

5. The apparatus as claimed in claim 4, wherein said ratio is not greater than 0.6.

6. The apparatus as claimed in claim 2, further comprising means for forwardly directing the divergent radiation to the sample, said divergent radiation being the radiation which is not directed from each of said sources to those spheroidal surface portions at said one focus of which said source is placed.

7. The apparatus as claimed in claim 6, wherein said divergent radiation does not include the radiation emitted axially inwardly from each of said sources within a predetermined small solid angle around said major axes.

8. The apparatus as claimed in claim 7, wherein said concentrating means comprises a plurality of inwardly reflecting substantially prolate confocal spheroidal surface portions disposed between the adjacent ends of the coaxial spheroid portions of each of said pairs of spheroidal surface portions, said confocal spheroids having foci at said two conjugate foci of said pairs of spheroidal surface portions.

9. The apparatus as claimed in claim 8, wherein the outermost one of said confocal spheroidal surface portions is continuous to the adjacent coaxial spheroidal surface portions of each of said pairs of spheroidal surface portions,

the end peripheries of said outermost one of said confocal spheroidal surface portions lying along the respective intersections of said adjacent coaxial spheroidal surfaces of the pair with a first circular conical surface having an axis coincident with said major axes, a vertex at the common focus of the pair, and a predetermined vertical angle,

the end peripheries of each of the inner ones of said confocal spheroidal surface portions lying along the respective intersections of said first conical surface with second circular conical surfaces having their respective axes coincident with said major axes and the respective vertices at said two conjugate foci of the pair and passing through the respective end peripheries of the outwardly adjacent one of said confocal spheroidal surface portions, the last-mentioned end peripheries being in staggered relation to the end peripheries of each of said confocal spheroidal surface portions.

10. The apparatus claimed in claim 2, further comprising an inwardly reflecting substantially hemispherical surface disposed axially outwardly of that coaxial spheroidal surface portion of each said pair of one focus of which the source of radiation energy is placed, each of said hemispherical surfaces having a center at said last-mentioned one focus and a diameter greater than the axial dimension of said sources.