US2076026A - Primary light source - Google Patents

Description

April 1 ,M. M. HIRSCH 7 2,076,026 I PRIMARY LIGHT SOURCE I Filed Nov. 8, 1935 7 I F z'g.j 7 I av I I IEQOOO L Q Ii:

MIXTURE OF MONATOMIC GASES III (((J I I I Z=' Zz'g.2.

l I I TUBE I POSITIVE {NEGATIVE I gLRIKES COLUMN I COLUMN" 1 I I ARGON I I I I I I l I l I I I I 2* 1 I I m 'I a: I I I C! I D I U I l w I I I I 0'- I l l 22-- I I :I: I I 3i I I l a I I I BLOCKS I I::::I':::::::':::: INVENTORY CATHODE EMISSION MELWLLE M HRSCH ATTORNEYS.

Patented 6, 1937 ATENT OFFICE;

2,075,026 PRIMARY LIGHT souncn Melville Manuel Hirsch, San Francisco, Calif.

Application November 8, 1935, Serial No. 48,880 Km Australia March 9,1934

' 9 Claims. (01.176-124) My invention relates to a primary light source, and more particularly to a primary light source of the gaseous conductor class, emittin a substantially white light.

The problem of how to obtain substantially white radiation in any type of primary light source has long been before those skilled in the art, and many attempts have hitherto been made to utilize some combination of the so-called "rare m gases in positive column lights to produce a spectrum of full color component in practical form. The monatomic gases such as neon, helium; argon, krypton and xenon are all utilized today in positive column lamps, both for general illumination 5 under certain conditions and for advertising purposes in the sign industry. There are also utilized modifications of the positive column lamp wherein mercury vapor is used to obtain the blue mercury spectrum. In present form, these positive column lamps may be energized by high voltage applied to the column through cold electrodes, or by lower potentials applied through heated electron-emissive cathodes, thus reducing the cathode drop.

In these prior art lamps, however, irrespective of cathode structure; the distinctive line spectrum of the gas involved is obtained when the gaseous filling is confined to a single element, and in each case the light produced is completely lacking in one or more spectral regions so that distinctive colors are radiated, the light being therefore unfit for many uses where a white light, comparable to sunlight, is required.

It might be supposed, however, that mixtures of these various gases, properly chosen, could be energized in a positive column in orderthat the spectrumof one would reenforce the other to fill out the missing spectral regions of either, but 4 the electrical characteristics of the gases themselves have hitherto prevented the simultaneous ionization of a plurality of gases. All references agree that it has not hitherto been possible to obtain any desired hue of light by mixing the monatomic gases and thereafter ionizing them in a positive column. The reason for this lies in the fact that the ionization potentials difier widely, each gas requiring a particular voltage value to cause radiation. If several of these gases are present in a tube lamp, the gas requiring the lowest voltage for ionization will ionize and mask those which require higher ionization potentials; thus the characteristic spectral lines of the gas requir ing the lowest ionizing potential will dominate in the hue of the radiation. If the ionizing potentials 5 are relatively far apart, the spectral lines of the gashaving the highest ionization potential will not show at all. This phenomena has become well known in the art under the name of hogging, and heretofore the only mixtures in which both gases could be simultaneously ionized to give a mixture spectrum are those whose ionizing potentials are relatively close together. For example, neon has only a slightly lower ionizing potential than helium, and it is therefore possible to dilute helium with a small amount of neon and. 10

- ionize the mixture in such a manner that the line spectrum of helium is modified by the red line spectrum of neon to produce an orange light.

It is also possible under certain conditions of temperature to ionize certain mixtures. As an ex- 5 ample, if we consider a tube containing a stable gas, and mercury, whichwill evaporate as the tube warms up, it is known that at low temperatures the spectrum of the gas only will show, while at high temperatures only that of the mercury will 20 show. At-some intermediate temperature, however, both the gas and the mercury vapor are radiative to produce amixture line spectrum. There 7 are grave practical difficulties, however, in maintaining the specific temperature conditions re-' 25 quired toassure that the radiation of. both elements will remain in any definite required ratio. When, however, the gases are such that the mixture relation must remainconstant, or at least vary inthe same degree, no such mixture spec- 30 trum has, as far as I have been able to ascertain, been obtained. The farther the ionizing I potentials are apart, the more difiicult it is to ionize the gas having a higher ionizing potential.

If the line spectra alone of these rare gases be 35 considered and compared,it will be seen that a very attractive theoretical mixture for the pro- I duction of aspectral spread, having a distribution similar to that of sunlight,'would be argon and helium. Helium has abright line in the orange 40 and a well defined red line, some yellow, and only a fair amount of blue. Argon is well represented in the blue and violet, and also in the orange and red, and it would be natural to expect from a true mixture spectruma well represented.line

spectrum deficient, prehaps, in yellow between 5200and 5500 Angstroms.

Unfortunately, however, the combination of helium and argon combines gases whose ionizing potentials are far apart. Reference books say 5 that the final ionizing potential of argon is approximately 16 volts as compared ,to helium at 24.48 volts. These constants differ so widely that ordinarilyargon only is ionized in a mixture, and the statement is repeatedly found in prior art literature that no less than one part of helium in two parts of argon can be detected by spectroscopic examination of the ionized mixture, and even then is only detectable at the cathode and 5 not in the positive column. The theory of this action appears to be that in such a mixture not only is the helium harder to ionize, but the mean free path of the helium is so much longer than that of argon that the helium atoms remain inert, the entire current being carried by the argon. Argon is 'also a large molecule and helium a small one. Thus, all physical and electrical conditions apparently combine to prevent simultaneous ionization of both gases in a mixture of argon andhelium, having the proportions desired for the production of a combined line spectrum, when energized in a positive column. Furthermore, even if it be considered that both gases could be ionized simultaneously, the spectrum produced might be expected to be a line spectrum with no banding and no approach to a continuous'spectrum such as is present in sunlight.

The present invention has two main objects:

First, to simultaneously ionize two or more monoatomic gases at the same or diiferent pressures in a tube light; and second, to produce this simultaneous ionization in such a manner that a band or continuous spectrum is produced, thus giving rise to a light having a very close approximation to that of sunlight.

My invention possesses numerous other objects and features ofadvantage, some of which, together with the foregoing, will be set forth in the following description of specific apparatus embodying and utilizing my novel method. It is therefore to be understood that my method is applicable to other apparatus, and that I do not limit myself, in any way, to the apparatus of 40 the present application, as I may adopt various other apparatus embodiments, utilizing the method, within the scope of the appended claims.

Before giving a description of a preferred means and method for accomplishing the result herein to be described, a. greater understanding thereof may be had by a brief reference to the atomic theories developed to explain radiation from ionized gas atoms, and as there is, as in all new research work, some difference of opinion regarding these theories, the explanations given are those found in standard reference books of the art to which the reader is referred for details beyond those herein given. I

The rare gas molecule is said to have a nucleus 55 around which electrons revolve in orbits of different radii. Under stimulus, either from electron bombardment or ion impact, electrons are dislodged from inner orbits outwardly to other orbits, and then fall back to inner orbits again. 0 Energy is radiated during these transitions, characteristic spectral lines appear in the radiation according to the length of the transitions, and the energy radiated is the difference between the initial and the final path. The character- 5 istic spectrum, therefore, of the normal ion of the gas will show lines in a certain related series which are related to orbital distances. There are, however, within the atoms certain orbits where the electron cannot return toward the center unless there be a. positive addition of energy from without the ion. These are called "metastable orbits, and when an electron does leave this orbit the ion is said to be in a m'tastable state. 7 The ion of the various rare gases, when in such a metastable state, has an altogether different ionizing potential than in the normal state, and if in a mixture of the rare gases metastable ions can be produced, all of the gases could be ionized'simultaneously. For example, metastable helium will ionize with argon; metastable neon will ionize with argon; and metastable argon will ionize with mercury vapor. Thus it can be stated that if the monoatomicgases in a mixture can be so ionized as to produce metastable ions in the proper quantities, all of the gases in the mixture may be simultaneously ionized, irrespective of their percentage relation.

Such a mixture, suitably ionized to produce metastable ions, would only partially solve the problem of producing a white light; it would produce only the combined line spectrum of the gases involved. It then becomes necessary to so energize the ions that either a continuous spectrum is produced, or at least a spectrum having wide bands, the bands being a widening-of the lines present to the extent that they completely, or at least partly, fill up the dark spaces between the original lines. My invention not only produces the metastable ions-necessary for simultaneous ionization of a gaseous mixture, but also produces a. continuous, or band, spectrum.

I will next consider what I have found necessary to do to the ionized atom in order that the lines of the spectrum may be broadened to bands, and as far as I know this can only be done by producing what is known as a recombination spectrum, although I am not aware that such a recombination spectrum has ever actually been produced in the monatomic gases,

although theoretically it can be predicted if conditions for the production of such a recombination spectrum can be fulfilled. For example, if an ion has at the moment a free orbit, it is known that an electron from one of the outermore orbits can return thereto, giving the characteristic spectral line at a definite wavelength. If, however, an electron from outside the ion can be moved into the open orbit, with an energy only minutely different from the energy, let us say, released by an electron traveling from the outermost orbit to the one to be filled, then a radiation would take place differing from the normal line only by a very minor degree. It therefore follows that if electrons can enter an orbit with all degrees of energy varying from exactly that released during the passage from the outermost one to the one in question, to some value greater than that producing the normal line, then there will appear on one side of the line a continuous band filling the gap between the spectral lines.

As each orbit of the ion becomes free at various times to capture an electron, each one of the spectral lines will have associated therewith this band; and if a sufiicient number .of electrons can be supplied to the ions having an energy content clcsely approaching that of the normal interatomic traversal producing the lines, then a continuous, or at least widely banded spectrum, will appear.

Practically, however, such a condition is not easy to achieve, and as faras I know never has been achieved in a practical manner, because the electrons entering the ion, to produce this banding, must have almost zero energy which means that their velocity must be extremely low; in fact, they must almost be standing still. Such a condition is impossible in a positive column lamp, first, because in the positive column there are very few free electrons; and second, because those that aovaoae a recombination spectrum, the two phenomena combining to produce a white light.

The above theory is offered as at least a partial explanation of the operation of my invention, but I do not wish to be bound solely by this explanation as further observation of the peculiar characteristics of my invention may determine that this theory is wrong in one or more respects, but at least at the present time, and in view of present knowledge, it serves to explain the peculiar action of the tube of my invention of which I will now describe a specific'preferred embodimentz' Figure l, accompanying this specification and made a part hereof, shows in diagrammatic form a hot cathme luminous tube and operating circuit. i

Figure 2 shows a characteristic curve, showing discharge current decreasing with cathode emission. The curveis idealized and not drawn to scale.

The tube itself is preferably of standard construction having enlarged ends 5, a luminous tube section 2, and hot cathodes in both ends. These hot cathods preferably comprise a thimble 3, heated by a heating coil 3, and I prefer to coat the thimble 3 with an electron-emissive material of the highest possible eificiency. Many of these are well known in the art and may comprise various mixtures of strontium and barium oxides. The heating coils 4 may conveniently be made of tungsten wire and are energized by heater wind.- ings 5 mounted on transformer core '6, the primary l of which is energized directly from alternating current mains 8, the primary preferabhr being adjustable, as indicated by arrow 9. Each thimble 3 is connected directly to a main lead 8, and a choke ill is preferably included in one cathode connection for purposes hereafter to be explained. A meter H may be placed in one cathode connection to measure tube current.

Inasmuch as I prefer to operate my tube directly from 220 volt alternating current mains, the length of the luminous portion 2 is made relatively short in order that ionization may take place therein; and in a specific example of the light, as built by me, the distance between cathodesis approximately four inches. The tube, after assembly, is placed on a pump line, thoroughly evacuated, the filaments burned, and the cathodes heat-treated to remove all unwanted gas content, as is well known in the art, and the tube is then filled with a mixture of monatomic gases. While I do not wishto be limited in my choice of either gaseous elements or strict proportions, I prefer to mix such gases as would, under ordinary conditions, and separately excited and combined,

produce a line spectrum covering, to a large extent, the visible range; and I have found one preferred mixture to be approximately fourteen per cent purified helium, the balance being purified argon, at a total pressure of fourteen millimeters of mercury. This mixture, it should be noted,

falls far below, in the proportion of helium, that v at which the spectral hues of helium can be detected when a mixture of these two gases is ionized in a positive column. I have'found that from five to twenty per cent of helium may be respect to each other.

used satisfactorily in the mixtures at finalpressures of from ten to forty millimeters of mercury. The final spectrum is practicallyidentical; ad-

justments in cathode heating, tube length and Due to the fact that it is necessary, inmost instances, for me to obtain the rare gases from standard commercial sources, Lhave found that helium, as supplied by them, always contains a trace of neon; and furthermore, I have found that if the tube is evacuated by a mercury pump, even though liquid air traps are used, there will also be a trace of mercuryvapor in the tube. As my method of operation-of the tube causes emission from all gases present, even these traces will contribute toward the final spectrum, bolstering up the red and the blue end, although I do not find that their presence is in any way necessary, being merely a factor in the emission which can be ignored. I may, however, desire to increase the amount of these gases in order to produce a light for specific purposes.

To start the lamp gradually, and to illustrate and explain meanwhile the peculiar action thereof as discovered by me, the heaters are energized at a low temperature and the line potential applied across the cathodes.. Under these conditions the heat of the cathodes may be slowly raised until there is a sufficient electron emission therefrom to start the lamp and ionize gas in tube 2; and the normal way of operating any lamp of this sort as a positive column lamp is to operate the cathodes 3 at a temperature where sumcient electrons will be fed into thecathode dark space to maintain a positive column discharge between the two negative dark spaces, two spaces being referred to because of the rapid alternation of potential on the cathodes with Under these conditions the lamp as described above operates as a positive column lamp, ionizing only the argon in tube 2; and the current carried by the tube between cathodes, as indicated by meter H, i relatively high. The light emitted by the tube operating in this condition shows the characteristic argon spectrum with no trace,of the spectral lines of the other gases in the tube; in other words, the tube is'operating purely as a prior art lamp of the positive column type, and giving results such as would be ordinarily expected from such a setup. The argon spectrum is weak, of a purplish-gray color, and utterly useless as an illuminant.

I next proceed to heat cathodes 3 to higher and higher temperatures, supplying to the tube 2 arf enormous excess of electrons beyond the value needed .to maintain positive column ionization. These electrons permeate the positive column, and as the heat of the cathodes is increased there comes a series of electron emission values at which the positive column; is destroyed, the

electrons having migrated entirely throughout the positive column and created a negative space charge therein, lowering the current through the column eventually to the point where the gas will cease ionization and extinguish by blocking. Thus the current through the lamp varies inversely with the cathode emission, as shown in Figure 2.

It should be noted that Angstroms assertion that the "Intensity of light is proportional to the first power of the current is entirely incorrect as applied to the lamp of my invention, although it should also be noted that the total column energization, metastable ions are apparently produced in the gases and all of the gases therein become simultaneously ionized, with a tremendous increase in brilliancy in the column and the emission of a substantially white light of high intensity. A balance can be found where the cathode temperature, or in other words the electron supply to the column, is such that a relatively low current is passing through the tube, with maximum brilliancy of radiation. After this condition has been reached, if the I tube is examined with a spectroscope, it will be seen that the line spectrum has disappeared and that each line has changed to a broad band extending from the line wavelength as a head and 5 gradually decreasing in intensity as the band widens, leaving practically no dark regions in the spectrum and thus giving evidence of the theory of the production of a recombination spectrum as outlined above. As each line of all the gases contained in the column has associated with it the band referred to, it will be seen that a continuous spectrum of almost any color content including white can be produced from a mixture of gases whose spectral lines are distributed in 5 the proper manner.

paths along the glass tube as highly resistant as possible; second, it is desirable to make the column relatively short in order that diffusion of electrons through the column may not take too long; third, it is desirable that the voltage applied between the cathodes be relatively low'in order that the electrons may not be swept out; and fourth, it is desirable that a relatively high gas pressure be used to maintain the electrons in position.

' In ordinary commercial use, this lamp is energized with an amount of filament current such that the cathodes will immediately rise to their preferred operating temperature to produce a negative column light. Under these conditions the transition from a straight argon positive column high current discharge, to a negative col- ;5 umn low current mixture discharge takes place relatively rapidly, and the sequence as described immediately above is automatic, making it unnecessary that a variable tap such as is shown at 9 be used on the transformer. The choke I0 is not needed when the lamp is running as a nega tive column lamp for reasons which are obvious from the falling current characteristic. The choke It, therefore, is only necessary during the instant that the tube is operating on the straight positive column argon current; that is, during Thus I am not only able to obtain with the monatomic gases a white the starting of the lamp, the lamp thereafter being completely stable.

The light produced by the lamp operating as described has the appearance of north light, although the radiation to the eye is less blue than north light with perhaps a very slight greenish tinge. I have, however, succeeded in entirely eliminating this greenish tinge by a slight increase in helium content or by the addition of a minute quantity of sodium. Color matching is possible, practically all objects maintain their natural colors under the lamp, and exhaustive spectroscopic examination of the emission has shown that the spectrum, with the gases as outlined above, is practically continuous. The spectrum may also be made still more continuous by widening of the hands by refinement of variables,

as outlined above, tending to stabilize electrons within the column more and more; and my invention makes it possible, in its optimum application, to obtain a completely continuous spectrum by widening the lines to broad bands in many gases or combinations of gases which do not completely cover the visible spectrum with lines, such a widening of bands being only a matter of degree and only involving the problem of electron stabilization within the column.

While I have as yet no actual evidence to prove the fact, I believe also that it is desirable to supply a sufficient number of electrons to the tube and to obtain such an electron balance within the column between the cathodes that free electrons will remain in this space between the alternations of the current. The reason for this is that, as far as I have been able to investigate, after the negative column has once been established the alternations are sufliciently fast with respect to the electron leakage out of the column, so that each illumination takes place with the gas lighting at once both in the metastable state and with immediate recombination occurring. Inasmuch as it takes an appreciable length of time, even though relatively small, to produce the initial production of the negative column when the control switch is closed, it is my opinion that the negative column would not have time to be extended from the cathode at each altemation, and therefore that the electrons must remain in the space between the cathodes during the dark portion of the cycle.

The fact that a primary light source of the type described above is extremely efllcient can be readily deduced from the behavior thereof. Light intensity without change of spectrum increases almost continuously up to the point where blocking occurs and the lamp goes out at the specific voltage involved. When this occurs the voltagebetween cathodes may be increased and the light reestablished, blocking again occurring as the cathodes are still further heated to supply more electrons to the column. In the interest of efliciency, therefore, it is desirable that relatively low voltage between cathodes be used in order that too great a current consumption be not needed to heat the cathodes themselves.

Therefore, not only have I produced a primary light source dependent upon ionized gas for radiation, wherein various gases of different ionization potentials may be simultaneously ionized, but I have also produced a light wherein continuous and band spectra are produced with an extremely high radiation efliciency. The lumens per watt obtainable with the lamp of my invention are nearly twice that of the best commercial hot cathode lamps hitherto produced, even though the latter give monochromatic radiation.

The negative column lampdescribed is not tricky to operate, is reproducible in quantities, and life tests show that the conditions producing the negative column radiation are not in any way conducive to short life. Tests under commercial conditions have shown the phenomenon to be as reliable after 1000 hours of energization reducing the cathode drop until the helium is ionized and the positive column resistance increases.

2. The method of obtaining a substantially white light in a gaseous conduction lamp which comprises mixing argon and helium at a pressure of the order of ten millimeters of mercury with helium at less than twenty-five per cent of the mixture, and supplying energy at a fixed voltage to said mixture with an over-supply of electrons at the energization cathodes.

3. The method of obtaining a substantially white light in a gaseous conduction lamp which comprises mixing argon and helium at a pressure of the order. of ten millimeters of mercury with helium at less than twenty-five per cent of the mixture, and supplying electrons to the luminous column until its resistance rises.

4. The method of obtaining a substantially white light in a gaseous conduction lamp which comprises mixing argon and. helium at a pressure of the order of ten millimeters of mercury with helium at less than twenty-five per cent of the mixture, supplying electrons to the luminous column until its resistance rises, and there after operating said lamp above the point where substantial increase in luminous column resistance occurs.

5. The method of obtaining a substantially white light in a gaseous conduction lamp which comprises mixing argon and helium at a pressure of the order of ten millimeters of mercury with helium at less than twenty-five per cent of the mixture, supplying electrons to the luminous column until its resistance rises, and thereafter operating said lamp at the point where the curve of light intensity crosses the curve of luminous column resistance.

6. The method of simultaneously ionizing both helium and argon in a gaseous conduction lamp when the percentage 01 helium is less than twenty-five per cent, which comprises the step of adjusting the electron supply at the cathodes until both helium and argon are ionized with a fixed voltage across the lamp.

7. In a gaseous conduction lamp, the method of changing a positive column to a negative column with an accompanying spectral change, which comprises energizing said lamp at relatively low alternating voltage with a relatively high gas pressure, and supplying electrons to each end of said column in sufilcient quantity to permeate said column and neutralize the greater portion of the positive charge therein.

8. In a gaseous conduction lamp, the method of changing a positive column to a negative column with an accompanying spectral change, which comprises supplying an excess of electrons over that required to maintain a positive column, and applying a voltage across said column sufiicient to cause said electrons to enter the column and raise the resistance thereof.

9. In a gaseous conduction lamp, the method of changing a positive column to a negative column with an accompanying spectral change, which comprises supplying an excess of electrons over that required to maintain a positive column, applying a voltage across said column sufficient to cause said electrons to enter the column and raise the resistance thereof, and thereafter adjusting the voltage to the point where electron activity is at the minimum necessary to maintain said charge within the column.

MELVILLE MANUEL HIRSCH.

Images