I _ — — — — _ — — — — — — — WU 1W .THS THE EXCESS ENERGY i’ COLLISIONS BETWEEN RARE GAS iONS AND M15 F M 4$ 1_ *5/15}- ?N 1'4 5.4M \t-.?i_.'§]»’abE AH Thesis for the Degree of M. 5 Earl Kenneth Van Tassel 1938 lllllllllllllllllllllHIHIIHHIIIHIIHIJlIHUIIWIHHIHII 1293 01774 9742 F. , f LIBRARY . Michigan State , University. PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINE-3 return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE THESIS ON THE EXCESS ENERGY AT COLLISIONS BETWEEN RARE GAS IONS AND NEUTRAL ATOMS IN A LOW VOLTAGE ARC. by Earl Kenneth Van Tassel A Dissertation Presented to the faculty of Michigan State College in partial fulfillment of the requirements for the degree of Master of Science. 1928 1-0‘§U:3() TABLE OF CONTENTS Introduction 'Part I. Types of Impacts. Types of the first kind Types of the second kind Positive ion impacts Mechanism of positive ion impacts Part II. Low Voltage Arc. Part III. Experimental. Apparatus. Manipulation Part Iv. Results. Figures 1. Apparatus 2. Wiring diagram. 3. Vamfin system. Charts 24,25,26, .Plates Computations Bibliography 1. 3. 3. 4. 8. 11. 14. 16. 19. 13. 15. 17. 87. 20. 28. 39. I am indebted to and very grateful for the services of Ir. J. G. Black who suggested the problem and blew the glass for the vacum system and aided greatly with timely advice. I further wish to express my appreciation for the personal interest and cc—eperation given by Professor 0. I. Chapman at all times during my work. I am indebted to Professor I. E. Layeock for the help that he offered in connection with the photographic work. INTRODUCTION In an attempt to uncover mother nature's secrets man has only a few tools with which to work. Some of the tools merely probe the cloak of mystery that surrounds the structure which he is studying. Other tools blow the structure to pieces and then he can watch it build up again; noting as it builds up the laws that govern its behavior and fitting a.physical concept that will obey these laws. One of these tools is the low voltage are. A device in which man can shoot high velocity bullets at atoms and find what happens when they strike. As an electron, the shooting bullet, hits an atom, the atom sends out messages called radient energy. lhen we can preperly interpret these messages we will be a long way towards the end of our task of uncovering the secrets of mother nature. It is the pur- pose of this investigation to make a study of colliding electrons, ions, and atoms. 3. PART I IMPACTS OF THE FIRST KIND The term “impact of the first kind' applies to collisions between electrons and molecules and collisions between two molecules when there is a transfer of kinetic energy into atomic energy. The most common method by which electrons can obtain energy is by falling through an electric field. The electron will acquire kinetic energy equal to imvz . ev where e is the charge on the electron and V is the potential difference through which the electron falls. If the electron collides with a molecule, while paesing through space and has not gained enough energy to place the molecule in one of its excited states, the electron will lose practically no energy and the collision is said to be an elastic impact. If, how~ ever, the electron has sufficient energy to lift the molecule into one of its excited states, then the electron loses the necessary energy and goes on with a diminished velocity. Such impacts are inelastic and are the most common impacts of the first kind. A molecule which has been excited or put into a higher energy state will usually dissipate this energy in the form of electromagnetic waves or radiation, as it re- turns to its normal configuration. The energy going out 3. is equal to hv where h is Planck's constant of radiation and v is the frequency of the emitted light. IMPACTS OF THE SECOND KIND 'Impacts of the second kind' is a term first used by Klein and Roseeland1 to describe a collision which is Just the reverse of an impact of the first kind. Klein and Rosseland show that, in a region where we hate impacts of the first kind which result in excited molecules and slow moving electrons, it is possible to have a collision between these two in such a.manner that the excited molecule returns its energy to the electron. Franokz extended the meaning of the term impact of the second kind to include impacts between excited mole- cules and neutral molecules. This impact can explain some cases of fluorescence, and.photo chemical processes. Cario3, working under Franck, discusses two exper- iments in which he has used the term impact of the second kind. Both of these experiments are dealing with impacts between excited molecules and neutral molecules. The first experiment is of interest to us because he was using a.mix- ture of the same two elements that were used in this invest- igation. He had a.mixture of argon and mercury. He excited 4. the mercury,by flooding it with monochromatic light of wave- length 2536.? angstroms. This wayelength is one of the- resonance lines and will place mercury in an excited state. The light from this mercury vapor was then analyzed by a spectrograph and it was found that the emitted radiation was weakened in the presence of argon. The article states that this proves that all of the energy of excitation may be taken up as kinetic energy of translation. It must be remembered that argon does not have an energy level equal to or less than the energy level which he was using in mercury. In the second experiment Cairo mixed mercury and thallium vapors and analyzed the radiation from these vapors in a spectrograph. Besides the strong mercury line 3536.7, he obtained six thallium lines. The energies of all six of these lines were less than 4.9 volts, which corresponds to the mercury resonance line 2536.? angstroms. This experiment shows that energy can be transferred from an excited molecule over to a neutral molecule haying lower energy levels and then be emitted by radiation. POSITIVE ION IMPACTS In a region where electrons possess energy sufficient to tear an electron completely away from the molecule, they are said to ionize the molecule. In this region we hare ions, 5. fast and slow moving electrons, and neutral molecules. Ihen an ion collides with a free electron, the atom goes through a series of decreasing energy values and upon each change in energy the difference goes out in the form of radiation. The composite radiation is known as the are spectrum.' The other possibilities for collisions fall into two classes: 1. Impacts between positive ions and molecules of the same kind, or molecules with larger ionizing potential. Out of such a collision very little can happen. 3. Impacts between positive ions and molecules with smaller ionizing potential. It is this class of impacts with which this investigation is chiefly concerned. IMPACTS BETWEEN SLOW NOVING POSITIVE IONS OF GREATER IONIZATION POTENTIAL WITH HOLECULES OF SMALLER ENERGY OF IONIZATIOH Franck‘ discusses an impact between a slow moving mercury ion with energy equivalent to 10.3 volts with a ceasium molecule which only requires energy carried by an electron after falling through 3.5 volts to ionise it. In this case he suggests that the ion could lose 5.4 volts of energy of the 10.3 volts and be left in the excited state. In the loss of this 5.4 volts, ceasium would use 3.5 volts 6. in ionization and Franck says that the remaining 1.9 volts of energy goes into kinetic energy of translation. Further, this excited atom of mercury possessing only 4.9 volts of energy could collide with another ceasium molecule and ionise it, using 3.5 volts and dissipating 1.4 volts into kinetic energy of translation of the two molecules. Further proof that rare gas ions can rob other neutral atoms of an electron is given by an article written by Smyth and Hornwells. In this experiment they used a Demeter positive ray apparatus with a mixture of rare gases. As a result they stated that the rare gas ion of helium would rob the neon of an electron. At the same time Smyth and Hornwell were working at Princeton on their experiment, there were two other men working in the University of California, by the names of Hogness and Lunn. Hogness and Lunn6 published their work and reported that their results checked those of Smyth and Hornwell. very shortly O. 8. Duffendack of the University. of Hichigan conceived the idea that when a rare gas ion robbed another atom of smaller ionizing potential, that the excess energy, if large enough, might go to excite the first spark spectrum of the atom that is being robbed. It should be called to mind at this place, that the composite light emitted by an electron configuration in which all of the 7. electrons are present is known as the arc spectrum. The light emitted from an electron configuration in which one electron is missing, and the energy changes take place in the remaining electrons, is called the first spark spectrum. If we obtain light from a configuration in which two electrons' hare been removed from the original nucleus, this series of wavelengths of light is called the second spark spectrum. Dr. Duffendack's idea relative to an impact between a neutral atom and a.positive ion possessing sufficient energy to ion- ize and excite the first spark spectrum, can be easily proven or disproven. The critical potentials and energy levels of the various elements are fairly well known. By careful selection from these different elements we can choose two elements, one of which when ionized will possess enough energy to ionize the other and change the remaining electron configuration. Black? of Michigan State College, in connec- tion with Duffendack has recently published an article in Science which contained the results of an experiment using mixtures of neon and capper, and neon and mangenese. The results of this experiment show very clearly that when a neon ion collides with a capper or mangenese atom that the neon will rob it of one electron and shift the remaining config- uration into an excited state which upon returning to the normal state emits the first spark spectrum. Smith and Duffendacke also hate performed work of a similar nature and 8. have given proof that positive ions upon collision with atoms of smaller ionizing potential will transfer energy in the above manner. The question now Opens up: what would be the result if we should choose two elements; the ionizing potential of one being greater than the other but the excess energy not being sufficient to excite the first spark spectrum? Franck, Duffendack, Smyth, Hornwell, Hogness, Lunn, and many others have all predicted that the excess energy would go into kinetic energy of translation of the two molecules. It is the purpose of this investigation to find out what becomes of this energy. For this purpose we selected two easily accessible elements, argon and mercury. The ionizing potential of argon is about 15.2 volts and that of mercury is 10.398 volts. The difference between these two is 4.808 volts of excess energy. This is insufficient to excite the first spark spec- trum of mercury so that the energy could not be used in this manne r . THE NECHANISM OF IMPACT If we can picture an atom of argon which has lost one electron and which by induction is distorting the fields about a.normal mercury atom and making a dipole of it, we can see how an ion can attract a neutral body. As we reason about 9. the matter we can see that there are several possible ways for these two bodies to collide and for the argon to rob the mercury of an electron. The simplest way in which we can picture the argon robbing the mercury of an electron is to hare the argon exert a greater force on the nearest outer electron than it exerts on the other electrons. This greater force will pull the electron away from the mercury while the two atoms are at a relatively great distance. There will be an instant in which the electron is in neither the mercury atom nor the argon atom and we will have two ions repelling each other. Some energy must be spent in the throwing of these two ions apart. When we have an electron that does not belong to any particular atom we call this a free electron. Ihenever a free electron is captured by an ion and moves in to the normal state the arc spectrum of the element is given off. In this mechanism Just described there is an instant in which the electron is a free electron and when it takes its place in the argon atom it should emit the argon arc spectrum. But from thermodynamic considerations we see that this type of collision is impossible. Because at the instant the electron is a free electron we hare stored in the system, 15.2 volts in the argon, 10.393 volts in the mercury, plus an additional unknown amount of kinetic energy in the two masses, making a total of over 25.798 volts of energy. This amount of energy is more than we started 10. with and there is no source from which to obtain this energy. A second possible series of events that might take place in the chamber would be for the argon ion to pick up a free electron which it might get from the chamber walls or in space. Then as this electron made its way in to the normal level, it would emit a quantum of energy. This radi- ant energy would be sufficient to ionize the mercury atom and therefore we would get the transfer of energy from the argon over to the mercury by Photo-electric action and any physical impacts as we have been discussing would not hate to occur. The excess energy referred to above, would probably be tied up in some unabsorbed argon radiation and in kinetic energy of the electron ejected from the mercury. The photo- electric effect will be almost impossible to eliminate be- cause we cannot remove all the free electrons from the collis- ion chamber. Still a third way might be thought of as a.possible means of exchanging this electron and energy. This would be a collision in which the atoms would come very close together and the electrons of the mercury would intermingle with the electrons around the argon nucleus. If this is the case, at some time or other, an electron originally associated with the mercury will find itself located with respect to all the other electrons in a position that is not permitted by the argon configuration. The argon will then have 4.808 volts of 11. excess energy. This energy will be given off in one of two ways; 1. As kinetic energy of translation. 8. As radiant energy. If the excess energy should go into kinetic energy, it is shown in the computations, that the velocity of the molecules would have to increase by 1.99 x 105 cm/sec. which is rather unreasonable. 0n the other hand if the energy was to be rad- iated a wave 2551 angstroms long would be sent out. PART II THE LOW VOLTAGE ARC In the early development of the study of atomic structure glow discharges or geissler tubes were used. This type of discharge is unsatisfactory and very difficult to control or interpret. Franck and Hertz9 in 1914 first pub- lished articles on critical potentials and inelastic impacts, describing at the same time the low voltage are. Franck was given the Nobel Prize for this work which was a direct test of Bohr's theory of atomic structure. In the low voltage are conditions can be more easily and definitely controlled and measurements more easily taken. This is the type of discharge that was used for this experiment. 12. The cathode or negative terminal is a hot filament, the filament consists of a 18 mil tungsten wire coiled in 3 or 4 turns of 3 m.m. diameter. The electron emission from this hot filament may be controlled by the current through it. The anode can be of any design necessary to meet the needs of the particular experiment. They can, however, be placed in two general classes, 1. Anode in the form of a.plate or flat surface. 2. Anode in the form of a hollow box with one or more sides made of gauze. This produces a constant potential chamber. In this eXperiment it was only necessary to use an anode of the first type. The speed of the electron is controlled and hence the kinetic energy is determined by the potential be- tween the anode and the cathode or plate and filament. Therefore we have a large stream of electrons flowing between the filament and plate and we can control the approximate number of these and the energy carried by each. As these electrons travel across this space between the filament and plate, which is from 4 to 7 m.m. long, they encounter numer- ous impacts with molecules such as we have just described. If the potential through which the electron has fallen is in- sufficient to give the electron energy enough to lift a mole- cule to one of its excited states the impact is an elastic impact. If, however, the electron.possees sufficient energy to lift the molecule to a state of higher energy, the impact H assault mask. efine.