Volume 21, Issue 1-2

Owing to their increased dispersion magnetic analysers with inhomogeneous magnetic fields may be useful for various applications e. g. nuclear spectroscopy, mass spectroscopy, and isotope separation. This inhomogeneity may be aquired by conically shaped pole faces. Boerboom and Tasman gave an analytical treatment of these fields, which allows the calculation of aberrations for particles along the medium radius. For some applications e. g. simultaneous collection of more isotopes in an isotope separator it would be important to know how the quality of focus varies with increasing distance from the mean radius. In this paper a numerical method is described which allows a three dimensional ray-tracing throughout inhomogeneous magnetic sector fields. Some of the results have been published elsewhere 1 .

The matrix representation for calculating the ion trajectory is extended to include the secondorder aberration terms. Properties of the matrix and relations between its elements are discussed. As an example of the application of the matrix method, the second-order coefficients of a homogeneous magnetic field of arbitrary shape are derived and the relations between the matrix elements and Hinterberger's notations are shown.

The most important advantage of the Mattauch—Herzog type mass spectrograph is double focusing for all masses. One disadvantage, however, is the fact that the energy slit cannot control the velocity spread (β) independently of beam divergence (α). This disadvantage is removed by substituting a two-stage electrostatic field for the usual single one. General formulae for determining the distances between the elements of the optical system are derived. Combinations of two cylindrical electrostatic fields with equal radii are chosen as a practical example. The condition to be fulfilled for a physically significant solution and the resolving power of this system are discussed. The study of the second-order aberrations shows that α 2 focusing for all masses can be achieved under suitable conditions. In addition, α β and β 2 aberrations can be made to vanish simultaneously at a point on the focal plane. The design parameters are numerically computed and tabulated for several favorable examples. The total image defect is calculated for a typical example, and found to be very small for a wide range of masses.

A new type of mass spectrograph with large mass dispersion has been constructed. The distinguishing feature of this apparatus is that the mass dispersing action of a r -l magnetic field is utilized. The instrument has a r -1 magnetic field of 198.1° sector type (22 cm mean radius) as the dispersing field, a toroidal electric field of 118.7° sector type (30 cm mean radius) and a 30° uniform magnetic field (120 cm mean radius) as the focusing fields. The principle, the design and important parts of the apparatus are described. The dispersion on the photographic plate was estimated to be 14 cm for 1% mass difference and the maximum resolving power of about 500,000 was obtained.

Part I. The transparency curve, produced by ions of 10 - 100 keV, can be described by T=T s +(1-T s ) T = transparency, T s = transparency at saturation, e = photographic sensitivity, N = number of ions incident per unit area, V = parameter of the size-frequency distribution of the AgBr grain impact areas. The equation is derived from basic principles by suitably modifying Kinoshita’s assumptions 1 . Agreement with experimental curves is demonstrated. Part II. The shape of the transparency curve (T vs. log of ion density) was investigated for several developers at various developing temperatures and developing times, for the improved Ilford Q2 emulsion. Plates were exposed with 28 keV Kr+-ions. For short developing times, only those grains are developed which have been struck by r and more ions, r in the order of 50. With increasing developing time, r tends towards 1. This statistical effect, observed for all developers, is superimposed by two developer-dependent effects: in regions of high density of developable grains, the developing action is hampered by local consumption of developer reagents; furthermore, some developers result in a superposition of curves of very different values of r: the transparency curves appear “flatter” than those demanded by theory. Part III. The consideration of background fog necessitates the differentiation of two types of background: a fog due to light quanta or plate storage, and a surface fog due to particles with limited penetration depth into the emulsion. Equations are derived correcting measured transparencies for both types of background. Background fog reduces the photographic sensitivity. A theoretical equation gives the loss of photographic sensitivity for ion-produced fog. A similar equation is valid for light- and plate storageproduced fog; it then contains a parameter which has to be determined experimentally. Agreement with experiment is demonstrated. Part IV. Grain size and grain density of Ilford Q2 emulsions are determined from photometric measurements of statistical fluctuations of the blackening for various developing conditions. For most developers, the grain density increases with increasing developing time, the grain size remaining constant. The maximum attainable precision of photometer transparency and ion current density measurements for a given line area is stated.

During the past year, certain of our experimental techniques have been re-evaluated and modified. In the course of these investigations, we have re-determined two atomic mass differences involving isotopes of cadmium and one involving isotopes of lead. The new values are consistent with reaction data and with our older measurements. Thirteen mass differences involving isotopes of neodymium and samarium have also been determined, several for the first time. These new results are consistent with reaction data and with other recent mass spectroscopic data, but indicate that our previously-reported mass differences in the neodymium-samarium region are systematically too large by small but real amounts. An unusually concordant situation now exists in this region of the mass table.

Very probably, the masses of all nuclides decaying by an α— β decay chain into 213 Bi are low by about 150 keV.

The dilemma is described which exists at the present time between the two present best sources of information as to the numerical value of the Sommerfeld fine structure constant, α. These two sources are the fine structure splitting in deuterium, determined in 1953 by Triebwasser, Dayhoff and Lamb, and the hyperfine structure splitting in hydrogen, measured more recently using the Ramsey hydrogen maser. The theoretical connection between the fine structure measurements and α is subject to little question but the experimental difficulties to obtain a precision of a few ppm are considerable. The relative precision obtained with the hydrogen maser on the other hand, is phenomenal (of order 10 -11 ) but the theoretical connection between the hyperfine splitting and α is subject to a controversial correction for the internal field structure of the proton. Assuming this correction term to be correct at its present value, the hf splitting in Η implies a value of α 26 ppm higher than the fs splitting in D. Present existing sources of evidence, some favourable to the lower and some to the higher value of α, are presented and discussed and the key importance of a better knowledge of this fundamental constant is stressed.

The theory of decaying systems of FRANZ 1 — which seems finally to settle the basic questions left open so far—has been reformulated assuming a slightly altered initial situation. The currently used formulae for the decay rates (see MANG 2 ) have been derived as an approximation to this practically rigorous theory.

In order to determine the half-life of the 130 Te double β-decay, the amounts and isotopic composition of xenon extracted from tellurium ores, from the Oya gold mine in Japan, have been measured with a high-sensitivity mass spectrometer. Compared with atmospheric xenon an excess was definitely found at mass numbers 129, 130 and 131 in the extracted xenon. The excess of 130 Xe is predominant, the average amount in three samples being (1.32 ± 0.09) × 10 -11 ccSTP/g 130 Te. Attributing the excess 130 Xe to the double β-decay of 130 Te, the half-life is estimated to be (8.20 ± 0.64) × 10 20 years, assuming an age of (9.06 ± 0.29) × 10 7 years for the Te ores. The latter value is the K-Ar age of porphyrite, which is in close geological connection with the Te ores. In order to investigate the other excesses than that of 130 Xe, isotopic analyses were also carried out on Xe from three other Te ores from the same mine. The ratios ( 129 Xe/ 131 Xe) excess=1.58 and ( 129 Xe/ 130 Xe) excess = 2.1 were found to be the same for all samples. The origin of these excesses is discussed. In addition a small excess of 128 Xe was found. If this is attributed to 128 Te double β-decay, the half-life of 128 Te is estimated to be 3 × 10 22 years, a value shorter by about three orders of magnitude than the theoretically expected half-life. The above estimated half-life may be a lower limit of the 128 Te half-life. The general tendency of the isotopic abundances (except for the above excesses), of the xenon extracted from Te ores seems to be to slightly increase in excess as one moves toward the lighter isotopes (as compared with atmospheric xenon).

Im allgemeinen hält man die Schwierigkeiten, die den Theorien der Elemententstehung anhaften, durch die Annahme für überwunden, daß die Elemente unseres Sonnensystems Mischungen darstellen von Kernsorten, die auf verschiedene Weise, zu verschiedenen Zeiten und an verschiedenen Orten entstanden sind. Gegen diese Vorstellung läßt sich jedoch u. a. ein grundlegender Einwand erheben : Durch Wahl einer geeigneten Klassifizierung der Kernsorten läßt sich zeigen, daß in bestimmten Massengebieten die Kernhäufigkeiten mit dem Gang der Kernbindungsenergien nach MATTAUCH et al. 1 , zum Beispiel mit den Bindungsenergien der letzten α-Teilchen, in einem, wenn auch derzeit nicht quantitativ deutbaren, augenfälligen Zusammenhang stehen.

Knowing the cosmic ray exposure ages of a sufficiently large number of meteorites and using the earth as analyser with special assumptions, criteria can be found to distinguish between a lunar or asteroidal origin of meteorites. Several of the following arguments are based on new and unpublished results of rare gas measurements by HINTENBERGER, SCHULTZ und WÄNKE 70 . Bronzite-chondrites : 1. Arguments for an origin near the surface of the parent body. a) Porosity of the chondrites 0—20%. b) Many bronzite-chondrites contain light primordial rare gases, originating from the exposure of the single meteorite grains to the solar wind. c) Primordial rare gas content always connected with light-dark structure. d) In the distribution of the cosmic ray exposure ages certain groupings can be distinguished. The age distribution of bronzite-chondrites with light primordial rare gases is identical with the distribution of the cosmic ray exposure ages of all bronzite-chondri-tes. The bronzite-chondrites containing primordial gas therefore are probably coming from the very upper layers, and the other bronzite-chondrites from somewhat deeper layers of their parent body. 2. Arguments for an origin close to the earth’s orbit. a) Bronzite-chondrites with high cosmic ray exposure ages show a slight tendency to fall in the afternoon (noon until midnight) . b) For the bronzite-chondrites, which are morning falls (midnight until noon), diffusion losses of 3 He and 4 He are higher and more frequent compared to the afternoon falls. The reason for this can be found in a closer approach to the sun of the first ones. Hypersthene-chondrites do not show this effect. c) Bronzite-chondrites with light primordial rare gas content concentrate among the afternoon falls. d) The mean cosmic ray exposure age of the bronzite-chondrites is considerably lower than that of the hypersthene-chondrites. 3. Arguments concerning the size of the parent body. Light primordial rare gas and their connection with light-dark structure indicate a parent body of the size of the moon or a large asteroid. None of these arguments are strictly conclusive. In some cases they are based on observations, which can only be obtained by using statistical methods. Most of these effects are close to the mean error. Adding, however, all observations together, a lunar origin of the bronzite-chondrites becomes nearly undoubtable. A lunar origin of stone meteorites was in recent times first proposed by URET 3 . Hypersthene-chondrites : Hypersthene-chondrites with low cosmic ray exposure ages are rare among the morning falls. Their parent body therefore probably has to be found outside the earth’s orbit. Their distribution of the cosmic ray exposure ages may also lead to this conclusion. As proposed by ANDERS 4 , the Mars asteroids could possibly be the parent bodies for the hypersthene-chondrites. Mars itself might however be considered also. A lunar origin of the hypersthene-chondrites seems to be completely out of question.

The production rates of 36 Cl and 39 Ar by the cosmic radiation on Fe and Ca +K have been determined in the chondrite Leedey. The values of 19.4 atoms/min · kg ≦ P Fe ( 36 Cl) ≦ 26.6 atoms/min · kg and P Fe ( 39 Ar) = (23.6 ± 1.8) atoms/min · kg are in good agreement with those found in iron meteorites of comparable size. A large contribution of Ca+K to the production of both isotopes is indicated by the ratios 9.2 ≦ P Ca + K ( 36 Cl) /PFe ( 36 Cl) ≦ 13.6 and P Ca + K ( 39 Ar) /P Fe ( 39 Ar) = 12.1 ± 1.6. The bearing of these results on the interpretation of 39 Ar-data obtained on other chondrites is being discussed.

“Absolute” values for the isotopic abundance ratios of atmospheric neon have been determined by mass spectrometry. Calibration standards were prepared by mixing of nearly pure separated nuclides of neon. The percentage abundance of the three nuclides of neon are: 20 Ne = 90.514 ± 0.031, 21 Ne = 0.266 ± 0.005, 22 Ne =9.220 ± 0.029. The calculated atomic weight of neon on the unified physical and chemical atomic weight scale ( 12 C=12) is 20.1794 ± 0.0030.

The K-Ar-ages of five ultrabasic rocks from the Baltic Shield were determined. K-Ar-ages of 3.7 to 8.8 x 10 9 y were obtained. Two samples were cut into fractions with different potassium content and a partial correlation of the argon content with the potassium content in the separated fractions was noticed. The question of extraneous argon causing higher age values is discussed. Taking into consideration the previous results obtained by GERLING et al., it may be said, that extremely high Ar 40 /K-ratios are the rule for ultrabasic rocks from the Baltic Shield. As a consequence of this fact, it is not likely that all these rocks have the age only of the precambrian pluton in which they are incorporated. Assuming that the alpinotype peridotites were transported upwards in the form of solid fractions from the upper mantle, it is possible, that actually radiogenic argon exists more or less in the sample. The highest Ar 40 /K-ratios (according to ages ranging from 5.9 to 10.9 x 10 9 y) exhibit the samples with potassium contents below 100 ppm. Samples with potassium contents more than 1‰ have much smaller Ar 40 /K-ratios (according to ages from 4.2 to 5 x 10 9 y) . Excess argon probably causes the extremely high “ages” of more than 7 x 10 9 y. The ages from relative potassium-rich samples will eventuible show the true age. The most likely value for this age is 4.2 x 10 9 y, measured for a peridotite with 0.49% potassium content.

By direct comparison of the signals of a particle multiplier and a Faraday cage at the collector end of a mass spectrometer, the secondary emission coefficients of a copper-berilium alloy were determined for single and multiply charged noble gas ions. The ions involved were He + and He 2+ , 20 Ne + up to 20 Ne 3+ , 22 Ne + , Ar + up to Ar 5+ , Kr + up to Kr 7 + and Xe + up to Xe 9+ . The energy of the ions varied between 3 and 90 keV. The secondary emission factors show an almost exact linear dependence on the velocity of the ions, with a threshold at 5.5 × 10 6 cm/s. The slopes are proportional to the square roots of the ion masses.

The electronic stopping cross sections in boron for atomic projectiles with Z≤ 11 have been determined in the energy interval 15 to 140 keV. Reasonable agreement is found with theory, however the previously observed periodic dependence of S ε on the atomic number of the projectile is also evident. Results for the relative straggling in energy loss are reported for hydrogen projectiles in boron, carbon, and aluminium targets and for helium projectiles in boron and carbon. Theoretical straggling estimates agree reasonably well with the experimental results.

A mass spectrometric investigation was carried out on the direct photolyses of propene, 1-butene, and hydrazine at 1849 A with a field ion source in a flow system. Comparisons were made with Pt tip and wire emitters. It was found that, without illumination, mass spectra obtained with the wire were accompanied by a number of fragment peaks amounting to almost 1%. Since these peaks interfere with those produced photochemically, the tip emitter was used mostly for the photochemical studies although it gave 100 times less current and was less stable. The photochemical products formed at a gas pressure of 10 μ by a low pressure mercury lamp were detected after approximately 10 m sec. The three main peaks observed in the propene photolysis were at masses 27, 28, and 56, indicating the processes: C 3 H 6 +hv→C 2 H 3 +CH 3 , C 3 H 6 +hv→C 2 H 4 +CH 2 , CH 2 +C 3 H 6 →C 4 H 8 . The photolysis of 1-butene gave four main peaks at masses 40, 41, 42, and 70, suggesting steps, C 4 H 8 +hv→C 3 H 4 + (H+CH 3 ) or CH 4 , C 4 H 8 +hv→C 3 H 5 +CH 3 , C 4 H 8 +hv→C 3 H 6 +CH 2 , CH 2 +C 4 H 8 C 5 H 10 . The only peak found with the photolysis of hydrazine was at mass 17, indicating the step, N 2 H 4 +hv→NH 3 +NH. The possibility of forming these products by secondary processes is discussed.

The angular distribution of atoms and molecules field ionized at Platinum tips is determined. First, with negative polarity of the tip the angular distribution of the field emitted electrons is measured in order to determine he radius of curvature and the shape of the tip, using calculations of DRECHSLER and HENKEL 2 . Then the angular distribution of field ionized Argon is measured at a tip temperature of about 400 °K. The curve agrees qualitatively with one derived from GOMER’s 6 equations on field ion currents as a function of field strength. In contrast with this curve which shows a continuous decay with polar angle, the angular distribution curve of field ionized n-hexane, acetone, water and methanol shows a side-maximum lying at a polar angle of about 40°. This is explained as the consequence of adsorption and particle supply phenomena.

The dynamic polarization of the 19 F nuclei in free radical solutions of C 6 F 6 , C 6 H 5 F and C 6 H 4 F 2 is much less negative than expected from investigations of the proton polarization in similar systems. In the limits of “extreme narrowing” the enhancement factor is of the order of —100 instead of —350; in strong magnetic fields it is even posiitve. Data obtained in fields of 16.2, 176 and 3230 gauss are interpreted in terms of a combined dipole-dipole and scalar coupling between nuclear and electron spins. For the time dependence of the dipolar interaction the model of an independent "translational" random diffusion of spin-carrying molecules is used whilst the scalar interaction is described by a spectral function proportional to τ Sk /(1+ω 2 τ Sk 2 ). The results are compared with measurements of RICHARDS in strong magnetic fields and with theoretical models of HUBBARD for the scalar interaction. Diffusional and scalar correlation times, the relative contribution of the scalar interaction and the order of the coupling constant are calculated.

The dynamic proton polarization, and the proton and electron spin relaxations have been studied in the organic polycrystalline free radicals bis-diphenylene-p-chlor-phenyl allyl (BPA), diphenyl picryl hydrazyl (DPPH), WURSTER'S blue perchlorate (WBP) and galvinoxyl (GAV). Enhancement factors between +370 and +79 were observed; extrapolation to infinite microwave power yielded values near the theoretical γs/γI ratio. Electron spin-lattice relaxation times are of the order of 10 -7 to 10 -8 sec; proton relaxation times are of the order of 10 -3 sec. With the exception of galvinoxyl at low temperatures, the proton resonance line shapes are not much influenced by the local field of the unpaired electrons. The ESR spectra show a typical exchange-narrowing. All the results are discussed in terms of a scalar nuclear-electron interaction in which electron-electron exchange couplings provide a time dependence of random character.

The decomposition of equimolar NaCl—KCl mixed crystals was studied at different temperatures with calorimetric and capacitance measurements. Comparison with other studies indicates that the decomposition proceeds in three steps. The main heat emission and change of the dielectric constant take place during the second step. The activation energies of the steps are nearly equal.

Die im Bereich von 180°—220 °C gelegene Thermolumineszenz-Emission des NaCl wird für Ca, Sr, Ba, Cu, Fe, Mn und Cr-dotierte kristalline Proben untersucht. Insgesamt sind die erhaltenen Spektren mit Ausnahme des NaCl : Mn uncharakteristisch für die jeweiligen Aktivatorzusätze. Diese Messungen zeigen eine Analogie zu den Ergebnissen vergleichbarer Untersuchungen an natürlich thermolumineszierenden Flußspatmineralien. Auch hier sind Anzahl und Lage der auftretenden Banden weitgehend unabhängig von der Art der eingeschlossenen Fremdionen.