The level structure of ¹²⁹Cs was investigated by measuring gamma-rays and internal conversion electrons following the decay of ¹²⁹Ba with a Ge(Li) detector and a \sqrt2π air core beta-ray spectrometer. About one hundred fifty gamma-rays were observed. The L-subshell ratios of the 6.5, 53.2, 129.1, 182.3 and 202.4 keV transitions were measured, and the multipolarities and the mixing ratios of these transitions were determined. For the 6.5 keV transition, zero energy loss peaks were observed. The low-lying level energies, spins and parities are as follows: 0(1/2+), 6.545(5/2+), 135.6(3/2+), 188.87(7/2+), 208.9(5/2+) and 220.8 keV(3/2+). The experimental results of l-forbidden M1 transitions are compared with the theoretical predictions.

View full abstract

The transitions to the ground state and the first 2⁺ state in the reaction ¹⁰⁰Mo(t, p)¹⁰²Mo were calculated in the coupled-channel Bornapproximation (CCBA). The calculations of the microscopic spherical phonon model and the deformed medel were carried out for the structure of the nuclei ¹⁰⁰Mo and ¹⁰²Mo. The inelastic process was treated macroscopically for both models. The results of the two models were compared. Concerning the ratio of the cross section of the ground state to that of the first 2⁺ state, the spherical phonon model is preferable to the deformed model. In order to see the change of angular distribution and magnitude of the direct and indirect transitions to the first 2⁺ state due to the variation of the coupling parameter and to examine the validity of the DWBA treatment of the inelastic process numerically, model calculations were performed with the form factor obtained by assuming two-neutron (g_7⁄2)² configuration in the Woods-Saxon well for the states of ¹⁰²Mo.

View full abstract

Angular distribution of tritons in (d, t) reactions induced by 21.4 MeV deuterons on ⁸⁸Sr, ⁸⁹Y, ⁹⁰Zr and ⁹²Mo targets was measured to obtain the l-values and the spectroscopic factors of transitions. In the reaction on ⁸⁸Sr, all of the expected spectroscopic strengths for T_< states are found for the transitions leading to 1g_9⁄2, 2p_1⁄2, 2p_3⁄2, and 1f_5⁄2 shell-model states. In contrast with this situation, severe fragmentation of the strength as a function of proton numbers was found in the other isotones. Any apparent spectra for 1f_7⁄2 neutron shell-model orbit have not been observed so characteristic as for 1g_9⁄2, 2p_1⁄2, 2p_3⁄2 or 1f_5⁄2 orbits.
The shell-model forbidden l=2 transitions to (2d_5⁄2)⁻¹ states were observed in the ⁸⁹Zr and ⁹¹Mo nuclei. From these strengths the amount of shell-model impurities in the ground-state correlation of the target N=50 nuclei is estimated as a function of proton numbers.

View full abstract

The photoneutron cross section of ²⁹Si has been studied by irradiating silicic anhydride, enriched in ²⁹Si, in the X-ray beam of a 15 MeV linear electron accelerator. The photoneutron cross-section shows maxima at about 10.6 and 12.2 MeV. The integrated cross-section form the threshold to 13.0 MeV is 11 MeV·mb. The angular distribution of photoneutrons has also been measured to fit with a curve of the form a+bsin²θ. The values for b⁄a are found to be 2.1±1.0 and 3.9±0.8 for measurements with bremsstrahlung having maximum energies of 11.7 and 12.9 MeV, respectively. The significant features of the photoneutron cross-section and the angular distribution are discussed and compared with theoretical calculations for the E 1 transition of a valence neutron.

View full abstract

The energy for the ground state and the three first excited states for one F center perturbed by a negative interstitial halide ion in KCl, KF, Kl and KBr is given. Using this results an approximate calculation was made for the ratio of the non-radiative to radiative transition of F center at room temperature. It is founded that in an approximate way an F center is not stable if a negative interstitial halide ion is within six lattice sites of the vacancy.

View full abstract

Theoretical study of Mizuno and Ishikawa on emission spectra is extended beyond the behavior precisely at threshold. Following them, the perturbation series of the spectral density functions are calculated up to the third-order term for L₂₃ spectra while up to the fourth-order for K, the perturbation being a static potential caused by a localized inner-shell hole that is present only before emission. In the present paper, however, the part, which goes to zero at threshold, of each of the terms is also calculated out explicitly without approximation.
After rearrangement of the obtained ill-behaved perturbation series, it is found that spectral densities below threshold may be expressed in the form of a generalized power law:
w(ω)=γ(ω)|ω_F−ω|^α(ω).
Here, ω is a frequency of emitted photon, ω_F being the value at threshold. Both α(ω) and γ(ω) denote certain well-behaved functions which are regular in particular about ω=ω_F.

View full abstract

Paramagnetic susceptibilities, χ_i(i=a,b,c) of single crystals of rare-earth orthoferrites RFeO₃ (R=Nd, Sm, Gd, Dy, Er, Yb and Y) were measured along the a, b and c axes at room temperature and 77 K. The main parts of the effective magnetic moments μ_eff,i’s were found to originate from rare-earth ions. All rare-earth ions except Dy³⁺ and Er³⁺ showed rather isotropic effective magnetic moments, which are nearly equal to those of the free ions. Er³⁺ and especially Dy³⁺ ions showed large anisotropy and the anisotropy increases with decreasing temperature. For Dy³⁺ μ_eff,b>μ_eff,a>μ_eff,c and for Er³⁺ μ_eff,c>μ_eff,a>μ_eff,b. The values of μ_eff,i for Dy³⁺ and Er³⁺ were obtained by using reported g-factors derived from optical experiments and the observed large anisotropy was found to result from the anisotropy of g-factor of the ground state.

View full abstract

Ferromagnetism and ferromagnetic spin waves in a magnetic crystal such as doped LaMno₃ are investigated by using the Hamiltonian derived in the first paper of this series. A close resemblance is pointed out between the present model and the strongly correlated, almost half-filled Hubbard model studied previously by Nagaoka. The results obtained are as follows: for simple cubic (s.c.) and body centered cubic (b.c.c.) structures the ferromagnetic state with the maximum total spin is stable for excitation of a spin wave, while for face centered cubic (f.c.c.) structure the ferromagnetic state with the maximum total spin is not the ground state.

View full abstract

Electron spin resonances in the hexagonal layer type compound NiCl₂ in the ordered state have been studied in detail by using microwaves of 10∼70 GHz regions. Considering the resonance data together with other facts, a possible model for the spin arrangement in NiCl₂ in which spins lie in the c-planes by making adjacent ferromagnetic layers which alternate in direction along the c-axis is obtained. The temperature dependence of the resonance point (H₀⁄⁄c-axis) can not be explained by the molecular field theory. A spin-wave calculation which takes into account a two-dimensional property in this salt gives the following temperature dependence of the resonance frequency,
(Remark: Graphics omitted.).
A quantitative agreement between the theory and the experiment is attained.

View full abstract

The nuclear relaxation of proton has been studied in an antiferromagnet CoCl₂ 2H₂O (T_N=17.2 K) which is characterized by ferromagnetic linear chain of Ising-like spin S=1⁄2. With the external magnetic field applied along the spin axis, the field and temperature dependences of the nuclear relaxation time (T₁) were measured in the three metamagnetic (antiferro-, ferri- and ferro-magnetic) states. The experimental analysis was made by using a simple phenomenological model for the spin fluctuation in onedimensional Ising spin system. It was found that the proton nuclear relaxation in CoCl₂ 2H₂O is closely related to the Ising spin excitation in its ferromagnetic linear chains.

View full abstract

An expression of the effective dipole moment operator which causes intraband transitions accompanied by spin flips is given in the second quantization scheme. The method adopted in this note is the very same as that established by Gondaira and Tanabe. The present expression, however, includes some new terms which arise from the lowest order perturbation processes overlooked by them. It is pointed out that the expression thus obtained, although it is accurate to the lowest order, cannot satisfy the selection rules for the optical transitions correctly. The fault of the formulation on this point is closely related to the yet unsolved problem of the asymmetry of Hartree Fock Hamiltonian for general orbital configurations.

View full abstract

Oxygen-deficient perovskite system SrFe_xMo_1−xO_3−y (2⁄3≤x≤1) was prepared, and the influence of heat-treatment on structural and magnetic properties and also on the valence state of cations were investigated. The lattice size of the system is reduced by air-oxidation of the sample which was prepared in an inert atmosphere of nitrogen gas, and the Néel temperature was lowered remarkably (e.g. from 660 K to about 80 K for SrFe_0.9Mo_0.1O_3−y) by heat-treatment without changing its crystal symmetry. These results are consistently explained by assigning the valence states of the sample before and after the oxidization as SrFe_x³⁺Mo_1−x^5+,6+O_3−y and SrFe_x^3+,4+Mo_1−x⁶⁺O_3−y respectively.

View full abstract

Frozen aqueous ferric perchlorate solutions with low pH were studied by the Mössbauer effect. From the temperature and concentration dependence of the spectra, the spin relaxation of Fe³⁺ ions is concluded to be mainly caused by magnetic dipole-dipole in interaction. Fe³⁺ ions were found to be localized in regions of a constant concentration independent of average concentration.

View full abstract

Electrical resistivity, the Hall and thermoelectric effects and proton nuclear magnetic resonance in the Ta–H system prepared electrolytically were measured in the temperature range between 77 and 300 K. X-ray diffraction analysis showed that these samples have the pseudo tetragonal structure which is slightly distorted to the monoclinic, γ=90° 28′ and γ=90° 40′ at 300 and 77 K. In the low temperature phase, the proton resonance line widths are about 11 Oe, and the diamagnetic shift of the resonance field suggests that hydrogen atoms make the partial covalent bond with surrounding atoms. Above 150 K the motional narrowing in the line width occurs and this result shows the rapid motion of hydrogen atoms and its activation energy E_a is obtained to be 0.144 eV.

View full abstract

The existence range of the one-dimensional long-period ordered alloy Cu₃Pt (L1_2−s) on the equilibrium diagram has been determined by X-ray diffraction, single crystal specimens having been prepared and quenched at various temperatures. As the result it has become clear that there occurs a two-step transition L1₂→L1_2−s→A1 by heating-up in this alloy system, and this transition has been confirmed by specific heat measurements. The half long-period M of L1_2−s phase has been measured from separations of split superlattice reflections. M increases with decreasing platinum content, and the behavior of this concentration dependence is well explained by the relation between M and e⁄a based on Sato-Toth’s idea. M decreases as the temperature approaches T_c, in the same way as in Cu₃AuII but contrary to the behavior in Ag₃Mg. Some argument is given on the temperature dependence of M and the degree of order in the long-period lattice of Cu₃Pt.

View full abstract

The low temperature specific heat of Th–Y and Th–La alloys have been measured in order to compare with that of La–Y alloys. The change of the Debye temperature of Th–Y alloys with Yttrium concentration is very similar to that of La–Y alloys. The atomic property of McMillan-Hopefield’s parameter η is also discussed.

View full abstract

Self-consistent band calculations are carried out for iron in paramagnetic and ferromagnetic states. Wave functions are expanded in terms of a set of functions constructed by the modified tight-binding method and the orthogonalized-plane-wave method. The modified Slater’s approximation is adopted for the exchange term of the crystal potential. Wave functions and crystal potentials are determined self-consistently by an iteration method. Band structures, density of states curves and the Fermi surfaces are obtained and compared with previously reported results.

View full abstract

The photocurrent transient in copper oxinate was measured. Copper oxinate has two photoconductive regions: one of them corresponds to the lowest singlet optical absorption band and the other lies in the low wavenumber side of the same band. Under irradiation, the photocurrent in both wavenumber regions abruptly rises to a maximum, and then it decreases gradually and settles to the steady state. In mixed single crystal (69 mol % palladium oxinate—31 mol % copper oxinate), such decreasing of photocurrent is also observed in photocurrent due to copper oxinate, while it is not observed in photocurrent due to palladium oxinate. The decreasing of photocurrent is not caused by the presence of carriers but caused by the excitation of specific molecule (copper oxinate) in the crystal, and the photoconduction in low wavenumber region is attributed to the excitation of copper oxinate.

View full abstract

The luminescence Br⁻-ions in NaCl single crystals was studied for concentrations of 10⁻² to 10⁻¹ mole fraction of NaBr at liquid nitrogen temperature. With ultraviolet excitation, two emission bands were observed at 4.0 eV and at 5.15 eV. These emission bands are known to be of Br⁻-monomer (4.0 eV) and of Br⁻-dimers (5.15 eV). By an analysis of the excitation spectra for the 5.15 eV luminescence, an absorption band of Br⁻-trimers was identified in the low energy side of the Br⁻-dimer absorption band. Excitation in this trimer band causes the only emission band at 5.15 eV which is known to arise from excitation in the Br⁻-dimer band. Therefore, the relaxed state of localized excitons at Br⁻-trimers is supposed to be identical with that of localized excitons at Br⁻-dimers.

View full abstract

An apparatus for detecting small second and third harmonics due to the dielectric nonlinearity is devised. A reactance filter network is used to obtain pure sine wave voltage and high detection sensitivity. The higher harmonic level applied to the crystal is suppressed to the order of 5×10⁻⁷ times below the fundamental voltage.
This apparatus is used for measuring the nonlinear dielectric constants of BaTiO₃ single crystals above and below the Curie point (120°C). The experimental results show that the coefficient of the P⁴ term in the Devonshire’s equation increases gradually with increasing temperature, which is against the previous result reported by Drougard et al.

View full abstract

The phase diagram is studied by X-ray and thermal analyses. The high temperature phase (α) is the solid solution of b.c.c. or f.c.c. in which Ag⁺, Cu⁺ and electron are mobile. The galvanic cell Ag|AgI|specimen|Pt is available to controle the excess or deficit Ag density and to detect the electrochemical potential of Ag⁺, since the interdiffusion of Cu⁺ into AgI is nearly forbidden except near Cu₂Se. With the help of the cell and utilizing the weight-difference between Ag and Cu, the polarization and the interdiffusion are measured and interpreted with the diffusion theory. The electronic and ionic conductivities are measured with varying temperature and x. Further the various properties of α(Ag, Cu)₂Se such as the conductivity, Hall coefficient, thermoelectric power, thermal conductivity etc. are measured and compared with the theory, indicating that the energy gap decreases with increasing x.

View full abstract

The central NMR spectrum of ²³Na in the ferroelectric silver sodium dinitrite (AgNa(NO₂)₂) is investigated. The second-order quadrupolar-shifts in the single crystals are measured. The rotation patterns about the three crystalline axes in the paraelectric and the ferroelectric phases are analysed by Volkoff method. For the calculation of the EFG, the chemical shift of ²³Na ions as well as the distribution in the EFG must be taken into account. The ²³Na EFG tensors in a unit cell of AgNa(NO₂)₂ are identical in the paraelectric phase. They split into the two kinds of EFG whose principal components are equal in magnitude and differ only in the orientation of the principal axes. The powdered samples show that nonstoichiometry does not affect the electric field gradient seriously. The temperature dependence of the EFG tensor is discussed in comparison with that of NaNO₂.

View full abstract

The emission spectra of phthalocyanine single crystals and evaporated films were investigated in the near infrared. In a H₂Pc single crystal, several fluorescence peaks were observed between 760 mμ∼1 μ. In a CuPc single crystal, one phosphorescence band with fine structures was observed at 1.12 μ. However, no fluorescence was observed between 700 mμ∼1 μ. In ZnPc single crystals, both the fluorescence peaks and the phosphorescence peak at 1.15 μ were observed. In CoPc and NiPc, no emission was observed. These emission spectra were discussed in connection with S–T interaction caused by central metals.
The phosphorescence was decreased markedly with raising temperature. In the evaporated film, a fairly broad emission band was observed which was attributed to the excimer emission tentatively. By the thermal treatment of this film, this broad band changes to the emission band of the single crystal.

View full abstract

The abasorption edge of lead sulphide has been measured over the temperature range from 300 K down to 18 K, using thermoabsorption measurements. Based upon the analysis of the thermoabsorption spectra the absorption edge energy E_g, the temperature coefficient dE_g⁄dT and the broadening factor Γ have been obtained. Typical values of E_g are as follows; E_g=0.290±0.002 eV at 20 K, E_g=0.308±0.003 eV at 85 K and E_g=0.418±0.003 eV at 292 K, respectively. The absorption edge E_g decreases linearly with decreasing temperature down to about 100 K and the temperature coefficient is 5.5×10⁻⁴ eV/deg. The temperature coefficient of the broadening factor dΓ⁄dT is about 0.3×10⁻⁴ eV/deg., less than one tenth of dE_g⁄dT. These values are discussed by comparison previously reported data.

View full abstract

Trapping of hydrogen molecules by M centers in KCl:U crystals has been investigated. The thermal bleaching of the M band converted optically from the F band was observed in the dark at 20, 30 and 40°C, and was explained by kinetic theory on diffusion of hydrogen molecules. The temperature dependence of the diffusion coefficient of molecules is qualitatively in agreement with the theory.

View full abstract

The large angle noncollinear three-phonon interactions have been studied by using longitudinal (L) and transverse (T) pulsed ultrasonics with megahertz frequencies. Two types of process, T[100]+T[\bar110]→L and L[100]+T[010]→L, were observed in the (001) interaction plane of a copper single crystal. The incident ultrasonic beams were approved to be cylindrical in shapes and almost non-diverging. The displacement amplitude of the produced sound was found to be proportional to the product of the amplitudes of two incident sounds, to vary wholly as the cosine of the angle between the polarization direction of T⟨100⟩ incident wave and the interaction plane, being agreed with theoretical predictions. The ratio of the amplitudes of the produced sounds in two types of interaction was disagreed with the theoretical value perhaps due to the difficulty in transmitting the transverse waves into the specimen.

View full abstract

Athermal motion of a dislocation in a lattice is studied. Using Nabarro force which represents the anharmonicity of the potential in the slip plane, a formula which describes the energy dissipated from a moving dislocation is given. Results of computer simulations of a one-dimensional dislocation are analyzed by the formula. It is shown that there are three types of motion: (i) a dislocation emits lattice waves, (ii) the displacement field around a moving dislocation is almost the same as the static one, (iii) dislocation velocity changes remarkably in the period of the lattice parameter. The boundaries between velocities of these types of motion depend strongly on the shape of the potential and damping of lattice waves. The continuum approximation is justified for the second type of motion which appears if a lattice has large damping and the low Peierls stress.

View full abstract

The temperature dependences of elastic stiffness constant C₂₂ and attenuation coefficient α₂₂ of thiourea single crystals were measured around the IV–V phase transition. It was found that the anomalous parts of the attenuation coefficient Δ α₂₂ and of the elastic stiffness constant Δ C₂₂ are both proportional to log(T−T₀), in agreement with the formula given by Levanyuk et al. for uniaxial ferroelectric materials.

View full abstract

Complex continuum potentials for channeling due to non-local inelastic scatterings are introduced. The relation between the complex continuum potential and the Yoshioka term for electron diffraction is studied. For an example, phonon excitations are considered in detail for numerical calculations of the imaginary continuum potential (absorption coefficient). The local diffusion constant for dechanneling theory is also discussed.

View full abstract

Several comments are given on the diffraction effect of annihilation gamma rays in a single crystal (cf. Hyodo et al.: J. Phys. Soc. Japan 31 (1971) 563). It is shown that the feature of this effect on the angular-correlation curve depends on the type of the apparatus. It is proved to be impossible to measure the momentum distribution of electron pairs along the direction strictly (within the error of less than a few milliradians) perpendicular to the lattice plane. The integrated intensity of the diffraction in copper has been remeasured with a better statistics. It agrees with the theoretical value within the experimental error of 5%.

View full abstract

Experimental study is described of the ionization of energetic neutral atoms on an untreated, metallic surface. The energy loss associated with the ionization is found to depend strongly on the incident angle of the particles onto the surface. The efficiency of ionization is of the order of 10⁻⁶. Angular distribution of the ions leaving the surface is also measured.

View full abstract

The complementary variational-bound theory for the dynamic polarizability, developed by Robinson, is extended to various mean excitation energies. The theory in the present paper provides both upper and lower bounds on the mean excitation energy for the grazing collision of a fast charged particle with an atom or a molecule (I₋₁), for its stopping power (I₀), for its mean fluctuation (I₁), and for the Lamb shift of an atomic energy level (I₂). Application to the 1s and 2s states of a hydrogen atom by use of 15-term trial functions yields bounds on lnI₋₁, lnI₀, lnI₁, and lnI₂ with a relative error of the order of 10⁻⁷, 10⁻⁵, 10⁻³, and 10⁻¹, respectively, if the I_k’s are given in Rydberg units. An extrapolation procedure increases accuracy by about two orders of magnitude.

View full abstract

Atomic resonance charge transfer Kr⁺+Kr→Kr+Kr⁺ (1) and molecular near-resonance processes Kr⁺+CO→Kr+CO⁺ (2) and CO⁺+Kr→CO+Kr⁺ (3) were studied by injected-ion drift tube technique in the energy range from 0.04 eV to 3 eV. The cross section of the process (1) was determined as 1.0±0.15×10⁻¹⁴cm² and found to be independent of the collision energy in the range studied. The cross sections of the processes (2) and (3) were found to be also independent of the collision energy and to have nearly equal magnitude of 3.7±0.7×10⁻¹⁵cm² and 3.3±0.7×10⁻¹⁵cm², respectively. The constant behaviour of the cross sections is explained by acceleration effect due to the attractive potential of each system. The large difference of cross section magnitude between the process (1) and the processes (2) or (3) is explained by polarized Franck-Condon factor. The threshold behaviour of the process (2) was seen below 0.1 eV.

View full abstract

The properties of plasmas produced by high power laser were investigated. The laser beam was irradiated on the various targets of beryllium Be, carbon C and polyethylene (CH₂)_n of high z number. The influence of the charge number z to the heating of the plasma was discussed.
The electron temperatures of plasmas from different targets were estimated by the measurement of soft X-ray.
The time variations of the reflected laser light from plasma were also investigated.
From the time of flight measurements with a Faraday cage and an electrostatic energy analyser, the different charged ions could be separated and ion temperatures and drift velocities were obtained from their velocity profiles.
No influence of laser pulse duration to the temperature was observed in the cases of durations 2 ns, 4 ns and 10 ns.
The experimental results for fast rising laser pulse were discussed and compared with those for slow rising one.

View full abstract

The theory of nonlinear wave modulation in a collisionless plasma is developed on the basis of the Vlasov description, with particular attention to the nonlinear Landau damping process associated with resonance effects at the group velocity of the wave. Contributions from resonance particles, moving at the group velocity, change the nonlinear Schrödinger equation, derived in previous investigations, into the following equation,
(Remark: Graphics omitted.),
where φ(ξ,τ) is the small but finite potential amplitude, and τ and ξ are stretched variables in the “reductive perturbation” theory. (Remark: Graphics omitted.) denotes a Cauchy principal part. Besides giving rise to the nonlocal-nonlinear term, the contribution of these resonant particles reverse the sign of the nonlinear coupling coefficient, q, relative to the value calculated neglecting their effect.

View full abstract

Obliquely propagating cyclotron harmonic waves are excited and their behavior is studied. The waves can be excited within the region that satisfies the condition, ω_p²>ω²−ω_c². Contrary to the commonly accepted notion, it is confirmed that the cyclotron harmonic wave is in a propagating mode rather than in a standing mode. The wavenumbers in the direction parallel to the magnetic field lines are determined by the geometrical length of the exciter.

View full abstract

Hybrid resonance due to two-ion species (H⁺ and H₂⁺) has been observed in a fully-ionized plasma for the density range (2×10¹¹ to 1.6×10¹² cm⁻³) using Stix type induction coil. The dependence of the resonance condition, which was found experimentally, on the concentration ratio of the two ion species, was in good agreement with theoretical predictions.

View full abstract

The study of two-dimensional, irrotational and steady state motion generated by an obstacle placed on the bottom of a running stream is made in terms of the linearised theory. It turns out that for a constant value of the parameter (gh⁄U²)>1, U and h being respectively the velocity and the depth of the stream, the wave length of the waves does not change with different obstacles whereas the amplitude does so. The dependence of amplitude on different shapes of obstacles leads to an interesting situation in the studies where it is observed that for certain shapes no wave-like motion occurs even in the case \sqrtgh>U, a condition where the wave motion is always expected to take place. The physical consequence is that such a body, when placed in a running stream, would not experience any resistance due to the wave formation.

View full abstract

The stability of a plane shear flow of an inviscid, incompressible and ideally conducting fluid of variable density in the presence of a non-uniform magnetic field inclined to the direction of the streaming is investigated, using the method which was taken by Howard in the non-magnetic case. Sufficient conditions for the stability, a semi-circle theorem and a bound on the growth rate for unstable modes are established.

View full abstract

The semicircle theorems obtained by Howard and Gupta are generalized. A sufficient condition for the stability and a semicircular region for the complex rate of growth of unstable modes are obtained with reference to the situation discussed by them in §6. The dual role of the uniform axial magnetic field is manifested.

View full abstract

In ion optical systems which contain ideal uniform sector-shaped magnetic fields, the expressions for positions X and widths W of image lines for arbitrary configurations of a magnet and an ion trajectory are derived using six (or eight) independent variables V_i. When an ion optical system is set in some configuration and if one regards the ideal configuration as the origin for measurements, the differential changes of positions or widths of images can be obtained by the linear combinations of the partial differential coefficients ∂X⁄∂V_i or ∂W⁄∂V_i. Comparing calculations of these changes with experiments following the methods described in this paper, one can estimate the order of magnitude of the effect due to, such as, higher order aberrations, fringing fields, or room temperature fluctuations, which influences to positions and widths of images.

View full abstract