An inequality for the trace of matrix products

In this paper we consider a product of n complex m×m matrices A_k (k=1,…,n) with singular values ∝^(k)_i ordered in decreasing magnitude. Using the spectral resolution for the operators $\text{A}{}^{\mathrm{dagger;}}{}_{\mathrm{k}}\text{A}{}_{\mathrm{k}}$, it is shown that $\text{|}\text{Tr}\text{A}{}_1\text{…A}{}_{\mathrm{n}}\text{|≤}\text{∑}\text{i=1}\text{m}\text{Φ}\text{i=1}\text{n}\text{α}{}^{\mathrm{(k)}}{}_{\mathrm{i}}\text{.}$This inequality is an extension of an inequality of von Neumann in the simple case that n=2. The necessary and sufficient condition for the equality sign to hold is established. Application of Hölder's inequality leads to further inequalities which can be useful in statistical mechanics.     

A study of the reactions π+p→ K+∑+ and K−p→π−∑+ at 10 GeV/c

For the first time, the reactions π⁺p→K⁺∑⁺ and K^?p→π^?∑⁺ have been studied in the same apparatus. This has been done at an adequately high momentum (10.1 GeV/c) to allow a check of the prediction of exchange degeneracy, that the differential cross sections should be converging at high energy. We have measured the cross section for momentum transfers t between t_min and t = ?0.3 (GeV/c)². We find that for both reactions the differential cross section shows an exponential fall, with no deviations right in to t =t_min (where some other experiments have shown a dip in the cross section). Furthermore, we find the magnitude of the differential cross sections to be closely similar at t = 0, with a ratio

$\text{R=}\text{(}\text{d}\text{σ}\text{d}\text{t)}{}_{\mathrm{t=0}}\text{(}\text{K}{}^{\mathrm{?}}\text{p}\text{→π}{}^{\mathrm{?}}\text{∑}{}^{\mathrm{+}}\text{)}\text{(}\text{d}\text{σ}\text{d}\text{t)}{}_{\mathrm{t=0}}\text{(π}{}^{\mathrm{+}}\text{p}\text{→}\text{K}{}^{\mathrm{+}}\text{∑}{}^{\mathrm{+}}$

However, the slope for the positive reaction is about 19% steeper than that for the negative reaction.     

Influence des marches atomiques sur le transport de matière à la surface du nickel

Surface atomic transport on Ni was measured by mass transport technique on a sinusoidal profile. One of the studied surfaces was within 15' of a singular (001) orientation. Others were vicinal surfaces. Kinetic damping coefficients $\text{1}\text{τ}$ are shown to be dependent on the profile parameters, i.e. of the density n_tot, of all the monoatomic steps and also of the density σ_c of kinks of the monoatomic steps. If the parameter describes the random motion of adatoms on isolated terraces and if α and k are two coefficients linked to adsorption/emission processes from steps and kinks, a sinusoidal profile with 5 μm periodicity obeys to the relation: $\text{1}\text{τ}\text{= (}\text{1}\text{τ}\text{)}{}_{\mathrm{T}}\text{[1+αn}{}_{\mathrm{tot}}{}^2\text{(1+ kσ}{}^2{}_{\mathrm{c}}\text{)]}$. This result demonstrates the importance of surface atomic structure which has been neglected in Mullins' theory, which may be non-negligible in all processes where surface mass transport is involved.     

On a least upper bound Tc of superconductors with paramagnetic impurities

We have obtained a least upper bound, $\text{k}{}_{\mathrm{B}}\text{T}{}_{\mathrm{c}}\text{? c(μ}{}^1\text{, t)A}$, on the critical temperature T_c of an isotropic superconductor with paramagnetic impurities described by the scattering matrix t for fixed values of μ¹. We have also obtained the corresponding optimal spectrum $\text{α}{}^2\text{F(m) = Aδ[ω?d(μ}{}^1\text{, A]}$. The numerical results for the functions $\text{c(μ}{}^1\text{, t)}\text{and}\text{d(μ}{}^1\text{, t)}$ are presented for $\text{α}{}^1\text{= 0.1}$ and 0.16 in the form of universal curves representing $\text{c(μ}{}^1\text{, t)}\text{and}\text{d(μ}{}^1\text{, t)}$ as functions of the reduced impurity concentration $\text{t}\text{= t/A}$. We have also established an upper limit to the reduced critical concentration $\text{t}{}_{\mathrm{crit}}$ for an arbitrary shape of α²F(ω)1.     

Transition magnetic moments of hadrons and sidewise dispersion relations

Systems of linear relations for the magnetic moments of the nucleon and transition moments μ_ωπγ, μ_?πγ and ${}^{\mathrm{\mu }}\text{N}{}^1{}_{1470}\text{N}\text{γ}$ are obtained on the basis of sidewise dispersion relations in the one-pion approximation. In such an approach the magnetic moments of the transition ω(?)→ πγ and $\text{N}{}^1{}_{1470}\text{→Nγ}$ are expressed in terms of the anomalous magnetic moments of the nucleon and the strong interaction coupling constants.     

Non-coherent scattering

The paper is devoted to the consideration of the problems of non-coherent scattering in a spectral line for an arbitrary redistribution law. The representation of the redistribution function r(x′,x) in the form

$\text{r(χ′,χ) =}\text{∑}\text{k = 1}\text{∞}\text{A}{}_{\mathrm{k}}\text{α}{}_{\mathrm{k}}\text{(χ′)α}{}_{\mathrm{k}}\text{(χ)}$

is used to study the various mechanisms of line broadening. The form of representation plays an important role in this investigation. New methods for solving radiative transfer integro—differential equations, as well as the corresponding integral equations are suggested. The solution of integral equations involves factorization of matrix integral operators. The generalized classical problems of absorption-line formation in stellar atmospheres are formulated and solved (Shuster's and Eddington's models). The results of numerical calculations, carried out to the fourth order, are given for Doppler broadening.     

The closure of cones in the algebra of test functions

We prove that the topological closure of the cone K of positive elements in the algebra of test functions is the same for several topologies (K?τo° = K?τ∞). Further we show that all elements of the closure of K are of the form

$\text{∑}\text{I=1}\text{∞}\text{f}{}^{\mathrm{(i)1}}\text{f}{}^{\mathrm{(i)}}\text{with f}{}^{\mathrm{<mtext>(</mtext><mtext>i</mtext><mtext>)</mtext>}}\text{?}\text{L}{}_{\mathrm{?}}$

and the sum converges. Applying these results, we give a characterization of the elements of K?^τ_o too.     

Hyperfine structure of CH3OH

Hyperfine structure of the (0, 0, 1) - (1, 0, 1) transition of methanol has been investigated by beam absorption and of the (J, 1, 3?) → (J, 1, 3+) transitions for J = 2, 3, and 6 by beam-maser spectroscopy. The best-fit results for the spin-rotation and spin-spin coupling constants $\text{C}{}_{\mathrm{JK\tau \pm }}{}^{\mathrm{(i)}}$ and $\text{D}{}_{\mathrm{JK\tau \pm }}{}^{\mathrm{(i)}}$, respectively, are in kHz¹: $\text{C}{}_{101}{}^{\mathrm{(1)}}\text{= 2.4(10)}$, $\text{C}{}_{101}{}^{\mathrm{(2)}}\text{= ?0.6(10)}$, $\text{D}{}_{101}{}^{\mathrm{(1)}}\text{= ?13.8(9)}$, $\text{D}{}_{101}{}^{\mathrm{(2)}}\text{= 7.0(9)}$, $\text{C}{}_{\mathrm{213?}}{}^{\mathrm{(1)}}\text{= ?5.0(10)}$, $\text{C}{}_{\mathrm{213?}}{}^{\mathrm{(2)}}\text{= ?5.5(10)}$ and ($\text{C}{}_{\mathrm{J13?}}{}^{\mathrm{(2)}}\text{- C}{}_{\mathrm{J13+}}{}^{\mathrm{(2)}}\text{) = 0.98(9)}$.     

Low momentum transfer theorem for two photon exchange in lepton hadron scattering

The two photon exchange contribution to lepton-hadron scattering is considered. Under the assumptions of Lorentz covariance, gauge invariance, unitarity and analyticity, we prove a low momentum transfer theorem for the relevant amplitudes. Fixed energy dispersion relations tell us that their nonanalytic part, in the neighbourhood of t = 0, is given by the contribution of the two photon cut in the t-channel. The t-channel absorptive parts are obtained from unitarity. Their calculation has as input the two amplitudes corresponding to Compton scattering on the hadron with a pole contribution, and the continuum controlled at low t by the electromagnetic polarizabilities. By means of the dispersion integral, one proves the expansion $\text{k}{}_1\text{(s)+k}{}_2\text{(s)}\text{?t}\text{+k}{}_3\text{(s)t}\text{log}\text{(?t)+O(t)}$ for the continuum contribution, where k₁(s) is the only unknown. Explicit expressions are obtained for the pole contribution as M → ∞, where M is the hadron mass, and for the continuum when (?t) <Λ and (?t) < 4m², where m is the muon mass and Λ is a characteristic parameter of the hadron structure.     

Excited vibrational state microwave spectra,structure, and force-field of trifluorosilane

The J = 2?1 microwave spectrum of six isotopic species of HSiF₃ has been observed and assigned in excited states of five of the six fundamental vibrations. The assignment is based on relative intensities, double resonance experiments, and trial anharmonic force constant calculations. Analysis of the spectra leads to experimental values for five of the $\text{α}{}_{\mathrm{r}}{}^{\mathrm{B}}$ constants, all three l-doubling constants q_t, one Fermi resonance constant φ₂₃₃, and one zeta constant $\text{ζ}{}_{\mathrm{6,\; 6}}{}^{\mathrm{(z)}}$.The harmonic force field has been refined to all the available data on vibration wavenumbers, centrifugal distortion constants, and zeta constants. The cubic anharmonic force field has been refined to the data on $\text{α}{}_{\mathrm{r}}{}^{\mathrm{B}}$ and q_t constants, using two models: a valence force model with two cubic force constants for SiH and SiF stretching, and a more sophisticated model. With the help of these calculations, the following equilibrium structure has been determined: $\text{r}{}_{\mathrm{e}}\text{(}\text{SiH}\text{) = 1.4468(±5)}\text{A}\text{?}$, $\text{r}{}_{\mathrm{e}}\text{(}\text{SiF}\text{) = 1.5624(±1)}\text{A}\text{?}$, ∠HSiF = 110.64(±3)°,     

Examinations on the radiative extinction functions in a cylindrical medium

Examining the derivation and physical meaning of the extinction function K(τ₀η, τ₀η₁) which prescribes the radiative contribution of fluid elements in the cylindrical homogeneous media, the authors introduce the alternative functions $\text{M(τ}{}_0\text{η,τ}{}_0\text{η}{}_1\text{) = ?}{}_1{}^{\mathrm{\infty }}\text{[K}{}_1\text{(τ}{}_0\text{η}{}_1\text{t) I}{}_0\text{(τ}{}_0\text{ηt)?t] dt (η ≦ η}{}_1\text{)}$ and $\text{N(τ}{}_0\text{η, τ}{}_0\text{η, τ}{}_0\text{η}{}_1\text{) = ?}{}^{\mathrm{\infty }}{}_1\text{[K}{}_0\text{(τ}{}_0\text{ηt)I}{}_1\text{(τ}{}_0\text{η}{}_1\text{t)?t] dt (η ≧ η}{}_1\text{)}$ which can eliminate the singularity of K(τ₀η, τ₀η₁) atη = η₁. The values of the functions M(τ₀η, τ₀η₁) and N(τ₀η, τ₀η₁) are numerically calculated and plotted for the parameter τ₀ and their characteristics are discussed in detail.Subsequently a numerical analysis is performed on the simultaneous radiative and convective heat transfer with uniform heat flux. As a result of this example, it is shown that the functions M(τ₀η, τ₀η₁) and N(τ₀η, τ₀η₁) are superior to the conventional extinction function K(τ₀η, τ₀η₁) for the numerical calculations.     

Energy levels of bound polarons

A new variational wave function to describe the ground state and the excited states of a bound polaron is proposed. It is of the form

$\text{|Ψ〉 = c|O〉|ø}{}_{\mathrm{n}}\text{〉 +}\text{∑}\text{g}{}_{\mathrm{k}}{}^1\text{V}{}_{\mathrm{k}}{}^1\text{(e}{}^{\mathrm{<mtext>i</mtext><mtext>k·r</mtext>}}\text{? ρ}{}_{\mathrm{k}}{}^1\text{)a}{}_{\mathrm{k}}{}^{\mathrm{+}}\text{|O〉|ø}{}_{\mathrm{n}}$

. It is argued that this form is reasonable for all electron—phonon coupling α and all strengths β of the Coulomb potential. Numerical and analytical results are derived for the energy of the ground state and compared to existing results. Results for the energy of the lowest p-type excited state of the bound polaron are obtained.     

Statistical flow-oscillator modeling of vortex-shedding

The very important engineering problem of modeling the fluid-structure interaction occurring during the shedding of vortices has defied, and will probably continue to defy, a closed form exact solution for the foreseeable future. Therefore, an attempt must be made to extract relevant information about the process in order to be able to have a basic understanding of it for the purpose of analysis. A useful method involves the flow-oscillator concepts of Hartlen and Currie [1] redefined here for stochastic processes. The fluid-structure system is assumed to be governed by the cross-coupled equations

$\text{x}\text{?}\text{(t)+2ξω}{}_{\mathrm{n}}\text{x}\text{?}\text{(t)+ω}{}^2{}_{\mathrm{n}}\text{=C}{}_{\mathrm{e}}\text{(t)pV}{}^2{}_0\text{(t)DL/2m (i)}$

$\text{C}\text{?}{}_{\mathrm{e}}\text{(t)+{α ? βC}{}^2{}_{\mathrm{e}}\text{(t)+γC}{}^4{}_{\mathrm{e}}\text{(t)}}\text{C}\text{?}{}_{\mathrm{e}}\text{(t)+ω}{}^2{}_0\text{C}{}_{\mathrm{e}}\text{(t)=b}\text{x}\text{?}\text{(t), (ii)}$

where these equations govern the structure and fluid oscillators, respectively. The fluid damping is non-linear. These equations are taken as stochastic differential equations because of the many unpredictable, random effects that determine the loading and response. The lift coefficient C_$\text{l}$(t) is assumed to be a zero mean, narrow band process and the velocity V₀, composed of a uniform, constant velocity current plus oscillating wave, a broad band process. The analysis is based on solving equation (i) for x(t) by using Duhamel's integral and substituting its derivative $\text{x}\text{?}\text{(t)}$ into equation (ii). This equation is then used to derive the Fokker-Planck equation for the process C_$\text{l}$(t). To obtain the Fokker-Planck equation, slowly varying variables are replaced by their long-time averages [2] and then the method of stochastic averaging is employed [3, 4]. The moment equation for the lift-oscillator process is derived from the Fokker-Planck equation and, as equation (ii) is non-linear, one finds the moment equation to be in terms of higher order moments. A truncation scheme [5] is used to derive the moment generating function. It is possible then to generate the first and second order statistics of the lift coefficient and the structure response in terms of the empirical parameters of fluid damping. This work was carried out in conjunction with an analysis of ocean wave-current forces with application to offshore fixed structures [6].     

The giant Gamow-Teller resonance states

The mean energy of the giant Gamow-Teller resonance state (GTS) is studied, which is defined by the non-energy-weighted and the linearly energy-weighted sum of the strengths for Σ^A_i = 1^τi?σⁱ? Using Bohr and Mottelson's hamiltonian with the ξl· σ force, the difference between the mean energies of GTS and the isobaric analog state (IAS) is expressed as$\text{E}{}_{\mathrm{GTS}}\text{?E}{}_{\mathrm{IAS}}\text{,≈ 2〈π¦Σ}{}^{\mathrm{A}}{}_{\mathrm{i}}\text{=1ξ}{}_{\mathrm{i}}\text{l}{}_{\mathrm{i}}\text{· σ}{}_{\mathrm{i}}\text{¦π〉/ (3T}{}_0\text{-4(k}{}_{\mathrm{\tau }}\text{?k}{}_{\mathrm{\sigma \tau }}\text{) T}{}_0$. The observed energy systematics is well explained by k_τ?k_{στ≈ 4/A MeV}. The relationship between the mean energies and the excitation energies of the collective states in the random phase approximation for charge-exchange excitations is discussed in a simple model. From the excitation energy systematics of GTS, the values of k_στ and the Migdal parameter g′ are estimated to be about $\text{k}{}_{\mathrm{\sigma \tau }}\text{=}\text{(16–24)}\text{A}\text{MeV and}\text{g′ = 0.49–0.72}$, respectively.     

Remarks on Mc Bryan's convergence proof for (α-4(cos(αϕ)−1+α2ϕ2/2))2-quantum fields

With the help of Mc Bryan's method we prove the existence of F(?)₂-fields, where $\text{F(x) ?}\text{∑}\text{n=0}\text{∞}\text{a}{}_{\mathrm{2n}}\text{x}{}^{\mathrm{2n}}$ is a power series with fast decreasing coefficients. Hypercontractive estimates show the existence of fields with a superposition of exponential or cosine functions as interaction, similar to Mc Bryan's interaction.     

Deviation from scaling in the triple-regge region of pp→ pX and fP universality

By generalizing $\text{f}\text{P}$ universality to Regge-particle “scattering” we obtain $\text{s}\text{d}\text{σ}\text{d}\text{t}\text{d}\text{M}{}^2\text{= F(}\text{s}\text{M}{}^2\text{,t)}$$\text{[1 +}\text{M}{}^{\mathrm{?1}}\text{b}{}_{\mathrm{<mtext>f</mtext>}}\text{(0)}\text{b}{}_{\mathrm{<mtext>P</mtext>}}\text{(0)}\text{]}$ for pp → pX, where b_f(t) and b_P(t) are the f and P Regge residues for, say, pp → pp. This agrees with the recent NAL data.     

A note on the momentum distribution function for an electron gas

The one-electron momentum distribution function $\text{〈a}{}_{\mathrm{k\sigma }}{}^2\text{a}{}_{\mathrm{k\sigma }}\text{〉}$ for an electron gas is investigated by a diagrammatic analysis of perturbation theory. It is shown that $\text{〈a}{}_{\mathrm{k\sigma }}{}^2\text{a}{}_{\mathrm{k\sigma }}\text{〉}$ has the following exact asymptotic form for large k (k ? p_F; p_F, the Fermi momentum): $\text{〈a}{}_{\mathrm{k\sigma }}{}^2\text{a}{}_{\mathrm{k\sigma }}\text{〉 =}\text{4}\text{9}\text{(}\text{αr}{}_{\mathrm{s}}\text{π}\text{)}{}^2\text{×(}\text{p}{}_{\mathrm{<mtext>F</mtext>}}{}^8\text{k}{}^8\text{) g?(0) + ?}$, where g?(0) is the zero-distance value of the spin-up-spin-down pair correlation function. The physical implications of the above asymptotic form are discussed.     

Long time expansion of two-dimensional correlation functions

We show that the long time behaviour of the velocity correlation function in a two-dimensional classical system with pairwise repulsive potentials can be represented by a series expansion of the form

$\text{〈υ}{}_{\mathrm{1x}}\text{υ}{}_{\mathrm{1x}}\text{(t)〉 = d}{}_0\text{t}{}^{\mathrm{?1}}\text{+ d}{}_1\text{t}{}^{\mathrm{?1}}\text{log}\text{t/t}{}_0\text{+ d}{}_2\text{t}{}^{\mathrm{?1}}\text{(}\text{log}\text{t/t}{}_0\text{)}{}^2\text{+ …}$

, where t₀ is mean free time between collisions. To lowest order in the density an exact expression has been obtained for d₁ employing the kinetic theory ofsystems with hard-core interactions. The significance of the series is discussed at low and intermediate densities.     

Theory of the decay process of metastable state

The short-time and long-time behavior of the distribution function P(x, t) are investigated in the laser model by using the generating function $\text{G(α,}\text{β}\text{;t) = σ(α-y(t)) Π}{}^{\mathrm{\infty }}{}_{\mathrm{n=2}}\text{σ(β}{}_{\mathrm{n}}\text{- M}{}_{\mathrm{n}}\text{(t)),}\text{where}\text{y(t)  ?xP(x, t)}\text{d}\text{x}\text{and}\text{M}{}_{\mathrm{n}}\text{(t)  ?(x - y(t))}{}^{\mathrm{n}}\text{P(x,t)}\text{d}\text{x}$.