Flow-injection technique for the determination of low levels of phosphorus in natural waters

An inexpensive flow-injection instrument for determining low concentrations of dissolved reactive phosphorus in natural waters is reported. The unique feature is the use of an inexpensive detector consisting of a flow cell and a simple photometer that incorporates a super-bright light-emitting diode as the source and a photodiode as the detector. The tin(II) chloride-molybdate method was optimized using a modified simplex optimization method. Silicate interference up to 5 mg Si l^?1 was removed by addition of 0.10% (w/v) tartaric acid. Using the tartaric acid-modified optimized reagents, a detection limit of 0.6 μg P l^?1 was achieved. The method was linear over the range 0–100 μg P l^?1 with an excellent precision (r.s.d. 2.9% at 2.0 and 0.5% at 50 μg P l^?1). An in-line pre-concentration anion-exchange column was used to obtain an even lower detection limit of 0.1 μg P l^?1 and applied to the analysis of real samples.     

Polymer characterization and gas permeability of poly(1-trimethylsilyl-1-propyne) [PTMSP], poly(1-phenyl-1-propyne) [PPP], and PTMSP/PPP blends

Pure gas and hydrocarbon vapor transport properties of blends of two glassy, polyacetylene-based polymers, poly(1-trimethylsilyl-1-propyne) [PTMSP] and poly(1-phenyl-1-propyne) [PPP], have been determined. Solid-state CP/MAS NMR proton rotating frame relaxation times were determined in the pure polymers and the blends. NMR studies show that PTMSP and PPP form strongly phase-separated blends. The permeabilities of the pure polymers and each blend were determined with hydrogen, nitrogen, oxygen, carbon dioxide, and n-butane. PTMSP exhibits unusual gas and vapor transport properties which result from its extremely high free volume. PTMSP is more permeable to large organic vapors, such as n-butane, than to small, permanent gases, such as hydrogen. PPP exhibits gas permeation characteristics of conventional low free volume glassy polymers; PPP is more permeable to hydrogen than to n-butane. In PTMSP/PPP blends, both n-butane permeability and n-butane/hydrogen selectivity increase as the PTMSP content of the blends increases. © 1996 John Wiley & Sons, Inc.     

Hydrocarbon/hydrogen mixed gas permeation in poly(1-trimethylsilyl-1-propyne) (PTMSP), poly(1-phenyl-1-propyne) (PPP), and PTMSP/PPP blends

The gas permeation properties of poly(1-trimethylsilyl-1-propyne) (PTMSP), poly(1-phenyl-1-propyne) (PPP), and blends of PTMSP and PPP have been determined with hydrocarbon/hydrogen mixtures. For a glassy polymer, PTMSP has unusual gas permeation properties which result from its very high free volume. Transport in PPP is similar to that observed in conventional, low-free-volume glassy polymers. In experiments with n-butane/hydrogen gas mixtures, PTMSP and PTMSP/PPP blend membranes were more permeable to n-butane than to hydrogen. PPP, on the other hand, was more permeable to hydrogen than to n-butane. As the PTMSP composition in the blend increased from 0 to 100%, n-butane permeability increased by a factor of 2600, and n-butane/hydrogen selectivity increased from 0.4 to 24. Thus, both hydrocarbon permeability and hydrocarbon/hydrogen selectivity increase with the PTMSP content in the blend. The selectivities measured with gas mixtures were markedly higher than selectivities calculated from the corresponding ratio of pure gas permeabilities. The difference between mixed gas and pure gas selectivity becomes more pronounced as the PTMSP content in the blend increases. The mixed gas selectivities are higher than pure gas selectivities because the hydrogen permeability in the mixture is much lower than the pure hydrogen permeability. For example, the hydrogen permeability in PTMSP decreased by a factor of 20 as the relative propane pressure (p/p_sat) in propane/hydrogen mixtures increased from 0 to 0.8. This marked reduction in permanent gas permeability in the presence of a more condensable hydrocarbon component is reminiscent of blocking of permanent gas transport in microporous materials by preferential sorption of the condensable component in the pores. The permeability of PTMSP to a five-component hydrocarbon/hydrogen mixture, similar to that found in refinery waste gas, was determined and compared with published permeation results for a 6-Å microporous carbon membrane. PTMSP exhibited lower selectivities than those of the carbon membrane, but permeability coefficients in PTMSP were nearly three orders of magnitude higher. © 1996 John Wiley & Sons, Inc.     

Chemical ionization and electron impact mass spectra of thiosulfinates,thiosulfonates and sulfinyl sulfones

The chemical ionization and electron impact mass spectra of some thiosulfinates, thiosulfonates and sulfinyl sulfones have been studied. The electron impact mass spectra of four of the six thiosulfonates show molecular ions of less than 1%. Inconclusive evidence was obtained for sulfenyl sulfinate type intermediates in the electron impact spectra of thiosulfonates. The electron impact spectra of thiosulfonates were similar to those of thiosulfonates. The chemical ionization (isobutane) mass spectra of thiosulfinates and thiosulfonates generally show protonated molecular ions [MH]⁺ as base peaks and [MH+1]⁺ and [MH+2]⁺ peaks.     

Correlations between eigenvalues of a random matrix

Exact analytical expressions are found for the joint probability distribution functions ofn eigenvalues belonging to a random Hermitian matrix of orderN, wheren is any integer andN. The distribution functions, like those obtained earlier forn=2, involve only trigonometrical functions of the eigenvalue differences.     

A Probabilistic Objective Function for R & D Portfolio Selection

This paper contributes to the literature on the problem of allocation of resources to a set of risky investments. Our objective is to develop the ideas in the context of a research and development laboratory. A mathematical programming approach to the resource allocation problem is taken, and various forms for an objective function under risk are discussed. A probabilistic objective function appropriate to R & D is isolated and tested on a small hypothetical example. Parametric linear programming is used to yield a near-optimum allocation.