BUNSEKI KAGAKU Abstracts

In this study, we focused our attention on breast fluid as specific perfluorinated compounds (PFCs) as an exposure source to infants, and developed a highly sensitive and accurate analytical method using high-performance liquid chromatography/tandem mass spectrometry (LC/MS/MS). For sample pretreatment, samples were determined using weak anion exchange (WAX) and solid phase extraction (SPE). The average recoveries of the analytes were in the range 94.3〜109.0％ (RSD ＜10.3％), and the limits of quantitation (LOQ) were determined to be 0.012 ng/mL for PFOA and 0.004 ng/mL for PFOS, PFHxS and PFNA, respectively. Relatively high levels of PFCs were detected from human milk, which ranged over 0.046〜0.098 ng/mL for PFOS and over 0.016〜0.270 ng/mL for PFOA. In addition, human milk and maternal plasma collected from the same donor were analyzed. As a result, the concentration of PFCs in human milk and maternal plasma showed significant correlations for PFOS (r＝0.78) and PFOA (r＝0.82).

Dispersive liquid-liquid microextraction (DLLME) coupled gas chromatography mass spectrometry was applied to the determination of benzodiazepines (flunitrazepam and nimetazepam). Some important parameters, such as the kind of extraction and the disperser solvent, the volume of the disperser solvent and the extraction time were investigated. Using the optimum conditions of DLLME for benzodiazepines, tricyclic antidepressant drugs (amitriptyline and nortriptyline) and phenothiazines (chlorpromazine and promethazine) were extracted from aqueous samples. In addition, the DLLME technique was applied to the extraction of flunitrazepam from alcohol drink ; simple and rapid extraction was possible. Thus, the DLLME technique is useful for the extraction of drugs in the forensic science field.

The water content of hair was measured by a dry-weight method, and was predicted by multiple-regression analysis using near-infrared (NIR) spectrum data after a spectra-pre-treatment. The correlation between the value by the dry-weight method and the NIR predicted value was r²＝0.957, and indicated that predicting the water content of hair from the NIR spectrum data is possible. In addition, we obtained the formula ∂(lnw)/∂(1/T)＝ΔH/R (w, water content ; ΔH, vaporization enthalpy change) by an analogy of the Clausius-Clapeyron equation, and we proposed that the individual variability of the moisture in hair (difficulty in evaporation of the water in the hair) is the difference in the vaporization enthalpy. The water content of the hair at 20〜100°C was measured for 10 people (21〜23 years old), and ΔH was obtained from the slope of plots of (ln [w/mol/kg]) vs. (1/T). As a result, the capacity of moisture in hair could be suggested by comparing it with ΔH (43.991 kJ/mol) of pure water.

A spectrophotometric method for the determination of nitrite ion was established based on a fading of the pyrogallol red-molybdenum(VI) complex in the presence of sodium dodecylsulfate, an anionic surfactant. This method can be used to determine 40〜460 ng mL⁻¹ nitrite ion ; the effective molar absorptivity at 535 nm and the relative standard deviation were 1.0×10⁵ L mol⁻¹ cm⁻¹ and 1.92％ (n＝6), respectively. This procedure is much more sensitive than that using a diazotization reaction. Recoveries of nitrite ion from real samples (rain water and commercial drinking water) were within 96.9〜100％.

A new, simple and sensitive spectrophotometric method for the determination of Ti(IV) was established. The method was based on complex formation among 4-Hydroxyderricin (Hy) or Xanthoangelol (Xa) and Ti(IV) in the presence of anionic surfactant. {Hy} : Beer’s law was obeyed in the range of 0.01〜0.96 μg/mL for Ti(IV) in a strong acidic medium. The effective molar absorptivity at 433 nm and 0.65％ (n＝6). {Xa} : Beer’s law was obeyed in the range of 0.02〜0.74 μg/mL for Ti(IV) in a strong acidic medium. The effective molar absorptivity at 429 nm and 0.53％ (n＝6), respectively.

We found a selective fluorescence quenching of perylene by gold(III) in a micellar system of polyoxyethylene-type nonionic surfactants. Nitric acid was essential for incorporating gold(III) into the micelles, but deactivated the excitation of perylene. The deactivation was minimized by solubilizing perylene in micelles of nonionic surfactants having no aromatic moiety. Among the nonionic surfactants tested, Brij-35, having the shortest hydrocarbon chain and the highest polymerization degree of oxyethylene, was the best choice for obtaining the greatest quenching effect of gold(III). A linear relationship was obtained in the range of 5×10⁻⁸〜5×10⁻⁷ M of gold(III) in the presence of 1 M nitric acid and 0.05％ (w/v) Brij-35. The Stern-Volmer quenching constant, K_sv, for gold(III) was 1.28×10⁷ M⁻¹. On the other hand, the K_sv (M⁻¹) values were 4.3×10⁴ for platinum(II), 8×10³ for palladium(II), and 5×10² for iron(III), being far smaller than the value for gold(III). Other metal ions including aluminum(III), manganese(II), cobalt(II), nickel(II), copper(II), zinc(II), silver(I), tin(IV), and lead(II) at 10⁻⁴ M had negligible quenching effect. The present method can be applicable to the selective determination of traces of gold(III).

A simple and selective method for the speciation of sub μg/L levels of inorganic chromium species in water sample is described. The proposed method is based on micro solvent extraction with an in situ extractant formation method and graphite furnace atomic absorption spectrometry. By mixing carbon disulfide and pyrrolidine in xylene prior to use, pyrrolidinedithiocarbamic (PDC) acid forms. The extraction behavior of Cr(III) and Cr(VI) caused by a variation of the pH and composition of a mixed solvent was investigated. At pH 3.0, Cr(VI) was selectively extracted from co-existing Cr(III) as a PDC complex. At pH 6.0, both Cr(III) and Cr(VI) were simultaneously extracted into the organic phase after heating at 35°C for 30 min. A 20 μL aliquot of the organic phase was injected into a graphite furnace. Under a 50-fold preconcentration factor by using 25 mL of water samples, linear calibration curves of Cr(VI) and total Cr [Cr(III)] were obtained over the range of 0 to 0.4 μg/L and 0 to 0.5 μg/L with detection limits (3σ) of 0.75 ng/L and 0.64 ng/L. The relative standard deviations (n＝10) were 1.8％ and 3.4％ for 0.16 μg/L of Cr(VI) and Cr(III), respectively. The proposed method was successfully applied to mineral water, river water and spiked samples.

A collection/concentration system controlled automatically by a lab-made program incorporated in ICP-AES was used to determine multi-elements in green tea leaves and their extracts into hot water. Twenty four elements (Cd, Ce, Co, Cu, Dy, Er, Eu, Gd, Ho, La, Lu, Nd, Ni, Pb, Pr, Sc, Sm, Tb, Th, Tm, V, Y, Tb and Zn) could be collected on a ME-3 chelating resin packed in a mini-column. This mini-column was installed in the selection valve on the collection/concentration system, which was connected on line to ICP-AES. Conditioning of the resin was carried out with a 1 mL portion of a 0.5 M ammonium acetate buffer solution (pH 5.5) at a flowing rate of 50 μL min⁻¹, collecting elements with 10 mL of the sample solution (pH 5.5) at a flowing rate of 70 μL min⁻¹, and washing with 0.5 mL of water. Elution was done by passing through 2.5 mL of 2 M HNO₃ at 35 μL min⁻¹. The enrichment factors of elements were from 40 to 74, and the detection limits (LOD) were from 0.01 ppb to 0.20 ppb. The RSDs of 7 measurements with 0.5 ng mL⁻¹ standard solution were between 2.9 and 8.9％. Tea leaves were digested by a microwave without hydrofluoric acid after predigesting and the digested solutions were measured directly by ICP-AES. The extracts from tea leaves in hot water could be measured with the proposed collection/concentration system.

The dissolved and acid-soluble concentrations of major and trace elements in deep and surface seawater samples near Kume Island in Okinawa Prefecture were investigated by ICP-MS and ICP-AES with two types of preconcentration methods using chelating disk preconcentration and lanthanum hydroxide coprecipitation. As a result, 16 elements in each sample were determined over the concentration range from％ to ng L⁻¹ (ppt). In addition, it was elucidated from multielement profiling analyses by using the obtained analytical results that the concentrations of Cd, Zn, Ni, Cu, and Co in deep seawater near Kume Island were higher than those at the surface. The acid-soluble particle fraction to the total concentration (P_acid) was calculated from dissolved and acid-soluble concentrations. The P_acid values for Cu, Co, Zn, and Ni in deep seawater were 10〜30％, and higher than those at the surface, while that for Cd was below 10％, and almost the same as those for the surface. The results suggested that portions of Cu, Co, Zn, and Ni exist as acid-soluble biogenic particles in deep seawater (water depth : 612 m), which were detritus and colloidal particles bound with organic matter, while most Cd exists as a dissolved form.

A novel spectrophotometric method was established for the determination of famotidine (FAM) and its related drugs. The method is based on fading reactions of ternary complex formation among palladium(II), o-carboxyphenylfluorone and hexadecyltrimethylammonium bromide (HTAB). In the determination of FAM, Beer’s law is obeyed in the range of 0.01〜0.7 μg cm⁻³. The apparent molar absorptivity and the relative standard deviation were 6.36×10⁵ dm³ mol⁻¹ cm⁻¹ and 1.89％ (n＝6), respectively. The method was successfully applied to assays of FAM and other H₂ receptor antagonists in pharmaceutical preparations. The proposed method should be useful for a simple and sensitive determination FAM and its related drugs.

The reduction of nitrate to nitrite is needed for a colorimetric determination of nitrate with the Griess reaction. The reduction rate of nitrate to nitrite by connecting some copper-cadmium reduction columns in a continuous-flow analysis was investigated. The reduction rate of nitrate did not depend on the concentrations of nitrate and nitrite. It was confirmed that a reduction rate of 95％ or more was usually realized using a single column. Plural reduction columns were connected and the obtained reduction rate was evaluated. Consequently, the connection of the columns increased the reduction rate and a steady value of the reduction rate (99.9％ or more) was achieved by three connected columns.

Matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) was applied to a structural analysis of condensed tannins extracted from wasted bark through a hydrothermal process under different temperatures. The mass spectra of hydrothermal extracts obtained at 100°C and 150°C showed a series of peaks for the tannin oligomers up to 8-mers, consisting mainly of the combination of three types of flavan-3-ol units. This result suggests that hydrothermal extraction under these temperatures leads to a rapid recovery of the tannin components while keeping their oligomer structure. On the other hand, the ion peak of the tannin oligomer was almost missing from the mass spectrum of extracts obtained at 200°C, suggesting that degradation and/or polymerization of the tannins occurred significantly through the extraction procedure under this temperature. Furthermore, the number-averaged molar masses of the tannin components in each extract sample were calculated on the basis of their corresponding mass spectra, and interpreted in terms of the antioxidant activity. This interpretation revealed that 1) the antioxidant activities for the tannin components increased with an increase in their number-averaged molar masses, and 2) the tannin components in hydrothermal extracts obtained under lower temperatures showed relatively higher antioxidant activities than α-tocopherol.

Polycyclic aromatic hydrocarbons (PAHs) including phenathrene, fluoranthene, and pyrene in sediment were extracted with aqueous micellar solutions of anionic surfactants. The surfactants examined were sodium dodecyl sulfate, sodium oleate, and sodium di(2-ethylhexyl) sulfosuccinate. The extraction yields of PAHs increased with increasing the surfactant concentration and the extraction time. They reached constant values (ca. 30％), which agreed with results obtained by the 1-butanol extraction method. The present results suggest the usefulness of surfactant-based extraction for evaluating the bioavailability of organic pollutants in sediments.

A simple and sensitive fluorophotometric method for the determination of resorcinols was examined utilizing a condencation reaction with 2,4-dihydroxybenzo-ylbenzoic acid in the precence of H₄P₂O₇. In the determination of resorcinol, the calibration curve exhibited linearity over a resorcinol concentration range of 1.0〜1100 ng mL⁻¹ at an emission wavelength of 515 nm with an excitation of 445 nm and with the relative standard deviations (n＝5) of 4.8％ for 110 ng mL⁻¹ of resorcinol.

Three solid-state ISEs (Cd, Pb, Cu) were evaluated for direct potentiometry of polyacrylate. Cd-ISE had a lower pH response but an appreciably high response to PA, while Cu-ISE had a remarkably high pH response but a rather low response to PA. Pb-ISE showed intermediate behaviors. These results suggest direct potentiometry of PA by simultaneous measurements of pH and the Cd-ISE response. Anions of foreign salts may cause positive errors by the interactions with Cd^2＋, whereas cations of salts may cause negative errors by the interactions with PA. The extent of such errors is, however, negligible in the detection of PA in the circulating water of cooling systems.

A cup-stacked carbon nanotube (CSNT) was formed by stacking cup-shaped carbon units with a hollow open channel of 80 to 100 nm in diameter, in contrast to the structure of a conventional multi-walled carbon nanotube (MWNT) made up of some seamless cylinders. The side of the cup is hydrophobic and the edge is hydrophilic due to the presence of carboxyl groups. Thus, a CSNT has a large hollow space in which one can incorporate enzyme molecules based on a physical adsorption. In this work, CSNT and other nanosize supports [MWNT, graphite, activated carbon and aminopropyl CPG (APCPG)] were packed into similar stainless-steel columns (4×3 mm), and glucose oxidase (GOD) was adsorbed by circulating an enzyme solution into each column. The CSNT adsorbed GOD both strongly and rapidly. The adsorbed amount was significantly larger than those onto MWNT, graphite, and activated carbon, and equivalent to that immobilized onto APCPG by a glutaraldehyde crosslinking reaction. As a result, it was found that CSNT has a high adsorptive activity for GOD. The flow-injection system with a GOD adsorbed CSNT packed column responded linearly to the concentration of glucose over a dynamic range of 5×10⁻⁷〜1×10⁻³ M.

A novel spectrophotometric method was established for the determination of iodide ion and organic iodine compounds. The method is based on a color-fading reaction of the ion-association complex of adenin silver(I) ion and eosin. In the determination of iodide ion, Beer’s law was obeyed in the range of 0.04〜1.3 μg mL⁻¹. The apparent molar absorptivity at 560 nm and the relative standard deviation were 1.38×10⁵ dm³ mol⁻¹ cm⁻¹ and 2.34％ (n＝5), respectively. The proposed method, which requires no pretreatments, should be useful for a simple and sensitive determination of iodide ion. In addition, this method could be applied to organic iodine compounds.

A new visual colorimetric and fluorimetric determination of trace zinc(II) has been developed by thin-layer chromatography (TLC) with tetrakis(4-carboxyphenyl)porphine (TCPP). In this method, the spots that are dependent on each metal ion concentration of zinc(II), copper(II), cadmium(II), and mercury(II) appear selectively on an octadecylsilanized (ODS) silica thin layer by chromatographic separation with 90％ (w/w) methanol. The yellow-colored zinc(II)-TCPP complex is distinguished visually from the other species with different colors. The zinc(II) concentration is determined by a visual comparison of the yellow-colored spot. The visual detection limit is 3×10⁻⁶ mol dm⁻³, defined as the minimum concentration that is visually distinguishable against a reagent blank. The calibration curve assessed with the densitometric responses is linear in the concentration range 0〜5×10⁻⁶ mol dm⁻³. The detection limit, which is calculated with 3 σ at the reagent blank, is 3×10⁻⁷ mol dm⁻³. On the other hand, the fluorimetric detection of the zinc(II)-TCPP complex is also available under the irradiation of UV light in the dark, and the visual detection limits is 8×10⁻⁷ mol dm⁻³. The calibration curve in the fluorimetric mode is linear in the concentration range 0〜3×10⁻⁶ mol dm⁻³, and the limit of detection is 3×10⁻⁷ mol dm⁻³. The effect for the determination of zinc(II) is not found in the presence of magnesium(II), aluminum(III), calcium(II), chromium(III), iron(III), nickel(II), and cadmium(II). The proposed method has been successfully applied to the determination of zinc(II) in river-water samples around an old mine.

The adsorptive removal of borate-boron from an aqueous solution using silica gel was studied using a batch equilibration technique. The effects of the pH and metal ions, such as alkali metal and alkaline earth metal, were clarified. The boron adsorption efficiencies increased with increasing the pH and the concentration of alkaline earth metal ions, and their maximum adsorptions were obtained at pH 10-12 in solutions containing calcium ion. The adsorption data obtained by changing the boron concentration and the adsorbent dosage were well represented by the Freundlich isotherm model.

The preparation of calibration curves using an aqueous standard solution was studied for the direct determination of trace levels of zinc in high-purity iron and steels by graphite furnace atomic absorption spectrometry (GF-AAS) using solid sampling technique. Certified reference materials of high-purity iron and steel samples were used to confirm the trueness of the proposed methods. The standard addition method for the solid sampling technique that we have proposed gave accurate analytical results. The calibration curve prepared by the standard addition method can be used as a kind of iron-matrix matching calibration curve, and it provides rapid and accurate determinations of zinc. The detection limit of zinc in iron and steel samples was 0.86 mg kg⁻¹ when 15 mg of one sample was measured.