The major method of conversion of methanol to a colorimetrically determinable derivative is its oxidation to formaldehyde with potassium permanganate and phosphoric acid. However, since the reaction with these methanol oxidizing agents to form formaldehyde is incomplete, the calibration curve of methanol by the conventional chromotropic acid method is not linear for 5100μg/ml. It is found out that an addition of potassuim permanganate, sulfuric acid and propionaldehyde promotes more oxidation of methanol to formaldehyde. The addition of propionaldehyde improves the determination of methanol by the conventional chromotropic acid method.
The recommended analytical procedure for methanol by the p-rosaniline method is as follows.
Oxidize 10ml of the test solution containing 360μg of methanol by adding 1ml of 0.2v/v% propionaldehyde solution, 1ml of 0.5w/v% potassium permanganate solution, and 1ml of 0.5N sulfuric acid. Swirl, keep at room temperature for 10minutes. Reduce excess methanol oxidizing agents by adding 1ml of 1w/v% sodium sulfite solution. Swirl, add 1ml of p-rosaniline hydrochloride solution (dissolved 0.16 grams completely in 24 ml of conc. hydrochloric acid, and diluted to 100ml). Swirl, stand for 60minutes at room temperature. Measure the absorbance at 580nm against the reagent blank obtained by the same procedure.
The calibration curve is linear for 360μg/10ml of methanol. The precision of this method is 0.025, the molar extinction coefficient is 5.4×10³.
The sensitivity of this method is the highest of those ever reported colorimetrically. Formaldehyde and ethylenglycol interfere seriously in this method.

View full abstract

An analytical method to determine the carbon black content in vulcanized rubber was investigated by means of pyrolysis technique. This method is based on the evolution of organic compound in a vulcanized rubber on the pyrolysis in a nitrogen (N₂) gas and on the combustion of the carbon black to carbon dioxide in an air. Pyrolysis of the sample in N₂ gas was carried out by using a commercial differential thermal analyser (Rigaku Denki Co., Ltd.)and them the combustion of the residue was performed by an electric furnace in an air. The pyrolysis temperature in N₂ gas was first programmed at maximum heating rate of 20°C/min. from room temperature to the fixed pyrolysis temperature (between 500 and 750°C), and then kept for 10 minutes isothermally at the desired pyrolysis temperature. After reweighing, subsequent combustion of sample in an air was performed in an electric furnace at 700 °C. The percentage of carbon black in vulcanized rubber was calculated as follows:
Carbon black content (wt%) =B-C/A×100
Where A=grams of specimen used,
B=grams of pyrolysis residue in N₂ gas,
C=grams of combustion residue in air.
Carbon black content determined between 500 and 750°C was independent of the pyrolysis temperature. For this reason, the optimum pyrolysis temperature of 600°C was experimentally decided.Under these experimental condition, this method is satisfactory for the determination of carbon black in a vulcanized rubber. The observed values of such hydrocarbonic vulcanized rubber as styrene-butadie necopolymer (SBR), natural rubber (NR), isobutylene isoprene copolymer (IIR) and ethylene-propylene terpolymer (EPDM) were in fairly good agreement withthe theoretical values calculated from the vulcanization recipe. While the observed values of polychloroprene rubber (CR) and acrylonitrile-butadiene copolymer vulcanized rubber (NBR) were in poor agreement with the theoretical values because of carbonaceous residue on pyrolysis in N₂ gas in addition to raw carbon black.Therefore, carbon black content of the former vulcanized rubber was determined directly from the experimental results, while the case of latter were determined successfully from the calibration curves by experimental correction. By using of calibration curves, EPDM which is representitative hydrocarbonic polymer, NBR and CR would be determined with 0.4, 1.0 and 0.5 wt% of mean absolute deviation, respectively, at the pyrolysis temperature of 600°C. This technique is very simple and rapid, and is applicable to the routine work.

View full abstract

The reaction gas chromatographic technique was applied for the determination of the non-volatile aromatic amines employing the Schiff's base formation reaction. The sample amine was packed in the reaction column by the following procedure; the sample amine dissolved in the adequate solvent (ethyl alcohol or DMSO) was injected into the reaction column in which the solid support (acid washed and DMCS treated Chromosorb W, 60/80) was packed, then the reaction column was inserted between the injection port and analytical column in a chromatograph, and the solvent was swept off from the reactoin column by the carrier gas flow at 100°C.
The constant amount of benzaldehyde was repeatedly injected into the injection port and carried into the reaction column. Consequent chromatogram showed only the water peak, by-product of the reaction, and the peak of excess benzaldehyde before and after the completion of the reaction, respectively. However, the both peaks emerged when one of the repeated injection of benzaldehyde stepped over the end point of the reaction
During the course of whole GC procedure, the sample amine and Schiff's base formed by the reaction remained in the reaction column. The amount of sample amine in the reaction column was calculated from the amount of excess benzaldehyde which was determined from its peak area in the reaction chromatogram. The analytical conditions such as carrier gas flow rate, the amount and the concentration of sample amine, and the amount of benzaldehyde injected at one time were investigated. The proposed conditions were: carrier gas flow rate; 30ml/min, reaction temperature; 120°C, other conditions such as the amount of benzaldehyde of each injection, the amount and the concentration of the sample amine did not affect the recovery, which was about 100%, and the standard deviation was 2.0 to 5.0%.

View full abstract

The complex formation reactions of Chromazurol S (CAS, H₄L) with aluminum or copper in the presence of zephiramine (ZCl) were studied by spectrophotometry. As the acid dissociations of CAS were effected by the concentration of zephiramine and the reactions between CAS and zephiramine were not stoichiometric, zephiramine was omitted from the chemical equations.
An excessive CAS and aluminum formed a 1:2 acidic complex, Al(HLZ)^-₂, (the absorption maximum at 620nm) below pH 5.1 and a 1:3 complex, Al(LZ²)₃^3- (the absorption maximum at 610nm) above pH 5.2 in the presence of zephiramine, whereas they formed AlL₂^5- and AlL₃^9- in the absence of zephiramine. Consequently, the 1:2 acidic complex of CAS with aluminum was possibly formed by an action of zephiramine. However, an excessive CAS with copper formed a 1:2 acidic complex, Cu (HLZ₂)₂, (the absorption maximum at 620nm), which has different structure from that of aluminum both in the presence and absence of zephiramine.
The absorbance of the 1:2 acidic complexes of CAS such as Al (HLZ)^-₂ and Cu(HLZ₂)₂, was very much enhanced by an action of micelle of zephiramine having a positive charge. Thus their molar absorptivity was 1.08×10⁵ and 1.40×10⁵, respectively.
Only the absorption spectrum of H₂LZ₂ which has unsymmetrical distribution of charge in CAS was shifted in the presence of zephiramine, hence we considered that the absorption of the 1:2 acidic complexes of CAS which have the same structure as H₂LZ₂ was also shifted largely by the micelle having a positive charge.
Since the 1:2 acidic complexes showed the highest absorbances in the presence of zephiramine among that of CAS, they were applied to the determination of aluminum and copper.
The calibration cruves obey Beer's law for 0.010.3ppm of aluminum in solutions containing 3.4×10^-4M CAS and 2.72×10^-3M zephiramine and at pH 4.55.1 and for 0.0250.6ppm of copper solutions containing 1.7×10^-3M CAS and 4.35×10^-3M zephiramine and at pH7 where the formation of the 1:2 acidic complex of the metal ion with CAS was concluded, respectively.

View full abstract

The authors had been reporting the tables and figures for the elucidation of C-13 nuclear magnetic resonance spectra, and the sub-charts were presented here, as a link of the series, for aromatic and heterocyclic carbons of mono and polynuclear molecules, with restrictions that the substitution of benzene was limited up to two, and in the case of complicated skeleton structure only the chemical shifts of basic molecule were listed.
The figure was produced in case of monosubstituted benzene so as to cover the chemical shifts of all aromatic constituent carbons, and about disubstitution system, the chemical shifts of ring carbons (s-carbon) hydrogen of which was replaced by a substituent were arranged in conjunction with the other substituent.
Fig. 1 was the sub-chart for monosubstituted ben zenes. It was observed that the chemical shifts of s-carbons were lowered in comparison with the unsubstituted except iodine compounds, and it caused higher shifts of adjacent ortho carbons. The degree of shifts of meta and para carbons were found less, and especially in the former case it had nearly the same transition level regardless of the species of the substituents.
Fig.2 was for disubstituted benzenes. There was a common feature between mono and disubstituted benzene derivatives that the chemical shifts of s-carbons were lowered according to the orders of -CO, -CH₃, -N and -O. The ranges of chemical shifts in the former case were 5563; 4356; 3854; 3443ppm respectively, and in the latter series the chemical shifts were distributed rather widely presumably influenced by the other substituents.
Although it was found very difficult to assign exactly all the lines of aromatic carbons from the figure directly, Savitsky's “Additivity relations” was useful for the assignment of spectra of disubstituted benzene derivatives and it was found that it might be applied to tri and polysubstituted systems if more deviations were allowed. Chemical shifts of 33 kinds of monosubstituted benzenes were listed in Table I for the aid of characterization of polysubstituted benzene derivatives, and an example of its application was also given.
Fig. 3 was for unsubstituted heterocyclic and polynuclear aromatic molecules. Every skeleton structure was found its own characteristic spectra pattern, and the assignment of the derivatives might be accomplished with the use of this figure and the information about the lines of substituent carbons and their effects on chemical shifts of the skeleton carbons.

View full abstract

It was observed in many organic compounds that carbon nucleus adjacent to halogen atoms shows an extraordinary C-13 NMR chemical shift, and the authors presented here, the figures for halogen containing compounds, as a link of the researches on the C-13 NMR sub-charts. The independent description of halogenated carbons was found to be convenient because by that, it is possible not only to refrain from complication of all the sub-charts, but also to rule the characteristic of chemical shifts of any kind of halogenated compounds easily.
The references cited here are the same as in the previous papers and the number of data is given in parentheses in the figure. Fig. 1 shows the general tendency of C-13 NMR chemical shift of halogen containing compounds. It was found to be clear in whole sorts of compounds that the relation between the chemical shifts and the element contained is, without any exception, in the order of fluorine, chlorine, bromine and iodine from the lower magnetic field.
Although the relationship between the functions and their chemical shifts is similar to the corresponding functions containing no halogen atoms, each element is found to show its own characteristic. Carbon adjacent to a fluorine shifts to the lower side than its unsubstituted one, but carbon neighbouring iodine atom shifted to the higher side. In case of carbons attaching chlorine or bromine, the degree of shift is between that of fluorine and iodine as shown in the figure. As an extreme example, Fig. 2 shows the chemical shift of halomethanes. Tetraiodomethane resonates at about 486 ppm from carbon disulfide, and this is the highest chemical shift ever reported.
Due to the lack of data and the difficulties of experimental procedures on fluoromethanes, the authors were obliged to presume the chemical shift of polyfluorinated methane derivatives using the data of chloro and fluoroacetic acid and monofluoromethane. The chemical shift of tetrafluoromethane is expected to reach 35ppm, and this is also the lowest value of sp³ carbons.
Fig. 3 shows the relation between the number of carbon and the chemical shift of α, β, and γ carbons of 1-halo-n-alkanes. Regardless of chlorine, bromine or iodine derivatives, the α carbons were observed to resonate at the lowest field when the number of carbon is three and this phenomenon is commonly observed with α carbon adjacent to the carbonyl in fatty acids. The chemical shift of α carbon adjacent to chlorine, bromine and iodine is observed to converge nearly 149, 160 and 187ppm respectively when the number of carbon of the alkane is beyond 6.

View full abstract

The surface of steel plate is not necessarily clean, as it is usually spoiled by surface products and adhesives on it. Therefore, it is important to identify these substances in the analysis of iron and steel. The substances are mainly classified as the following four groups; a) graphite, b) amorphous substance, c) crystal as estimated to result from dust and d) oxyhydroxides and oxides of iron.
Such surface substances can be usually identified by the combined use of optical and electron microscopic observation, electron and X-ray diffraction, infrared and emission spectroscopy, electron probe microanalysis and so on. Though the amorphous one is often estimated to be a free carbon in the substances of the above four groups, it was difficult to identify such an amorphous substance, because the substance is a quite little in amount and localizes on the surface of steel plate.
In this paper, an analytical method was investigated for the trace amounts of amorphous substance formed or adhered on the surface of steel plate (see Photo.1), which was difficult to identify directly with an X-ray diffraction (cf. Fig.1) or an electron probe microanalysis (see Photos. 24) until now. As the results, the effective method using an electron probe microanalyzer through a new sampling procedure (see Fig.2) was proposed and the presence of trace amounts of a free amorphous carbon on the surface of steel plate was proved by the results of electron probe microanalysis and electron microscopic observation.
The carbon on steel plate is extracted as follows; The objective carbon is extracted into an acetylcellulose film formed on the surface of a specimen. Beryllium is evaporated on the acetylcellulose film, which is dissolved in methylacetate. As the result, the evaporated beryllium films float in methylacetate.
The prepared beryllium film extracting the objective carbon is put on a beryllium bed and is analyzed with the X-ray scanning images of carbon in an electron probe microanalyzer (see Photos. 5, 7 and 9). Iron powder (see Photo.6) and zinc oxide (see Photo.8) was simultaneously confirmed together with the amorphous carbon by the combined use of the proposed method and electron microscopic observation. These results are summarized in Table I. Moreover, the proposed method can be applied to the analysis of amorphous substance including light elements such as nitrogen and oxygen besides carbon.

View full abstract

The method for the quantitative determination of benzidine by thin layer chromatography in direct dyes was described.
Direct dyes were dissolved in 500ml water using 1000ml separating funnel and o-tolidine was simultaneously added as an internal standard. The solution was extracted with 300ml of ether three times by shaking for 10 minutes. The fraction of ether layer containing the extract was concentrated to about 10ml in vacuo below 45°C. The residue was transferred into a 500ml separating funnel, then 50ml of diluted hydrochloric acid solution (pH 1.5) was added and shaked for 10 minutes, where ether soluble matters were removed. Extractions with 150ml of ether, in addition, were repeated three times.
The pH of residual acidic solution was adjusted to 8.5 with ammonia water and the extraction was repeated with 150ml of ether three times by shaking for 10 minutes. The ether containing the extract was concentrated to about 1ml in vacuo and the extract was spotted on thin layer plate.
Separation of the extract into the components using one or two-dimensional thin layer chromatography was carried out with cyclohexane-chloroform-ethyl alcohol-acetone (17.5:40:2:1, v/v), ethyl ether chloroform-acetic acid (20:20:0.4, v/v) developing solvents.
After development, the plate was sprayed with Ehrlich's reagent, and standing in the dark for 10 minutes.
The plate was directly scanned by a densitometer at 440nm. The area of colored part of thin layer was calculated by the counter, and then the quantity of benzidine was determined in comparison of area ratio with the previously scanned calibration curve of an internal standard.
Twelve different types (17 samples) of direct dyes and acid dyes produced at home and at foreign products were analyzed according to the proposed procedure. Benzidine was qualitatively analyzed for all samples, and quantitative analysis for benzidine was made for 12 samples of 7 different types of dyes.
The coefficient of variation of the data obtained by the above mentioned method did not exceed 9.0 percent.
The present procedure can be successfully applicable to the analysis of benzidine in direct dyes.

View full abstract

The computer-aided analysis of mass spectra of small sized organic compounds was carried out. The pattern recognizers, for this purpose, were prepared to distinguish alkanes and alkenes from many other types of compounds by the input of low resolution mass spectra.
First of all, the mass spectra of three hundred and sixty-two compounds consisted of C, H and O within the scope shown in Table I were collected and the ten most intense peaks in each spectrum were extracted (F₁F₆₆). On the other hand, the collected spectra were classified into seven patterns (S₁S₇) by the clustering. Then the interdependence between the types of compounds (ak) and the patterns (S₁S₇)was established as can be seen from Table II. Thus the appearance probabilities (P) and the weights of appearance probabilities (W) of F₁F₆₆ were given as shown in Tables III and IV, respectively. The recognizers of compounds belonging to S₁ (mostly alkanes), S₂ (mostly alkenes) and so forth were prepared with the weights of appearance probabilities (W), m/e values and the highest (H) and the lowest (L) relative intensities of the ions F₁F₆₆ (Table V).
By feeding the low resolution mass spectrum of an unknown compound within the scope shown in Table I to the computer, the spectrum is contrasted with seven recognizers (Table V) and the score (total of W) (G) for each recognizer is calculated. After that, the type of compound is deduced from the pattern which acquired the highest G value. The similar computation is repeated to predict the types of alkyls and alkenyls by the use of the second recognizers {s₁₁₃ (Table VII) and ss₁ss₆ (Table IX)} which were prepared by the same procedures as the recognizers for the prediction of the types of compounds. For instance, the contrasting mass spectrum of an unknown compound, m/e 27 (354), 29 (425), 39 (187), 41 (456), 43 (1000), 55 (140), 56 (340), 57 (472), 70 (374), 71 (468), (numerals in parentheses express the relative intensities), with seven recognizers showed the highest G value, 764, for the pattern S₁, therefore it was concluded that the compound should be alkanes or alkane-like compounds (Table X). Furthermore, the presence of sec-butyl group was also suggested by the comparing the same spectrum with thirteen recognizers (Tables VII) for the prediction of the types of alkyl groups (Table XI).
As a result of testing seventy-nine alkanes and sixtyfive alkenes, the types of compounds, alkyl groups and alkenyl groups of them were predicted satisfactorily and appropriately by the above-mentioned method.

View full abstract

In spite of its higher sensitivity, the extraction of copper(II) with dithizone-chloroform from an acidic solution is slow, while that with oxine is rapid. The ligand-exchange of the extracted copper oxinate with dithizone in chloroform was attempted and this exchange reaction was found to proceed rapidly and to be available for the determination of trace amounts of copper.
The procedure was as follows: An aqueous solution containing the copper(II), not more than 18 μg, was taken in a 50ml separating funnel; pH of the solution was adjusted to 3.13.5 with 2ml of 1M acetate buffer solution; 2ml of 0.01M oxine aqueous solution was added; the solution was diluted to about 20ml with water and 10.0ml of chloroform was added; the mixture was shaken for 30 seconds; the chloroform phase was transferred into an Erlenmeyer flask and dehydrated with anhydrous sodium sulfate; a 5.0ml portion of the chloroform phase was transferred into a 10ml measuring flask and 2.0ml of about 1.5×10^-4M dithizone chloroform solution was added; the mixture was shaken and diluted to 10.0ml with chloroform; the absorbance (A₁) was measured at the wavelength of 610nm against chloroform; the absorbance (A₂) of a reagent blank was measured at the same wavelength; the amount of copper was proportional to the difference between A₂ and A₁.
The calibration curve was linear from 0 to 18μg of copper in a sample solution. The presence of Ga, In, Zr, Al, Mo (VI), Co, Zn, Bi and U (VI) did not interfere, but Fe(III), V(V) and Hg(II) showed their interference. This interference could be removed by adding 1ml of 20% hydroxylamine hydrochloride solution in the sample solution. As small as 10μg of copper in the sample solution and 0.03% copper in an aluminum alloy could be successfully determined by this ligand-exchange method.
The ligand-exchange reaction seems to be: Cu(Ox)₂+2H₂Dz=Cu(HDz)₂+2HOx. For a quantitative transformation of copper oxinate into copper dithizonate, the following equation should satisfied:[Cu(HDz)₂]/[Cu(Ox)₂]=K[HDz]²/[HOx]²≥100, where; K is the exchange equilibrium constant. The value of log K was reported by J. Stary as to be 7.27±0.03 by the spectrophotometric method and the present author obtained 7.38±0.08 with same method. Therefore, copper oxinate was quantitatively transformed into copper dithizonate when the oxine and dithizone were used in their specified amount by the procedure.

View full abstract

Although a suitable composition of mixed stationary phase for the separation of components may be conveniently estimated by the investigation of the differences of retention indices of two components adjacent to each other, it is also practically necessary to consider the peak resolution which is influenced by the column efficiency. Therefore, considering not only peak distance but also peak resolution, in this paper, the program of the determination of an optimum composition to satisfy the separation of all components was made for a small computer and its applicability in practical analysis was investigated. PDP 8/I high speed digital computer (Digital Equipment Corp., core memory; 4, 096 words×2) was used and the program was written by means of FORTRAN language.
The determination of an optimum composition was performed in the following five procedures; (I) supply of the data for computating, i.e., the gradient (A) and intercept (B) of linear line which held between the retention index (I_R) and the composition (weight fraction, X) of mixed phase, (II) calculation of I_R at X using equation (1); I_R=AX+B and rearrangement of them in ascending order, (III) calculation of number of theoretical plate (N) required to separate two components adjacent to each other at X according to a specified peak resolution (R) using equation (15); N =[2R {exp (2.303I_R1C)+exp (2.303I_R2C)}/{exp (2.303 I_R2C)-exp(2.303I_R1C)}]², where C is coefficient for calculation, (IV) determination of an optimum composition by comparing N obtained in the procedure (III), and (V) type out of the computation results.
Equation (15) used in the procedure (III) had been deduced from the relation between R and retention times (t_R) or peak widths (W) of adjacent peaks, the relation between N and t_R or W, and the relation between t_R and I_R. Although N decreased as the value of C increased, it was recognized from experimental results that there was no change of C accompanied by the change of the composition to influence N under a fixed temperature. Therefore, it was quite reasonable to consider C to be wholly constant.
If the column efficiency is high enough to separate two peaks with the most close t_R out of all peaks, all of other components can be separated. Based on this logic, in the procedure (IV), N(max.) required to separate the most close components was found out at each composition and the composition when. N(max.) becomes minimum was determined as an optimum composition.
This program was applied to the presumption of separation of alkylphenols, alkylanilines, or alkylnitrobenzenes on mixed phase of silicone rubber SE 52 (SE)/polyethylene glycol 20M (PEG). As it was clarified from these results that 9 components of ethylphenols might be separated with R=1.00 at PEG 0.49 if the column of which N is 1, 160 was used, the author made a trial of separation of these components using 2m. column with SE 0.51/PEG 0.49 mixed packing practically, and recognized that the results obtained in this test gave fairly good agreement with the presumption.

View full abstract

A method for the determination of furamethrin (5-propargyl-2-furylmethyl-dl-cis, trans-chrysanthemate) in mosquito coils has been developed by gas chromatography. Usually insecticidal pyresroids, allthrin and proparthrin for example, were extracted with acetone by Soxhlet extractor or with 1% furfuryl alcohol in acetone solution by shaking and determined by gas chromatography. The more organic solvents used for the extraction of furamethrin were polar, the more furamethrin was extracted from mosquito coils by Soxhlet extractor, but furamethrin was not completely recovered. The extraction efficiency was raised when powdered coils were mixed with a mount of powdered and activated charcoal and the mixture was refluxed with organic solvents. The highest recovery of furamethrin was obtained by use of acetone as a solvent and more than 95% of theoretical contents of furamethrin was extracted from coils with coefficient of variation of 2.3%. Addition of n-hexane to acetone was effective to diminish the extracts from coils blank, which was easier to operate after extraction. Furamethrin was decomposed on usual silanized column packing coated with liquid phase and the base line drifted on the gas chromatogram. When 3% sodium borate was coated on the support, the decomposition of furamethrin was avoided and a fine chromatogram was obtained.
The recommended procedure was as follows. Powdered coil (4050 mesh through) containing about 12mg of furamethrin weighed and 1g of powdered and activated charcoal were mixed well and the mixture was refluxed with 100ml of n-hexane: acetone (1+1) mixture for 3 hours in water bath of 80°C by Soxhlet extractor. Allethrin was added as an internal standard, the solvent was removed in vacuo and the resinous matter was dissolved in 1ml of acetone. The furamethrin was determined by gas chromatography under the following operational condition. Apparatus: Yanagimoto GCG-550 with FID, column: 0.75m in length and 3mm in inner diameter glass column, column packing: 6080 mesh, acid-washed and silanized Chromosorb W coated with 3% sodium borate and 15% silicone DC HV grease, column temperature: 150°C, injection temperature: 150°C and carrier: nitrogen 20ml/min. Recovery of furamethrin extracted with methanol by shaking was a few percent lower than that by the method mentioned above. However the time necessary for extraction was shorter and this method was also useful for routine analysis.

View full abstract

New device for the determination of trace amount of mercury in rocks and soils as developed for use in our labaratory is a modification of the method reported by Vaughn and McCarthy.
In our system, sample taken in a porcelain boat is heated with electric furnace and the mercury is vaporized and carried up to the gold trap wire by an air stream. Only mercury is selectively trapped on the gold wire making amalgam, and decomposition gases resulted from burning of sulfides and organic materials, and water pass on out the system through the part of gold trap and to the exhaust pipe.
After removal of interfering gases, an air stream is stopped at once, the mercury trapped on the gold wire is vaporized by heating the gold with electric furnace, and the resulting mercury vapor is carried into the absorption cell by starting the flowtation of air stream. A thermocouple equipped on the gold trap control the starting time of air flowtation.
As a result of traverse of the mercury vapor through the absorption cell, atomic absorption signal of mercury at 2537 Å sharply increase to a maximum and quickly return to zero.
The gold trap is allowed to cool to room temperature with an air blow and set up to the another sample.
A number of parameters have been investigated which affect the determination of mercury in this system.
Mercury in rocks and soils is completely vaporized as monoatomic gas by heating with preheated electric furnace at 500°C, within 3minutes. 80μφ pure gold wire used for trapping the mercury was effective. Only 1.5 gram is an adequate amount in the analytical range 1100ng of the mercury. About ten times larger square plates of gold have been used by early investigators.
Adjusting the weight of sample taken, 11000ng of the mercury in rocks and soils has been determined routinely. Coefficient of variation is about 5%. Detection limit using 1 gram sample is 1ng. Time required for the analysis of cne sample is within 6minutes.

View full abstract

The separation of caffeine, acetanilide, phenacetin, and acetylsalicylic acid by taking advantage of the difference of affinity to strongly basic resins AG-I, X 8 (100200mesh) in acetate form, has been investigated. The capacity of adsorption by a batch method is increased as follows; caffeine, acetanilide, phenacetin, acetylsalicylic acid.
Caffeine is not adsorbed at all, acetanilide and phenacetin are a little, acetylsalicylic acid is much adsorbed. Acetanilide and phenacetin were not separated by means of ethanol (520v/v%) and dilute acetic acid (0.21N) or their mixture in the column operation. The separation of acetanilide and phenacetin was possible by using water as eluate, and column of 35cm (inside diameter 9mm), but phenacetin was eluted by using mixed solution of ethanol and dilute acetic acid (0.2N) in order to avoid tailing.
The aqueous solution of acetylsalicylic acid is slowly hydrolyzed, and it gives salicylic acid and acetic acid. The acetylsalicylic acid was eluted with 6N acetic acid as eluate, but salicylic acid was not eluted, this was applied to its separation.
The mixed sample (each 12mg) was adsorbed on the column (9×350mm) at the flow rate is 0.5ml/min. The water as eluate was passed through on the column for eluting caffeine and successively, acetanilide. The mixed solution of ethanol and 0.2N acetic acid (2:8) was then applied for eluting phenacetin, in the last, acetylsalicylic acid was eluted with 6N acetic acid.
The components in effluent after each fraction (10ml) in a single volume were determined by u.v. spectro-photometry, that is, the absorbance of caffeine was measured at wavelength 270nm, acetanilide at 235nm, phenacetin at 245nm, and acetylsalicylic acid at 275nm respectively against reagent blank and the determination was made by using their calibration curves.
The recovery of each component was 99105% in means (n=8) and coefficient of variation was ranged from 2 to 7%.
The separation and determination of aspirin(acetylsalicylic acid), phenacetin, and caffeine powders of J.P. (VIII) were performed by this method, and good results have been obtained.

View full abstract

Iodide ion-selective electrodes of the liquid membrane type using ion associates between iron(II) chelates of o-phenanthroline derivatives and iodide ion as the liquid membrane active substance have been prepared, and the performance of the ion electrode with the liquid membrane and the basic properties of the liquid membrane have been investigated.
The iodide ion electrode shows a Nernstian linear relationship between the membrane potentials and the logarithmic activities of iodide ion beyond the concentration of 1×10^-3M, but the membrane potentials decrease in the lower concentration range. The liquid membrane potentials in this range increase in the order of 2, 2'-dipyridyl<o-phenanthroline<bathophenanthroline depending upon the ligands combined with ferrous ion and are also dependent on the concentration of the ion associate in the liquid membrane. The sensitivity of ion electrode with the liquid membrane is larger in the order of the sort of the anions I^-<SCN^-<ClO^4-. Nitrobenzene is more sensitive as the liquid membrane solvent of ion electrode than 1, 2-dichloroethane or chloroform. These results may suggest that the membrane potential of ion electrode is mainly dependent on the extractability of the liquid membrane active substance, and the observed liquid membrane potentials are accord with the values calculated by considering the distribution ratio of liquid membrane active substance, transport number of iodide ion and liquid junction potential, etc.
The selectivity of ion electrode of the liquid membrane type is also closely concerned with the extractability of ions, it is found that extractive equilibrium constants are relative to the selectivity ratios obtained from measurements of liquid membrane potentials. The selectivity ratios are independent of the liquid membrane active substances in the case of nitrobenzene as a liquid membrane solvent, but the use of chloroform increases the selectivity ratios of halide ion to nitrate ion. The ion-selective electrode of liquid membrane type described in this paper is easy to prepare and work quite well for measurements of iodide ion.
Measurements of electric conductivity under the condition where the liquid membrane of ion electrode is used indicate that the liquid membrane active substance in the nitrobenzene dissociates to divalent iron (II)-chelate cation and iodide ion, respectively.

View full abstract

A method for the determination of three isomeric toluenesulfonic acids by nuclear magnetic resonance technique was investigated, and the application of this method to the determination of three isomeric sodium toluenesulfonates was also examined.
The NMR measurements were made with a JNMC-60HL spectrometer at a frequency of 60MHz and RS-201DY recorder. The temperature in the spectrometer probe was 30°C during the measurements.
The NMR spectra of toluenesulfonic acids and sodium toluenesulfonates in water gave three sharp singlets for the methyl peaks of each isomer. The chemical shifts of methyl protons of three isomeric toluenesulfonic acids showed interesting behaviour for the concentration dependence of water solution, that is, ortho methyl protons were shifted to low field and meta and para methyl protons to high field with increasing concentration. This concentration dependence of ortho methyl protons may be caused by the strong polar effect of SO₃H and those of meta and para methyl protons by hydrogen bonded systems of associated molecules between SO₃H and H₂O, or SO₃H and SO₃H. From the concentration dependence of the resonance, the mixtures of isomeric toluenesulfonic acids were found to give NMR spectra including the methyl peaks sufficiently resolved for the determination of each component at 40 weight % water solution. At this concentration the differences of chemical shifts between methyl peaks of meta and para and between those of ortho and meta were 0.05ppm and 0.41ppm, respectively.
Two good calibration curves between meta and para isomers and between ortho and para isomers were obtained by using these three methyl peaks measured with eight different kinds of known mixtures in 40 weight % solution.
This method was applicable for the determination of isomeric sodium toluenesulfonates, by using the concentration of 30 weight% water solution, with an average error of about 1%. The method was free from the effect of sulfuric acid in toluenesulfonic acids. According to the result of the analysis of a commercial sodium p-toluenesulfonate by this method, it was found that the isomers of the ortho, meta, and para being 28.3%, 6.6%, and 65.1%, respectively.
This proposed method would be more rapid and simpler than the other methods.

View full abstract

The authers propose a method for gravimetric determination of sulfur, which replaces hydrochloric acid and barium chloride by perchloric acid and baruim perchlorate, then they propose a new method which combines the above method with the modified method of Low-Pearce-Bennet for arsenic determination. The procedure of this method is as follows. A sample solution which contains sulfur and arsenic is alkalized by sodium hydroxide and oxidized by hydrogen peroxide. Then sulfur is oxidized to sulfate ion and arsenic is oxidized to arsenate ion. The above solution is made up to 200ml and acidified by perchloric acid to an acidity of 0.05N. Then the sulfate ion is precipitated with barium perchlorate in the solution containing arsenate ion. The produced percipitate of barium sulfate is filtered, washed, ignited and weighed.
For the determination of arsenic which exists as arsenate ion in the solution, the following procedure is continued. About 1ml of acetic acid is added to the above filtrate solution, and using phenolphthalein as indicator, the solution is neutralized by sodium hydroxide. Then acetic acid is added until the solution becomes colorless, then one or two drops of an excess acetic acid is added, and the solution is heated. After cooling, arsenate ion is precipitated with silver nitrate. The washed precipitate of silver arsenate is dissolved in nitric acid. The amount of silver is titrated with potassium thiocyanate (Volhard's method). The amount of arsenic is determined from the amount of silver.
In order to ascertain the above procedure, the authers have made some experiments on the effect of concentration of perchloric acid and barium perchlorate upon the analysis of solutions containing 40mg of sulfur and 40 mg of arsenic. In a wide range of the acidity of perchloric acid, 0.025N to 0.2N, excellent results are obtained for either of sulfur or of arsenic. In this case, 1.7times of barium perchlorate was used to precipitate sulfate ion. Up to 7.5times of excess barium perchlorate had no interference on the determination of sulfur and arsenic when this proposed method was carried out. For various concentrations of sulfur and arsenic, this procedure is also available in a better accuracy.
Compared with other methods, the characteristic of this method is as follows.
1) easiness of the procedure,
2) necessity of only a small anount of sample for analysis,
3) accuracy of silver arsenate determination.
This characteristic is due to no coprecipitation of sulfate ion with silver arsenate, which is possible by the previous elimination of sulfate ion.
This procedure has been applied to a successive determination of sulfur and arsenic in chalcogenide glasses and good results are obtained.

View full abstract

Elemental analyses of organic compounds using combustion by a gas injection nozzle give a highly efficient oxidative decomposition of a sample and combustion can be effected in a short period of time. The nozzle-type combustion system published to date includes a system in which gasified sample is injected into an oxygen stream and a system in which oxygen is injected into a gasified sample stream. Comparison of these two systems showed that, in the case of the former, the sample sometimes carbonizes in the capillary of the nozzle and block it up, although such a phenomenon was not witnessed in the latter system. Back-wash of combustion gas has a tendency to appear in both cases when the amount of a sample becomes larger but is found more often in the former case.
Oxidative efficiency of the nozzle was compared with the following four types: (1) single nozzle with narrowed outlet, (2) two single nozzles connected in series, (3) multiple capillaries sealed into a nozzle, and (4) two capillary-type nozzles connected in series. (2) was better than (1), (3) was better than (1), and (4) was better (3). Thus, multiple number of nozzles seemed to be better than one nozzle, and the capillary type seemed better than a single nozzle. therefore, three capillary-type nozzles connected in a series was examined for oxidative decomposition of a large amount of a sample with good efficiency and in a short period of time.
The samples used were standard substances for elemental analysis of organics containing halogen or sulfur, diluted with glucose, antipyrine, or liquid paraffin to prepare standard substances containing 0.10.005% of halogen and sulfur. The time required for combustion of 0.10.5g of the sample was 15min. In the case of an explosive sample, combustion was carried out manually, over a protracted time. Sulfur oxides were determined by absorption into 5% hydrogen peroxide solution and titrated with 0.02N barium perchlorate, using Thorin as an indicator. When halogen was to be determined at the same time, the absorption solution was boiled with a platinum net, cooled, and sulfur determined by titration with barium perchlotate, using Thorin as an indicator, Then a silver electrode and reference electrode were placed in the sample solution and halogen determined by potentiometric titration with 0.002N silver nitrate solution. Concurrent determination was successful with standard samples containing 0.10.005% sulfur and chlorine or bromine. Presence of fluorine and phosphorus interfered in this determination.

View full abstract

Characteristic effects can be observed when water or alcohol soluble oxidants are mixed with green suspensions of ferrous hydroxide. The same holds for nitro or nitroso compounds as manifested by a change from green to brown, due to the formation of ferric hydroxide. However, the color change from green ferrous hydroxide to brown ferric hydroxide is not sensitive.
In this study ammoniacal ferrous salt was colored with dimethylglyoxime and was disappeared by nitro compounds.
A drop of the test solution was mixed in a small test tube with ferrous salt solution and alcoholic dimethylglyoxime, and neutralized by ammonia, then allowed to stand a few minutes. Red color of ferrous dimethylglyoxime complex disappeared if nitro compound is present.

View full abstract

Reduction of trace water in N, N-dimethylformamide (DMF) by the use of liquid sodium amalgam (Na-Hg) and potentiometric and conductimetric titration with acetic acid was investigated.
It was well known that sodium amalgam reacted with water and formed equivalent sodium ion. In this report, therefore, DMF solution after reduction by sodium amalgam was titrated by potentiometry or conductimetry using acetic acid dissolved in DMF as a standard titrant Na-Hg reacted with H₂O but not with DMF owing to basic of DMF. DMF was stirred. with Na-Hg about 5 minutes and amalgam was separated. A separated DMF solution was titrated with acetic acid dissolved in DMF.
As the results, less than 10^-3% H₂O in DMF was measured accurately in comparison with Karl-Fischer method.
This method was applied to determine the H₂O in other solvent such as petroleum ether, MIBK, carbon tetrachloride, and acetone, and obtained satisfactory results.

View full abstract

A rapid routine analytical method for the determination of the amount of nickel on nickel plating specimen has been required for the control of the process of nickel plating plant. Therefore, the authors devised a simple method which treated the sample in a special dissolution vessel with mixed acid and the amount of nickel was determined by using high sensitive nickel test paper. In order to improve the accuracy of determination, the standard color scale of various gradation in the range of about 5 to 500ppm was prepared at an interval of 5 to 20ppm according as nickel content.
The procedure is as follow:
Prepared sample is fixed in a special vessel for dissolution and the nickel plating sheet is treated with 5ml of mixed acid. The solution is transfered into a 100ml measuring cylinder. The solution obtained contains nickel and ferric ion. In order to mask ferric ion and to control pH, potassium fluoride crystal is added until a white precipitate appears. The solution is diluted to 50ml with water and then nickel test paper is dipped into the solution. The resulting color of the test paper is compared with the standard. color scale and nickel content is determined.
Analytical results of this method agreeded with those of EDTA method. The time required for the analysis was about 3 minutes for one sample.

View full abstract