Precipitation of Oxalates from Homogenous Solution - Analytical

Gordon, and E. R. Caley. Anal. Chem. , 1948, 20 (6), pp 560–563. DOI: 10.1021/ac60018a019. Publication Date: June 1948. ACS Legacy Archive. Cite thi...
0 downloads 0 Views 558KB Size
ANALYTICAL CHEMISTRY

560 After the filtration is completed the rubber bands holding the cylinder are disconnected and the cylinder is carefully raised. Any particles adhering to the lower edges are recovered with a fine brush. The suction is then cut off, and the filter ring is removed and stored in a desiccator at 50% relative humidity for a t least 12 hours before weighing.

Table 11. Radioactivity Measurements of Magnesium Ammonium Phosphate Precipitates Containing Varying Amounts of P31and Paa Relative P * z Concentration Taken

Psi Taken, Mg. 10 20

1

Table I. Recovery of Phosphorus Weighed on Filter Rings as Magnesium Ammonium Phosphate Hexahydrate Phosphorus Taken 1Mg.

hlg”1PO1.6H20 Recovered

Phosphorus Recovered

MQ.

MQ.

2

30 10 20 30 10 20 30

,

3

ActivityQ, Counts/Sec.

8.01 8.44 8.21 15.70 15.95 16.01 23.80 23.76 24.02

a Activity corrected for background, radioactive decay, and r e c w e r y timr

of counter.

TYPICAL MEASUREMEVTY

The weighing of magnesium ammonium phosphate hexahydrate

as a procedure for determining phosphorus is not a commonly accepted practice. However, it has been recommended by Fales (8) and Treadwell and Hall (5). The results given in Table I show the recovery of phosphorus when weighed on filter rings as magnesium ammonium phosphate hexahydrate to be approximately equal to the phosphorus taken and to vary in direct proportion with the amounts taken. The radioactivity measurements of the filter rings containing a Layer of magnesium ammonium phosphate hexahydrate with a Geiger-Xiiller counter should not present difficulties if an arrangement for adequately reproducing the geometry is provided. A shield and sample holder similar to that described by Dauben et ul. (1) has been in use a t this laboratory. Srlf-absorp-

tion was tested on amounts of precipitate up to 300 mg. and, aE was expected, no correction appears necessary. Table I1 gives the radioactivity measurements of nine precipitates prepared to contain three amounts of P32in combination with three amounts of P31. The data show that the radioactivity of given amounts of P 3 2 can be determined with accuracy in the presence of varying quantities of Pal. These data are representative of the degree of reproduction that has been obtained. LITERATURE CITED

(1) Dauben,

\v.

G., et d., IVD.EXG. CHEM.,A N A L . ED., 19, 82s

(1947).

Fales, H. A., “Inorganic Quantitative bnalysis,” p. 222, New York, Centuiy Co., 1926. (3) Henriques, F. C., et al., ISD. ESG. CHEM.,A N ~ LED., . 18, 349 (2)

(1946). (4) (5)

Henriques, F. C . , and Margnetti, C., Ibid., 18, 415 (1946). Treadwell, I‘.P., and Hall, W.T., “Analytical Chemistry,” Vol 11,9th ed , p. 267, New Tork, John Wiley & Sons, 1945.

RECEIVED October 1. 1947.

Precipitation of Oxalates from Homogeneous Solution Application to Separation and Volumetric Determination of Magnesium LOUIS GORDON AND EARLE R . CALEY, The Ohio State University, Columbus, Ohio

THE

separation or determination of certain metals by precipitation with oxalate ion is an old but useful technique. A difficulty frequently encountered in such separations or determinations is the formation of finely divided precipitates that require retentive filter media. As a consequence filtration and washing mav become SIOR- and tedious operations. Though this manipulative difficulty may usually be overcome to a large extent by the familiar technique of adding the preripitant slowly and digesting the precipitated solution for a sufficiently long time, this does not always resolve the difficulty. lJ7hen an oxalate is precipitated by such technique in a medium that is essentially nonaqueous, as in the method for the determination of magnesium described by Elving and Caley (I), the precipitate is still too Bnely divided and causes difficulty in filtration. What is needed therefore is a fundamentallj different technique which will always result in the formation of an oxalate pfecipitate of large particle size that may be filtered and washed with ease and rapidity. For the special case of the precipitation of calcium as oxalate, Willard and Chan ( 2 ) devised a procedure that produces a coarse precipitate that is easily filtered and washed. In this procedure part of the necessary oxalate is added to a calcium solution of such acidity that little or no initial precipitation results. Urea is then added and the solution is boiled gently to cause slow hy-

drolysis of the urea. This brings about gradual decrease in the hydrogen-ion concentration and a corresponding gradual increase in the oxalate-ion concentration, which in turn causes the slow precipitation of calcium oxalate in coarse crystals. The remainder of the necessary oxalate is then added to ensure complete precipitation. For the separation of thorium and most of the rare earths from a solution of a monazite sand, Willard and Gordon ( 3 ) found that the hydrolysis of methyl oxalate was a means of producing a slow increase in the concentration of oxalate ion, and that this technique yielded a precipitate of the mixed oxalates which was of better physical form than that obtained by the usual direct precipitation with oxalic acid. Neither of these techniques for precipitation from homogeneous solution can be applied directly to the quantitative precipitation of certain other metal osalates, such as the precipitation of magnesium oxalate in 85% acetic acid solution for example, though the principle involved in the second one appears to be generally applicable. I n actual experiments on the use of methyl oxalate for the precipitation of magnesium oxalate in, 85% acetic acid solution it was found that this ester decomposed too rapidly, especially at the elevated temperature desirable in this precipitation for the purpose of obtaining the best possible separation from lithium and sodium. However, the resulting precipitate was

561

V O L U M E 20, NO. 6, J U N E 1 9 4 8 Magnesium oxalate may be quantitatively precipitated in a dense and coarsely crystalline form from 85% acetic acid solution by the gradual release of oxalate ion in this medium as the result of the slow decomposition of dissolved ethyl oxalate. In contrast to a precipitate obtained by the direct addition of oxalate ion, one obtained by this technique may be filtered off and washed rapidly. By the use of this special technique of precipitation, the indirect method for the determination of magnesium based upon the solution of magnesium oxalate in dilute sulfuric acid and titration of the released oxalic acid becomes a very convenient and simple procedure.

noticeably superior to that produced by the procedure of Elving and Caley, though not markedly so. Evidently what a,;s needed was a more stable, and yet sufficiently soluble, ester of oxalic acid that would decompose slowly in 85y0acetic acid solution a t an elevated temperature so as to bring about a much more gradual increase in oxalate-ion concentration. After various preliminary experiments, ethyl oxalate was :selected as the reagent that gave the best results under the optimum conditions for the precipitation of magnesium oxalate in this solution. These conditions are (1) the highest possible temperature that can be employed without danger of loss from bumping as the dense precipitate settles on the bottom of the precipitating vessel, and (2) a period of time for the formation of the precipitate of about 2 hours, as this yields a precipitate of very satisfactory character within a convenient working period. The precipitate of magnesium oxalate formed by this reagent under these conditions is strikingly different in physical character from that obtained by the precipitation procedure of Elving and Caley. Instead of being finely divided and voluminous it is coarsely crystalline and dense, and yet it does not adhere to the glass wall of the containing vessel as do some precipitates formed from homogeneous solution. This method of precipitation removes a principal objection to the analytical methods described b y Elving and Caley, as the precipitate of magnesium oxalate may 'be very quickly filtered and washed. These workers found it impracticable to use a filter crucible because it became clogged with precipitate, and for this reason they were forced to use centrifugal separation and washing for their otherwise convenient volumetric method. The precipitate of magnesium oxalate formed by the present method may be rapidly filtered off on a glass or porcelain filter crucible of medium porosity. By thus eliminating the need for centrifugal separation, the volumetric method of Elving and Caley may be made the basis of a n accurate and relatively rapid method for the determination of magnesium that requires no special equipment. The production of oxalate ion by the hydrolysis of ethyl oxalate in 8570 acetic acid is accompanied by a corresponding increase in hydrogen-ion activity, and unless some method of compensating for this is employed the solubility of magnesium oxalate in this medium is so increased that precipitation is not quantitative. The addition of a sufficient quantity of ammonium acetate is a convenient and satisfactory method of automatically controlling the hydrogen-ion activity within the necessary limits. COMPOSITION OF THE PRECIPITATE

Elving and Caley did not investigate the composition of the precipitate formed in 85y0 acetic acid solution. In the course of investigating the possibility of determining magnesium gravimetrically by collecting, drying, and weighing the precipitated magnesium oxalate in a filter crucible, a number of experiments were made on the composition of the precipitate formed in this medium. It might be expected that the precipitate formed a t high temperature in 85% acetic acid would be anhydrous, or at least contain considerably less water than the dihydrate precipitated from aqueous solution. Actually, the dihydrate is also

the salt precipitated from 857, acetic acid solution. However, as shown by the data in Table I, the composition of actual analytical precipitates does not correspond exactly to the theoretical composition of this salt. These precipitates were obtained by the quantitative precipitation of solutions of distilled magnesium (99.9970 purity) by the procedure given below, and were dried to constant weight a t 105" C. The magnesium content in terms of percentage was calculated from the magnesium taken and the weight of dried precipitate. This was found to be a more accurate procedure than analytical determination of the magnesium content of the precipitates by ignition to oxide and weighing. The oxalate content was determined by titration with permanganate; all the usual precautions were taEen to secure high accuracy. Table I.

Experiments on Composition of Precipitates

Precipitate No.

I I1 I11 IV Av. Calcd. for &fgCz04.2HaO

Mg,

CaOi,

16.20 16.22 16.23 16.29

58.55 58.66 58.63 58.83

%

%

16.23

58.67

16.39

59.32

Mg/CzO Ratio 1.001 1.001 1.002 1,002

MgCzO,

70

74.75 74.88 74.86 75.12 74.90 75.61

The ratio of magnesium content to oxalate content is close to theoretical, but the magnesium content, the oxalate content, and consequently the magnesium oxalate content fall appreciably below the theoretical. In other words, dried precipitates weigh more than they should, and high results may be expected in the gravimetric determination of magnesium on weighing such precipitates. In four trials on amounts of magnesium in the neighborhood of 50 mg., the results ranged from 0.3 to 0.6 mg. too high in terms of this element. Results sufficiently accurate for some purposes may, however, be obtained by this simple gravimetric method. The experiments indicated that the extra weight in such dried precipitates is due to small amounts of foreign substances adsorbed on or occluded in the precipitates when they are formed, and not removable by washing. For example, in an experiment where perchloric acid was used in the preparation of the magnesium sample, the almost indispensable procedure in actual practice, perchlorate ion was apparently present in the thoroughly washed precipitate, as was indicated by the presence of chloride ion in a nitric acid solution of the magnesium oxide obtained by ignition of the dried magnesium oxalate. It is likely also that the precipitates contain small amounts of occluded solvent not removable a t a safe temperature for drying the hydrated magnesium oxalate. The precipitate of magnesium oxalate obtained by this method may be ignited t o magnesium oxide for weighing, but the time and trouble involved in this ignition, and the lower accuracy often obtained in practice, make this gravimetric method less desirable than the volumetric. Because of the voluminous nature of the precipitate obtained by their method of precipitation, Elving and Caley found that 25 mg. was the highest amount of magnesium that could conveniently be determined volumetrically, but this limitation is now removed and 100 mg. or even more can be determined by the present method of precipitation. In other words,

ANALYTICAL CHEMISTRY

562 there is now no good reason to prefer the gravimet,ric method to the volumet>ric,and hence only the volumetric method is here described and recommendpd. PROCEDURE

Concentrat,e the neutral magnesium solution contained in a 250-ml. beaker to a volume of 14 to 15 ml., or dissolve thc residue of dried salts containing the magnesium in sufficient water to reach this same volume. Add 75 ml. of glacial acetic acid and stir until solution is complete. Then add 10 ml. of glacial acetic acid containing 1gram of ammonium acetate. Finally add 1.5 ml. of ethyl oxalate and stir thoroughly. Cover the beaker with n watch glass and place it on an electric hot plate so regulated that the solution may be raised to approximately 100" C. and held at this kmperature. From the time precipit,ation begins, heat for 2 hours if the amount of magnesium is 50 mg. or less, and for 3 hours if the amount of magnesium is as high as 100 mg. As a precautionary measure, especially if there is doubt as to the full mainknance of the proper temperature throughout the period of precipitation, add, 15 minutes before filtration, 5 ml. of 85% acetic acid that has been saturated with ammonium oxalat,e at room temperature. Filter while hot through a glass or porcelain filter crucible of medium porosity. Remove the precipitate from the beaker wvit,h the aid of a rubber-tipped stirring rod and warm (70"to 80" C.) 85% acetic acid, preferably delivered from an all-glass, bulboperated wash bottle. After transferring all of it, to the crucible, wash the precipitate with four or five additional portions of warm acid of about 5 ml. each. Rinse out the suction flask thoroughly with distilled water and after replacing the empty flask dissolve the precipitate and wash out the crucible with successive port,ions of warm (80" C.) 59" sulfuric acid until a total of 200 ml. has collect,ed in the'suction flask. Titrate this solut,ion a t once lvitli permanganate solution that has been standardized against sodium oxalate a t a t,itration t,emperature of 70" to 80" C. Notes on the Procedure. Inasmuch as each milliliter of R 0.1000 AT permanganate solution is equivalent to only 0.001216 gram of magnesium it is desirable, in order to avoid the inconvenience of refilling the buret, to take a sample of such size that) not more than 50 to 55 mg. of magnesium are present when an ordinary buret and permanganate solution of the usual concentration are used. For larger amounts the use of a 100-ml. buret or a more concentrated permanganate solution is a convenience. The substances that interfere in this procedure are essentially those listed by Elving and Caley. However, sulfate interferes more in this procedure than in theirs, and oxalate must be entirely absent. I n adapting the present method to the general quantitative scheme the practical procedure after separation of calcium as oxalate is to remove ammonium salts and oxalate completely by treatment with nitric and perchloric acids, and to fume off the excess of perchloric acid, so that there is finally obtained a dry neutral residue of the perchlorates of magnesium and the alkali metals which is then dissolved in the recommended volume of water. Instead of first adding the 75 ml. of glacial acet,ic acid and then the 10 ml. of the 10% solution of ammonium acetat,e in glacial

acetic acid it is equally satisfactory, and perhaps more convenient, to add a t once 85 ml. of glacial acetic acid in which has been d i e solved 1 gram of ammonium acetate. It is essential that the ethyl oxalate be pure diethyl oxalate and that it contain no ethyl hydrogen oxalate such as is formed b y contact of diethyl ovalate with water or moist air. This reagent as purchased from the usual commercial sources has been found to be of satisfactory quality. However, when pure diethyl oxalate is allowed to stand in a partly filled coqtainer that is opened frequently in humid weather sufficient hydrolysis takes place to render the reagent unfit for use. The partly hydrolyzed ester decomposes too rapidlv and causes the formation of a fine prccipitate. Instead'of using a filter crucible and a suction flask for the filtration it is also icasible to use a filter stick, though for work on \ the macro scale this technique is not advantageous. TEST ANALYSES

The results of a series of test analyses on pure magnesium solutions, prepared from vacuum-distilled magnesium of 99.99yc purity, are shown in Table 11. Essentially the same results were obtained on both types of solutions tried and the difference errors were small and of practically the same magnitude regardless of the amount of magnesium determined. In Table I11 are shon n the results of a similar series of test analyses in which magnesium was determined in the presence of varying amounts of lithium. The results are generally satisfactory, although theie is a definite tendency toward high results, especially in certain ranges. Elving and Caley iound that of all the alkali metals lithium interferes most in this general method. However, a comparison of the present results with those obtained by Elving and Caley shows that the interference from lithium is less by the present procedure than by their procedure. I t is likely that there is less coprecipitation of lithium oxalate by the present procedure, both because of the slower rate of precipitation and because of the larger particle size with correspondingly less total surface. Bv using the present procedure for the determination of magnesium i n the presence of lithium double precipitation is less often a iicressitv. Test analyses on the determination of magnesium in thr prcscnce of sodium showed that there is little interferenee

Table 111. Determination of Rlagnesium in the Presence of Li thirini L1

Anion Present

Anion Present Chloride Perchlorate Chloride Perchlorate Chloride Perchlorate Chloride Perchloratn Chloride Perchlorate

M g Found

Gram 0.0010

0.OOOQ

0.0011 0.0010 0.0111 0.0113 0.0225 0.0224 0.0227 0,0203 0,0506 0.0561 0,0562 0.0662 0.0562 0,0562

0.1008 0,1008

Difference, Error Gram -0.0001 - 0.0002 +o. 0001 10.0000 -0.0001 +o ,0001 +0.0001 A0 .oooo + O . 0003 +0.0001 +o ,0001 - 0.0002 -0.0001 -0.0001 -0.0001 - 0.0001 - 0,0002 - 0,0002

0,0505 o 0505

0.1010 0.1010

0,0113 0,0103 0 0507 0.0507 0.1010 0,1012

+0.0003 tO.0002 + O 0002 +o 0002 *o . 0000

0.0563 0.0563 0 ,0563 0.0563 0,0505 0.0505 0.0011 0.0011 0,0112 0.0112

0,0564 0.0565 0.0563 0,0567 0,050Q 0.0510 0.0010 0.0009 0,0115 0.0115

+ O ,0001 + O ,0002 t o . on02 +0.0004 + O ,0004 +0.0005 -0.0001 - 0,0002

Chloride Perchlorate Chloride

0.010 0.010 0,010

0 0112 0,0101

Perchlorate

0,010 0,010

Determination of Magnesium Alone Rlg Taken Gram 0.0011 0.0011 0.0010 0.0010 0.0112 0,0112 0.0224 0.0224 0.0224 0.0202 0.0505 0.0363 0.0863 0.0563 0.0563 0,0863 0.1010 0.1010

Difference, Error Gram

Taken Gram

Perchlorate

Table 11.

hI g Found G Iani

Present Gram

Chloride

Table IV.

o~o.i n.

~ .

0.020 0.020 0,040 0,040 0,070 0,070 0,100 0.100 0,100 0.100

+0.0002

+0.0003 f0.0003

Determination of Magnesium in the Presence of Sodium Mg Taken Gram

Found Gram

Difference Error Grnm

0.050R 0.0505

0.0506 0.0507

+0.0001 +o. 0002

0,100 0.100

0.0101 0.0101

0.0102 0.0103

+0.0001 +0.0002

0,100

0,030.5 0.0505

0.0607 0.0507

+o. 0002 +0.0002

Anion Present

Na Present Gram

Chloride Perchlorate

0.040 0.040

Chloride Pcrchlorate Chloride Perchlorate

0,100

,

Mg

V O L U M E 20, N O . 6, J U N E 1 9 4 8 ' from nioderat,e amounts of sodium, as is indicated by a few typical rcsults given in Table IV. I n ordinary cases double precipitation is not necessary when this procedure is used. In view of the prior experiments of Elving and Caley on the determination of magnesium in samples of actual materials by this general method, it seemed unnecessary to run test analyses by this special precipitation procedure on such materials. Preliminary experiments have bhown that the slow hydrolysis of esters of oxalic acid rnay be applied in a similar way t o the precipit.at'ion of the oxalates of zinc, cadmium, and lead in an

563 easiIy filterable state for the pbrpose of' tlieir qubntitative determination. LITERATURE CITED

*

(1) Elving, P. J., and Caley, E. It., IND.EX+. CHEM.,ANAL.ED.,9, 558-62

(1937).

(2) Willard, H. H., and Chan, F. L., in Willard and Furman, "Ele-

mentary Quantitative Analysis," 3rd ed., pp. 397-99, New York, D. Van Nostrand Co., 1940. (3) Willard, H. H., and Gordon, L., ANAL. CHEM.,20, 165-9 (1948). RE^^^^^^ ootober 22, 1947

Analysis of Simple and Complex Tungsten Carbides JOHN J. FUREY AND THOS. R. CUNNIR'GHAM Electro Metallurgical Company, ,Viagara Falls, N . Y . Jlethods for separating and estimating the principal constituents of tungsten carbides are described. Tungsten is precipitated with cinchonine and a-benzoinoxime after decomposition of sample by fusion with sodium peroxide. Columbium, tantalum, titanium, and iron are separated from tungsten and molybdenum by precipitation with sodium and ammonium hydroxides; and columbium and tantalum are separated from iron and the bulk of titanium by hydrolysis. Columbium and tantalum are precipitated with cupferron; also titanium after separation from iron by

T

HE analysis of tungsten carbide containing tungsten, niolyb-

denum, cobalt, nickel, columbium, tantalum, titanium, chrom i u h , iron, silicon, manganese, and carbon is time-consuming and tedious, and good technique is essential if accurate results are to be obtained. The literature on the analysis of complex tungsten carbides is limited and leaves much to be desired as regards the solution of the difficulties encountered in the analysis of these materials. By the procedures described hereinafter, interfering clcinents are either removed or determined and proper corrections are made. An analysis for the elements mentioned may be made in about 2.5 days. The fusion method for tungsten involves fusion of the alloy with sodium peroxide in a nickel crucible, solution in hydrochloric acid, and addition of cinchonine (or cinchonine-antipyrine) and a-benzoinoxime to ensure the comp1et)eprecipitation of the tungsten and molybdenum, followed by filtration, ignit,ion, and weighing. The precipitate is tested colorimetrically for molybdenum, and for columbium, tantalum, and titanium by precipitation with cupferron in the presence of tartrate ion t o ensure complete separation from tungst,en. h solution method for tungsten involves dissolving the sample in hydrofluoric and nitric acids in platinum, and evaporating to a low. volume, followed by transferring the solut,ion to a beaker containing boric acid, adding hydrochloric and nitric acids, evaporating, and complrting the determination for tungsten as for the fusion method. In the mcthod for rolumt)iuni, tantalum, titanium, and iron, the tungsten and inolybdcnum are removed by precipitation with a very small excess of ammonium hydroxide and filtration after the sulfuric acid solution of the sample has been treated with an excess of sodium hydroxide solution (200 grams per liter), boiled, and acidified wit,h hydrochloric acid. The ammonium hydroxidc precipitate is treated with nitric and perchloric acids, the solution is evaporated to dense fumes of perchloric acid, and columbium and tantalum are hydrolyzed by boiling with dilute hydrochloric and sulfurous acids, and separated from the iron and the greater

hydrogen sulfide. Cobalt and nickel are determined electrometrically after separation from interfering elements with potassium hydroxide. After separation from tungsten and molybdenum with ammonium hydroxide, chromium is determined by ammonium persulfate method. Manganese is determined colorimetrically with potassium periodate after separations involving potassium hydroxide and hydrolysis. Sodium peroxide fusion, followed by dehydration with perchloric acid and purification with sodium pyrosulfate-sulfuric acid, separates silicon, and carbon is determined by combustion in oxygen.

part of the titanium by filtration. The last traces of tungsten are removed from the columbium and tantalum by precipitation with cupferron in the presence of tartaric acid. The iron is separated from the bulk of the titanium by precipitation with ammonium sulfide in the presence of ammonium tartrate and completed by titration with potassium dichromate. The titanium in the ammonium sulfide filt,rate is recovered by precipitation nith cupferron. Cobalt and nickel arc determined clcctrolyt,ically after the tungst.en and molybdenum and the bulk of the tantalum and columbiuni have been removed by a double precipitation with potassium hydroxide and filtration. Cobalt is also determined potentiometrically after a similar precipitation 1Tith potassium 11)-droxide and filtration. The method for chromium consists in removing the tungsten and molybdcnuni by t,reatnient of the sulfuric acid solution of the sample with ammoniuni hydroxide and filt,ration, the destruction of the precipitate with nit,ric and perchloric acids, and t,he completion of the determination by the ammonium persulfate method. In the determination of manganese, the tungsten and molybdenum and the bulk of the columbium and tantalum are removed by treatment of the sulfuric acid solution wit,h potassium hydroxide and filtration. The remaining columbium and tantalum arc separated by hydrolysis and filtration, and the manganese in the filtrat,e is determined colorimetrically by the potassium periodate method. Silicon is determined by fusion of the sample with sodium peroxide, dehydration with perchloric acid, separation from tungstic, columbic, and t,antalic acids by fusion with sodium pyrosulfat,e, and treatment of the sodium pyrosulfate-sulfuric acid solution with diluted t,art,aricacid, followed by filtration and washing. Carbon is determined by combustion with ingot iron and cupric oxide, and the resulting carbon dioxide is absorbed in Ascarite and weighed.