Determination of lanthanum by flame photometric titration

Determination of lanthanum by flame photometric titration

Short communications 978 6. 7. 8. 9. 10. 11. A. Rant, J. P. Cali and H. D. Thompson, Anal, Chem., 1956,28,1867. J. P. Cali, Trace Analysis of Semic...

329KB Sizes 1 Downloads 140 Views

Short communications

978

6. 7. 8. 9. 10. 11.

A. Rant, J. P. Cali and H. D. Thompson, Anal, Chem., 1956,28,1867. J. P. Cali, Trace Analysis of Semiconductor Materials, Pergamon Press, Oxford, 1964. E. Ricci and F. F. Dyer, Nucleonics, 1964,22, No. 6, 45. A. Csiike and I. Peter, KFKl Kozlemen., 1965, 13, 343. D. F. Covell, Anal. Chem., 1959,31,1785. G. Iwantscheff, Das Dithizon und seine Anwendungen in der Micro- und Spurenanalyse, Verlag Chemie, Weinheim, 1958. 12. J. Starf, The Solvent Extraction of Metal Chelates, Pergamon Press, Oxford, 1964.

Tdanta,

1968. Vol.

IS, pp. 978

Determination

to 982. Perramon

Press.

Printed in Northern

Ireland

of lanthanum by flame photometric titration

(Received 15 January 1968. Accepted 16 April 1968) DETAILEDstudies have shown that in some flame photometric

interferences there is a linear relationship between emission and the concentration of interferant, up to a limiting value of the latter, and no change in emission at higher concentrations of interferant. The limiting concentration usually corresponds to a simple stoichiometric compound formed from the two reactants. This may be called a stoichiometric interference, and has been studied by Fukushimal who tried to explain the nature of these apparently chemical reactions. Yofe and Finkelstein* even applied the laws of chemical equilibria to such phenomena, Erdey et al.” described a method for the determination of phosphate, based on the shift caused in the equilibrium by addition of a third reactant, and Erdey and Svehla’ determined calcium by “flame photometric titration”. Although not free from interferences, such a titration offers an increased selectivity because the wavelength of emission is characteristic of the emitting species. In a study of the flame photometry of lanthanum we found another stoichiometric interference and therefore examined the possibility of titration of lanthanum with phosphate. EXPERIMENTAL Solutions Lanthanum chloride. Stock solution, 0.lM, prepared from lanthanum chloride heptahydrate and standardized complexometrically, and further diluted to 0.OlM, O*OOlMand 1000 ppm of La. Diammonium hydrogenphosphate,O*lM. Further diluted to O*OlM and OOOlM. Zirconyl chloride, thorium chloride, cerium(III) chloride and aluminium chloride, @lM, Prepared from the hydrated salts. Titanium(ZZZ)chloride, O*lM, prepared from commercial titanium chloride solution. Yttrium chloride, 0.lM, prepared by dissolving yttrium oxide in hydrochloric acid.

The Unicam SP 90 combined emission and absorption flame photometer was used with an acetylene-air flame. Recommended settings are, acetylene flow-rate 900 ml/min, air flow-rate 5 l./min, wavelength 560 rnp, slit-width 0.1 mm, burner height 2 cm, lowest damping. Consumption of sample The rate of consumption of solution under the conditions used is about 4.5 ml/min. The minimum time needed for full g&vanometer response to a change in signal was 1 set at the lowest damping setting (1). We read the maximum deflection after 1.2 sec. With 12 readings per titration the total consumption was about 1 ml, i.e., 2% of the initial volume (50 ml). Since most of these readings were near the end-point the negative error was reduced because most of the solution had already If the titration is repeated with as few readings as possible before the end-point been “titrated”. this error can be decreased considerably. interference by phosphate with lanthanum emission The suppressing effect of phosphate on the Bame emission of lanthanum was reported by Menis and Rains,5 and Yofe et al.%but its stoichiometric nature was not pointed out. We prepared a set of O*OlM lanthanum solutions each containing 10 ml of hydrochloric acid (1 + 1) per 100 ml to prevent precipitation. Various amounts of phosphate were added and the solutions were diluted to tinal volume; the final phosphate concentration ranged up to 0.02M. The emission of each

979

Short communications 100

It\

90

80

70

60

50

LO

30

20 0

10

0

0

I

I

I

I

-005

-01

.015

l02

cPO,l

FIG. 1.-Variation

of lanthanum

M

emission at 560 rnp with phosphate

concentration.

solution was measured at gain 54 and damping 2 and the readings were taken when steady. The results, plotted in Fig. 1, show a sharp intersection, termed the stoichiometric point, which corresponds to a 1:l molar ratio between lanthanum and phosphate, indicating formation of LaPO,. These. conditions are suitable for a flame photometric titration. Titration of lanthanum with phosphate Because of the design of the instrument used we were obliged to remove the sample from it between additions of titrant but this did not cause much inconvenience. We added the titrant in suitable increments and took galvanometer readings after each addition. Titration curves obtained with 0.1 and O.OlM solutions are shown in Fig. 2. The end-point is reached when there is no further change in the emission and can be determined quite easily without the titration curve being drawn. The results of titrations of 0.1, 0.01 and OGOlM lanthanum, shown in Table I together.with those for “unknowns,” indicate that accuracy and precision are both reasonable; the simplicity and rapidity offer advantages over other methods. The titration gives more precise results than the direct flame photometric determination of lanthanum, but reproducibility decreases considerably if ONtlM titrant is used. At the start of the titration the gain has to be adjusted to yield a full-scale reading (9&100°~. The readings near the end-point then become very low (say 10 %) but can be. increased by increasing the gain. The sensitivity is thus enhanced and the accuracy of end-point detection increased (Fig. 2). Effect of foreign ions We examined the effect of those metal ions which seriously interfere with the complexometric titration, namely cerium(III), yttrium, zirconium, titanium(III), thorium and aluminium, using them in about the same concentration as the lanthanum. Titrations with O*lM titrant showed that cerium

980

100 4

Short communications

-(cl

100

(b)

10 /

0

5

T1tran.t

FIG. 2.-(u)

20

15

10

25

[ml)

@-Flame photometric titration curve of 2000 ml of 0*1&f L&I8 titrated with O*lM (NHI),HPOI. (b) O-Flame photometric titration curve of 2000 ml of O*OlM L&Is titrated with O*OlM (NH4),HP04. (c) A-Flame photometric titration curve of 20430 ml of 0.1&f IA& titrated with O*lM (NH,),HPO,, increased sensitivity being used in the vicinity of the end-point.

981

Short communications TABLE I.-RESULTS OF FLAMEPHOTOMETRIC TITRATION OF LAL.&Is solution taken ml M 0.1

20*00

M 0.1

15*00 10.00 5.00 0.01

20.00

0.01

15.00 10.00 5.00 O-001

0-l

20.00

‘LUnk?lowns” 17.5 8.3 8.2 15.6 6.7 14.3

0.001

0.1

Titrant used ml 20.0 20.3 20-o 14.8 15.0 10.0 10.0 5.0 5.0 20-o 19.2 19.4 14.0 14.5 10.0 10.7 ::z 19.0 19.5

Std. devn., ml 19.6 19.8 19.5 15.5

0.3

10.25 5.15 19.0 20.4 20.5 14.5

0.6

10.0 49 18.0 21.4

17.4 8-O 7.8 15.5 7.0 14.1

and yttrium (and presumably other tervalent rare earth metalions) seriously interfere, but titanium(LI1) and zirconium show no interference at all. If thorium and aluminium are present the horizontal of the titration curve becomes oblique but is still a well-de6ned line: in that case, the end-point can only be determined reliably if the titration curve is plotted, Procedure Take a solution containing 70-300 mg of lanthanum, acidify it with 10 ml of hydrochloric acid (1 + 1) and dilute it with water to 50 ml. Titrate the solution with 0.1M diammonium hydrogen phosphate with constant stirring and take Same photometric emission readings from time to time, with the gain set to give initial full-scale deflection, until there is no further decrease in the emission readings. Determine the end-point graphically or from the readings themselves. If possible, carry out a second titration, adding about 80% of the titrant in one batch, before starting to take emission readings, with the gain set to give full-scale deflection at this point. One ml of 0.1M diammonium hydrogen phosphate is equivalent to 13.9 mg of lanthanum. Solutions containing 7-70 mg of lanthanum can be titrated in a similar way with O*OlM titrant but with reduced accuracy. Acknowle&ements-The and to the Government Chemistry Department Queen’s University Belfast, N. Ireland

authors wish to thank Professor C. L. Wilson for his interest in this work, of Northern Ireland for a research grant. G.SWHLA P. J. SLEVIN

982

Short communications Summary-The flame emission of lanthanum at 560 rnp decreases linearly with phosphate concentration until a 1:l molar ratio is reached, and then remains practically constant. Lanthanum can be titrated with phosphate, the equivalence point being detected from the change in emission intensity. Errors due to consumption of solution by th; atomizer can be kepclow by using short spraying times and low galvanometer damping. The average error is about -1% for O-1M solutions and less’th& -5 % for %*OlM. The method’ gives good results in the presence of titanium(III), zirconium, thorium and ahuninium but cerium(II1) and yttrium seriously interfere. Zusammenfassnag-Die Flammenemission von Lanthan bei 560 nm Wilt linear mit der Phosphatkonzentration ab bis zum Molverhlltnis 1 : 1 und bleibt dann praktisch konstant. Lanthan kann mit Phosphat titriert werden, wobei der Aquivalenzpunkt aus der Intensitatslnderung der Emission bestimmt wird. Fehler, die auf Lbsungsverluste im Zerstauber zuriickgehen, lassen sich mit kurzen Sortihzeiten und geringer D&npfuni des Galvanometers klein halten. Der durchschnittliche Fehler ist bei O.lM Lijsuneen etwa -lo/, und bei 0.01 M Liisungen weniger als -5 o/b. Die Merhode liefert in Gegenwart von Titan(III), Zirkonium, Thorium und Aluminium einwandfreie Werte; Cer(II1) und Yttrium stijren dagegen betrachtlich. R&nn&-L’emission de flamme du lanthane a 560 rnp d&xoPt lin&airement avec la concentration en phosphate jusqu’a ce qu’un rapport molaire 1:l soit atteint, puis reste pratiquement constante. On peut titrer le lanthane par le phosphate, le point d’equivalence &ant detect6 par le changement dans l’intensite d’emission. Les erreurs dfies a la consommation de solution par l’atomiseur peuvent &tre maintenues faibles en utilisant de courts temps de pulverisation et un faible amortissement du galvanometre. L’erreur moyenne est d’environ - 1% hour des solutions 0,lM et de moins de -5 “/, Dour O.OlM. La bethode donne de b&s resultats en la presence-de titane (III), zirconium, thorium et aluminium, mais le c6rium (III) et l’yttrium interferent serieusement. REFERENCES

1. 2. 3. 4. 5. 6.

S. Fukushima, Mikrochim. Acta, 1959, 596. J. Yofe and R. Finkelstein, Anal. Chim. Acta, 1958, 19, 166. L. Erdey, E. Gyory and G. Svehla, Proc. Anal. Chem. Conf. Budapest, 1966,3,243. L. Erdey and G. Svehla, Z. Anal. Chem., 1957,154,406. 0. Menis and T. C. Rains, Anal. Chem., 1959,31,187. .I. Yofe, R. Avni and M. Stiller, Anal. Chim. Acta, 1963, 28, 331.

Talattta, 1968. Vol. 15, pp. 982 to 985.

Persanlott Press. Printed in Northern Ireland

Improvement of the sensitivity of molybdenum and tungsten determination in niobium and tantalum* (Received 6 February 1968. Accepted 13 March 1968) RECENT years, great strides have been made in producing refractory metals of extraordinary purity. In 1962, the dithiol method for determining molybdenum and tungsten in niobium was introduced*.’ to cover the 50-1000 ppm range. Now, only five years later, commercially produced niobium frequently contains only a small fraction of the amounts of molybdenum and tungsten for which the method was designed. In addition, application of the method to the determination of IN

* Work performed

in part under Air Force Contract F33615-67-C-1524.