Analyst, June, 1974, Vol. 99, pp. 355-359 355 Application of the Spectrophotometric Determination of Nickel and Cobalt in Mixtures With Bipyridylglyoxal Dithiosemicarbazone to the Analysis of Catalysts BY J. L. BAHAMONDE, D. PfiREZ BENDITO AND F. PIN0 (Department of Analytical Chemistry, University of Seville, Seville, Spain) Bipyridylglyoxal dithiosemicarbazone forms a complex with nickel(I1) a t pH 5.2, which can be extracted into chloroform (Amax. = 410 nm). A similar complex is obtained with cobalt(II), but is not extractable in this solvent, thus allowing nickel and cobalt to be determined in mixtures. Two procedures are proposed for the accurate analysis of such mixtures in which 1 p.p.m. of one of the ions can be determined accurately in the presence of as much as 5 p.p.m.of the other. One of the procedures has been applied to the determination of nickel and cobalt in industrial catalysts and the results obtained have been compared with those obtained by atomic-absorp- tion spectrophotometry. Satisfactory results were obtained. THE work described in this paper forms part of an investigation into the use of dithiosemi- carbazones as analytical reagents. In a previous paper1 we have studied the reactions between iron(I1) and (111) ions and bipyridylglyoxal dithiosemicarbazone (BGT)- I H2N-C-NH S II In this paper the determination of nickel(I1) and cobalt(II), both separately and in mixtures, with the above reagent and its application to the analysis of catalysts are described. EXPERIMENTAL APPARATUS- S9ectrophotometers-Unicam SP800 and SP600 spectrophotometers, equipped with 1-0-cm quartz or glass cells, were used for ultraviolet and visible-light absorbance measure- ments.A Perkin-Elmer 402 atomic-absorption spectrophotometer was also used. Digital PH meter-A Philips PW 9408 instrument, with glass - calomel electrodes, was used. SOLUTIONS- All solvents and reagents were of analytical-reagent grade. Bipyridylgglyoxal dithiosemicarbazone solution-A 0.1 per cent, m/V solution in ethanol. Standardised solutions of n i c k e l ( I I ) and cobalt(II). Acetic acid - sodium acetate bufer solutiorlz, pH 5.2. The reagent is synthesised from bipyridylglyoxal and thiosemicarbazide.2 PROCEDURE- Determination of nickel-Up to 6 p.p.m. of nickel, 15 ml of 0.1 per cent. m/V bipyridyl. glyoxal dithiosemicarbazone solution in ethanol, 20 ml of acetic acid - sodium acetate buffer @ SAC and the authors.356 BAHAMONDE et al.: SPECTROPHOTOMETRIC DETERMINATION OF NICKEL [Analyst, Vol. 99 solution (pH 5.2) and up to 50 ml of water are poured into a separating funnel. The mixture is shaken briskly, left to stand for 30 minutes and then extracted four times with 5-ml volumes of chloroform. The chloroform extracts are collected in a 25-ml calibrated flask and diluted to the mark with chloroform. The absorbance at a wavelength of 410 nm is measured against a blank obtained by extraction of the reagents (containing no nickel) in the same way. Determination of cobalt-A solution with a concentration of up to 8 p.p.m. of the cobalt is placed in a calibrated flask, 15 ml of 0.1 per cent.m/V bipyridylglyoxal dithiosemicarba- zone solution in ethanol and 20ml of pH 5.2 buffer solution are added and the mixture is diluted to 50ml with water. The flask is shaken vigorously, allowed to stand for 1 hour and the absorbance is measured at 410 nm against a reagent blank. Determination of nickel and cobalt in mixtures: Method A-A neutral solution of nickel(I1) and cobalt(I1) is poured into a separating funnel; the pH is maintained at 5.2 by use of the buffer solution. An excess of 0.1 per cent. m/V reagent solution is added. After 30 minutes the complexes are extracted into chloroform as indicated above. The aqueous and chloroform layers are separated and the absorbance of each is measured at 410 nm, as described. The amount of nickel is calculated from the absorbance of the organic layer, and cobalt from that of the aqueous layer. Determination of nickel and cobalt in mixtures : Method B T w o identical aqueous sample solutions are prepared and in one of them the sum of nickel plus cobalt is determined at pH 5.2.With the other sample nickel is determined alone after extraction into chloroform. A cali- bration graph is plotted for the nickel ion in the aqueous layer and the amount of cobalt is obtained by difference. Determination of nickel and cobalt in industrial catalysts-A 0.5-g amount of catalyst is weighed accurately and is placed in a flask that is suitable for refluxing to which 20 ml of concentrated nitric acid and 60 ml of concentrated hydrochloric acid are added. The mixture is refluxed for 1 hour and is then concentrated to about 25 ml.The solution is adjusted to pH 5 to 7 and diluted to 100 ml with distilled water in a calibrated flask. Aliquots are taken from this solution and the nickel and cobalt complexes are developed and the analysis is completed as described under Method A . RESULTS AND DISCUSSION REACTION OF BIPYRIDYLGLYOXAL DITHIOSEMICARBAZONE WITH NICKEL AND COBALT- Bipyridylglyoxal dithiosemicarbazone forms yellow - green chelates with nickel( 11) and cobalt(I1) in a weakly acidic medium, the wavelengths of maximum absorption being 390 and 410 nm, respectively (Fig. 1). Both complexes are formed slowly and the solutions must be allowed to stand for a time for the development of a stable colour. Reducing agents do not Wavelengthhm Fig.1. Absorption spectra of solutions a t pH 5-2 of complexes formed with bipyridylglyoxal dithiosemi- carbazone: 1, 5 p.p.m. of nickel(I1) in a homogeneous medium; l’, 5 p.p.m. of nickel extracted into chloroform; and 2, 7 p.p.m. of cobalt(I1) in a homogeneous mediumJune, 19741 AND COBALT IN MIXTURES FOR THE ANALYSIS OF CATALYSTS 357 affect the cobalt complex spectrum, but oxidising agents, such as potassium persulphate and hydrogen peroxide, shift the maximum towards the ultraviolet, with a notable hyperchromic effect. The nickel complex is extracted into chloroform at pH 5.2 and the A,. shifts to 410 nm. The cobalt complex is not extracted into this solvent at any pH value. The different behaviour of nickel and cobalt complexes in regard to extraction into chloroform is probably caused by the charge on the cobalt complex.With iron(II), the bi- pyridylglyoxal dithiosemicarbazone - iron( 11) complex is extracted into chloroform both in an acidic and in an ammoniacal medium, giving an emerald-green colour, but the extraction is easier in an ammoniacal medium. The extraction of other ions has not been tested, but their extraction is to be expected as some of them produce positive errors in the study of the inter- ferences of the nickel(I1) and iron(I1) complexes. Iron(I1) ions interfere in the determination of cobalt and nickel, as can be seen from the results of the study of the interferences of both ions. The absorbance - pH graphs for the cobalt and nickel complexes are shown in Fig. 2. Despite its not being the optimum value, a pH of 5.2 has been chosen for the nickel complex because the sensitivity of the reaction is good, the acetic acid - sodium acetate buffer is more readily available, and some interferences are avoided at this pH value.The absorbance of a chloroform solution of the nickel complex remains stable for at least 24 hours. The ethanol (from the reagent solution) extracted with chloroform stabilises the solutions. The stoicheiometry of the complexes has been studied by the continuous variation method (Fig. 3). The reagent - metallic ion ratio found is 1 :1 for the nickel complex and 2:l for the cobalt complex. f i n e " - A - - - - v v - B I I I I I I I I I 1 (u 1.2 1.0 + 0.8 a 0.6 0.4 0.2 0 0 Fig. 2. Absorbance veisus pH graphs of nickel and cobalt complexes of bipyridylglyoxal dithiosemicarbazone : A, nickel complex extracted into chloroform for 6 p.p.m.of nickel (Amax. 410 nm) ; and B, cobalt complex in homogeneous medium for 3.2 p.p.m. of cobalt (Amax. 410 nm). Both graphs have been obtained with use of various amounts of hydrochloric acid and sodium hydroxide ANALYTICAL APPLICATIONS OF THE COBALT AND NICKEL COMPLEXES- Nickel com$Zex-The optimum conditions for the formation and extraction of the nickel complex have been indicated under Experimental. Beer's law is obeyed between 1 and 5 p.p.m. of nickel and the molar absorptivity at 410 nm is 1.17 x lo4 1 mol-1 cm-1. Ringborn's graph shows that 1.4 to 4.0 p.p.m. of nickel(I1) is the minimum range of error. The relative error (P = 0.05) of the method is 5 0.13 per cent.The interferences on 4 p.p.m. of nickel have been investigated: 100 p.p.m. of cobalt(II), manganese(II), chromium(VI), silver, platinum( IV), tungsten(VI), lanthanum, calcium, barium and aluminium and 10 p.p.m. of mercury(I1) ions gave errors below 5 per cent.; 2 p.p.m. of iron(II), copper(II), zinc, cadmium and osmium(1V) ions gave errors between 4 and 8 per cent.; 200 p.p.m. of fluoride, oxalate, citrate, phosphate and thiosulphate did not358 BAHAMONDE et d. : SPECTROPHOTOMETRIC DETERMINATION OF NICKEL [Analyst, VOl. 99 [Me"+] t [BGTI Fig. 3. Stoicheiometry of nickel and cobalt complexes of bipyridylglyoxal dithiosemicarbazone (continuous variations) : A, nickel complex extracted into chloroform a t pH 5.2 (Amax. 410 nm) ; and B, cobalt complex in homogeneous medium at pH 5.2 (Amax.410 nm). The initial solutions of nickel and cobalt were a t a concentration of 1 x M interfere or gave errors below 2 per cent. ; and 5 p.p.m. of EDTA and cyanide gave an error of 10 per cent. Cobalt comjdex-The conditions established in the recommended procedure have been determined empirically. Beer's law is obeyed at pH 5-2 for between 1 and 7 p.p.m. of cobalt and the molar absorptivity is 9.05 x lo3 1 mol-l cm-I. The optimum concentration range, evaluated by Ringborn's method, is 1.5 to 7.5 p.p.m. of cobalt. The relative error (P = 0-05) of the method is & 0.55 per cent. The interferences for 4 p.p.m. of cobalt were investigated: 100 p.p.m. of manganese(II), cnromiumt v 11, piarinumti v j , rungsrent v i j , ianrnanum, calcium, parium, aluminium, lithium, sodium, potassium, magnesium, rubidium and strontium ions did not interfere or gave errors lower than 2 per cent.; 10 p.p.m.of mercury(I1) and cadmium ions, 2 p.p.m. of zinc and osmium(1V) ions and 0-5 p.p.m. of iron(II), copper(I1) and nickel ions gave errors of over 15 per cent.; EDTA and cyanide ions interfered above 5 p.p.m.; and 200 p.p.m. of citrate gave errors of 5 per cent. At this concentration (200 p.p.m.) fluoride, oxalate, per- chlorate, phosphate and thiosulphate do not interfere. TABLE I DETERMINATION OF NICKEL AND COBALT IN MIXTURES BY METHOD A Determination of nickel Determination of cobalt r Nickel added/pg ml-l 0.5 1.0 1.0 1.0 2.0 2.0 3.0 4.0 4.0 3.0 1.8 - Ni :Co ratio 0.05 0.1 0.17 0.2 0.33 0.5 1.0 2.0 4.0 6.0 9.0 -l Nickel found/pg ml-I 0.55 1.05 1.0 1.0 2-05 2-05 3.06 4-06 4.06 3.0 1-8 f Cobalt added/pg ml-l 5.0 5.0 6.0 5.0 6.0 4.0 3.0 2.0 1.0 0.5 0.2 - Co :Ni ratio 20.0 10.0 6.0 5.0 3.0 2.0 1.0 0-5 0.25 0.17 0.1 1 1 Cobalt found/pg ml-1 5.3 5.2 6.0 5.0 6.0 4.0 3.0 2.0 1.0 0.55 0.25June, 19741 AND COBALT IN MIXTURES IN THE ANALYSIS OF CATALYSTS 359 Analysis of nickel and cobalt mixtures-We have applied method A to a series of eleven samples in which the nickel to cobalt ratio varied from 0.05 to 9.The absorbance should not exceed 1. Table I shows the results obtained. It can be deduced from these results that nickel can be determined, within the concentration range given in Table I, in the presence of up to ten times its own concentration of cobalt with an error of less than 5 per cent.It is possible to determine cobalt in the presence of concentrations of nickel up to four times greater with errors below 4 per cent. Table I1 shows the results obtained by use of method B when applied to thirteen samples in which the nickel to cobalt ratio varied from 0.14 to 7. In the presence of up to five times greater concentrations of one ion, the other can be determined with minimal error. TABLE I1 DETERMINATION OF NICKEL AND COBALT IN MIXTURES BY METHOD B Nickel Nickel Cobalt Cobalt added/pg ml-l found/pg ml-f addedlpg ml-l found/pg ml-1 1.0 1.5 7.0 7.5 1.0 1.5 6.0 6.1 1.0 1.0 5.0 5-0 1.0 1.0 4.0 4.0 2.0 2.0 6.0 6 0 2.0 2.0 4.0 4.0 3.0 3.0 3.0 3.0 4.0 4.0 2.0 2.0 6.0 6.0 2.0 2.0 4.0 4.0 1.0 1.0 5.0 5.0 1.0 1.0 6.0 6-1 1.0 1.5 7.0 7.6 1.0 1.5 Amlysis of industrial catalysts-The techniques described above have been applied to the determination of trace amounts of nickel and cobalt on aluminium oxide supports and in process catalysts such as the Unifining catalyst, Isomax-UOP-DHC-2, Filtrol 475-8 and Unifining Petresa.All of these supports have the shape of a ball of about 2 mm diameter, or a small cylinder of similar dimensions. For Unifining catalyst method A was used. The results are compared with those obtained by atomic-absorption spectrophotometry, and are shown in Table 111. It should be noted that neither an excess of aluminium oxide nor the presence of molybdenum, which is present in these catalysts, interferes. TABLE I11 ANALYSIS OF INDUSTRIAL CATALYSTS Sample* Methodt Unifining catalyst . . .. . . AA Isomax-UOP-DHC-2 . . . . AA Filtrol 475-8. . .. .. . . AA Unifining Petresa . . . . . . AA BGT BGT BGT BGT Nickel, per cent. Cobalt, per ccn?-. 0.30 2.20 0.33 2.20 0.82 - 0.81 - - 1-16 - 1.16 2.31 - 2-36 - *Unifining catalyst: nickel 0-30 per cent., cobalt 2.20 per cent., molybdenum 3.40 per cent. Isomax-UOP-DHC-2 : nickel 0-82 per cent., molybdenum 4.60 per cent. Filtrol 475-8: cobalt 1.20 per cent., molybdenum 2.45 per cent. Unifining Petresa: nickel 2.35 per cent., molybdenum 4.50 per cent. (According to standard of UOP.) t AA = atomic absorption ; BGT = bipyridylglyoxal dithiosemicarbazone. REFERENCES 1. 2. Bahamonde, J. I,., Pdrez Bendito, D., and Pino, F., Talanta, 1973, 20, 694. -, -, -, Infcidn Quipn. Analit., 1972, 26, 7. Received Juty l l t h , 1973 Accepted December 28th, 1973