Analyst August 1983 Vol. 108 pp. 933-938 933 Pyridine-2-aldehyde 2-Furoyl hydrazone as a Fluorogenic Reagent for the Determination of Nanogram Amounts of Gallium Eloisa Requena Jose J. Laserna Aurora Navas and Francisco Garcia Sanchez Department of Analytical Chemistry Faculty of Sciences University of Mdlaga Mdlaga-4 Spain The synthesis characteristics and analytical applications of pyridine-2-aldehyde 2-furoylhydrazone (PAFH) are described. This compound has been examined to evaluate its usefulness as a selective and sensitive spectro-fluorimetric reagent for gallium. The method is applied in 0.8% ethanolic solution a t pH 4.5. Under these conditions the fluorescent species have excitation and emission maxima at 380 and 445 nm respectively. The detection limit is 0.8 ng ml-1 and the range of application is between 1 and 50 ng ml-l.The method has been employed to determine gallium in syn-thetic mixtures and its recovery from human urine samples. Keywords; Gallium determination; pyridine-2-aldehyde 2-furoylhydrazone reagent ; spectro$uorirnetry Organic reagents containing the atomic arrangement -CO-NH-N=CH- namely aroylhydra-zones have been widely used for the spectrophotometric determination of metal ions because of their great complexing ~apacity,l-~ forming coloured complexes with transition metal ions. However a search of the relevant literature revealed that these compounds had not been widely applied to spectrofluorimetric analysis. The ability of these reagents to undergo conformational transformations in the reaction between their ionised particles and metallic ions leads to the formation of complexes with coplanar structures This allows the formation of rigid structures that facilitate the fluores-cent emission from the complex.Taniguchi et aL6 synthesised 32 aroylhydrazones and studied the relationship between fluorescence and structure of the chelates formed with gallium aluminium scandium and zirconium. By means of these studies they concluded that the =CONH- group has a relevant participation in the formation of fluorogenic chelates. On the other hand the possibility of keto - enolic tautomerism which increases the conjugation of the molecule in the enolic form, and also the ability to co-ordinate with metal ions contribute greatly to the production of fluorescent phenomena with the co-ordination process.This paper also gives details of the synthesis properties and analytical behaviour of pyridine-2-aldehyde 2-furoylhydrazone (PAFH) together with a description of the more interesting chromogenic and fluorogenic reactions of this compound with metal ions. The detailed development and testing of a fluorimetric method for the determination of gallium is described below. The method has been applied to the determination of gallium in synthetic mixtures and in human urine samples. The antiturnour activity of gallium,' and its use as a tumour-scanning agent,8 as well as gallium toxicity,gJo have created the need for sensitive and reliable methods of determining gallium in tissues and biological fluids.ll Experimental Reagents Unless otherwise stated the reagents were of analytical-reagent grade.PAFH was synthesised by refluxing equimolar amounts of pyridine-2-aldehyde and 2-furoyl hydrazide (Ega Chemie) at 60-70 "C for 2 h. After cooling to room temperature the reaction mixture was placed in a refrigerator and after a short time brown crystals formed. These were filtered washed with cold water and ethanol repeatedly and redissolved in ethanol -water (1 + 1). The reagent was characterised by its infrared spectrum purity being con 934 REQUENA et al. PYRIDINE-%ALDEHYDE 2-FUROYLHYDRAZONE Analyst Vol. 108 firmed by elemental analysis (required for C,,H902N3 C 65.67 N 13.93 H 4.48%. Found: C 65.60 N 14.00 H 4.40%). The yield of the synthesis was 75%. The melting-point of the reagent was 108 & 1 "C.Solutions of the reagent were prepared weekly in absolute ethanol (Merck) . A stock solution of gallium was prepared by dissolving 0.9978 g of Ga(N03),.8H,0 in 250 ml of 2 M hydrochloric acid solution. The exact gallium content was determined by titration with EDTA. Solutions of lower concentrations were made by dilution with de-ionised water. Buffer solutions of pH 4.5 consisted of sodium acetate - acetic acid (0.1 M). Apparatus Fluorescence measurements were performed on a Perkin-Elmer Model MPF-43A spectro-fluorimeter equipped with an Osram XBO 150-W xenon lamp excitation and emission grating monochromators 1 x 1 cm quartz cells R-777 photomultiplier (Hamamatsu) and a Perkin-Elmer 023 recorder. A set of fluorescence polymer samples were used daily to adjust the spectrofluorimeter and compensate for changes in source intensity.The fluorescence data are given without spectral correction. An ultrathermostatic water-bath circulator (Frigiterm S-382) was used for temperature control. The absorption measurements were performed on a Shimadzu UV-240 Graphicord spectro-photometer . Procedure Into a 25-ml calibrated flask transfer 2.5 ml of pH 4.5 buffer solution and 0.2 ml of the appropriate PAFH ethanolic solution to make the final solution between 2 x and 2 x 10" M. Add an aliquot of sample solution containing 0.025-1.25 pg of gallium and dilute to volume with de-ionised water. Store in the dark for 30min; afterwards measure the fluorescence intensity a t 445 nm with excitation at 380 nm against a reagent blank at 15 "C.Preparation of Urine Sample Solutions A 10-ml volume of human urine was digested by treatment with concentrated nitric acid, evaporated to a small volume and diluted to 50 ml with de-ionised water after an aliquot of gallium standard solution had been added. Results and Discussion Acid - Base Equilibria of PAFH The absorption spectra of 0.8% ethanolic solutions of PAFH registered in the 250-400 nm spectral range show that for values of pH ranging from 1.5 to 12.1 the absorption spectra are unaffected. They show a maximum at 310 nm and two isosbestic points at 330 and 280 nm. From these experimental data the Pease and Williams12 method allows the determination of pK = 3.10 &- 0.08 and pK = 9.60 & 0.13. In agreement with literature data13-15 for similar reagents these values can be assigned as pK for the de-protonation of pyridinium hydrogen and pK for the de-protonation of enolic hydrogen.The species distribution diagram which corresponds to the prototropic behaviour of PAFH is shown in Fig. 1. Fig. 1. Species distribution diagram of PAFH. cc = Molar ratio August 1983 AS A FLUOROGENIC REAGENT FOR GALLIUM 935 Main Chromogenic and Fluorogenic Reactions of PAFH with Metal Ions The reactions of PAFH with 30 cations at three pH values (4.5 7 and 10) were investigated in a 0.8% ethanolic solution. The characteristics of the most important complexes are summarised in Table I. The data were obtained from the appropriate spectra which were measured in the presence of a 5 M excess of the reagent at those pH values which facilitate the formation of the different complexes.As shown the chromogenic reactions of PAFH are not very sensitive but on the other hand the reagent is selective. TABLE I ANALYTICAL CHARACTERISTICS OF THE MAIN COLOUR REACTIONS OF PAFH Concentration/ Ion pg ml-l pH hmax./nm AX*/nm Emax. x 10-4/1 mol-l cm-I Ni . . 1 10 360 50 3.80 Zn . . 5 10 350 40 0.84 c o . . 1 10 350 40 3.33 Cu(I1) . . 2 10 355 45 2.05 CU(I1) . . 2 4.5 365 55 2.16 Pd . . 5 4.5 405 95 1.59 * Colour contrast (AX) = Am, (complex) - Amax (reagent) under identical conditions of measurement. The fluorescent reactions were tested in a similar way and for those complexes that fluor-esced most strongly a fluorescent titration was performed. A summary of the main spectral characteristics of the fluorogenic reactions of PAFH with metallic ions is given in Table 11.The concentration of all the metallic ions in these experi-ments was 1 pg ml-l with a reagent concentration 5 M in excess. It is noteworthy that the relative fluorescence intensity of the PAFH chelates with gallium and scandium is higher than with aluminium. This is unusual because the heavy-atom effect tends to make the reactions of the lighter metal ions more sensitive. TABLE I1 ANALYTICAL CHARACTERISTICS OF THE MAIN FLUOROGENIC REACTIONS OF PAFH Ion pH Xexc./nm hem./nm RFI* A1 . . . . 4.9 382 442 350 Ga . . . . 4.5 380 446 4 100 In . . . . 6.9 393 452 15 s c . . . . 4.5 382 460 3 400 Zn . . 7.8 395 454 20 Y . . 7.5 384 430 80 * The relative fluorescence intensity of 0.1 p.p.m.quinine sulphate is 22. Spectrofluorimetric Study of the Gallium - PAFH System an intense blue fluorescence is formed immediately. the gallium - PAFH complex in a 0.8% ethanolic solution at pH 4.5 are shown in Fig. 2. reagent excitation and emission spectra are also given. When dilute gallium(II1) solutions and a 1 x M solution of PAFH in ethanol are mixed, The excitation and emission spectra of The Effect of Experimental Variables In order to evaluate the effect of pH a fluorescent titration was carried out in 0.8% ethanolic solution. The pH was changed by addition of small amounts of dilute hydrochloric acid and sodium hydroxide solutions. The results are shown in Fig. 3 which shows that oscillations in fluorescence intensity about the peak at pH 4.5 are within 3% in the pH range 4.44.55; the intensity diminishes strongly at both lower and higher pH values 936 REQUENA et al.PYRIDINE-%ALDEHYDE 2-FWROYLHYDRAZONE Analyst VOZ. 108 Wavelengthlnm Fig. 2. Uncorrected excitation (A,A') and emission (B,B') of complex (A,B) and free ligand (A',B') a t pH 4.5. [Gal = 4 x M and [PAFH] = 8 X 1 0 - ' M . 3 4 5 6 PH Fig. 3. Effect of pH on the formation of the gallium -PAFH chelate. [Gal = 4 x M and [PAFH] = 8 x M. Other experiments were carried out at pH 4.5 to determine the optimum percentage of ethanol. The results show that the ethanol concentration in the reaction solution must be carefully controlled because the fluorescence intensity decreases continuously by 70% as the ethanol concentration increases from 0.8 to 40%.Between ethanol concentrations of 40 and goy, the fluorescence intensity decreases by 15%; there is little or no fluorescence in 90% ethanolic solutions. The 0.8% ethanolic solutions are satisfactory. The fluorescence behaviour of the gallium chelate with solvent composition suggests that an increase in the polarity (0.8% ethanol) raises the energy of the n-n* state above the T-T* state and thereby facilitates deactivation by fluorescence emission.16 Generally if a carbonyl-containing molecule has an Sn,x. lowest state it will not fluoresce but will phosphoresce. The phosphorescence can originate from either a Tx,n* or a Tnmx. state. If an Sx,x* state is lowest both fluorescence and phosphorescence occur.17 Solvents can affect the emission properties in terms of both A,,,.and quantum efficiency. This results from a change in the nature of the energy relationship between the lowest Sn,n* and SxIx* statesla During the experiments a chelate photolytic phenomenon was observed that seriously affects the time necessary for stable fluorescence intensity. In order to evaluate the effect of instrumental parameters on the stability two solutions were continuously exposed to lamp-light at 380 nm with different excitation slits; another was stored in the dark until its fluores-cence intensity was measured. From the results obtained it was evident that to ensure better reproducibility small excitation slits and storage of the solutions in darkness until measurement were important Under the recommended experimental conditions the fluorescence emission remains stable for 90 min 30 min after preparation.The effect of PAFH concentration on the intensity from the 4 x 10-6 M gallium solution was also studied under conditions similar to those of the recommended method. The intensity of the fluorescence increases with concentrations of PAFH up to 8 x 1 0 - 4 ~ . For greater concentrations the intensity decreases steadily because of fluorescence inversion phenomena. The wide anomalous range of PAFH concentrations in which the fluorescence intensity increases can be ascribed to the low stability of the chelate which necessitates a large reagent excess for full development of the complex. The 2 x 10-4 M PAFH concentration was selected for further investigation as this impedes the fluorescence inversion phenomena and ensures a sufficient reagent excess.The dependence of the fluorescence intensity on temperature is critical showing a dramatic fall (70%) on increasing the temperature from 18 to 30 "C. A temperature increase from 5 to 18 "C diminishes the fluorescence intensity by 5%. This effect can be explained by the higher internal conversion as temperature increases facilitating non-radiative deactivation of the excited singlet state. In addition the contribution of the higher inter-system crossing rate as temperature increases may be effective. The presence in the PAFH molecule of a carbonyl group lends some support to this hypothesis because compounds of this naturelg often show these energy-transfer mechanisms. This work was carried out a t 15 & 0.5 "C Azlgust 1983 AS A FLUOROGENIC REAGENT FOR GALLIUM 937 Composition of the Complex The metal to ligand ratio in the complex was studied under the established working condi-tions by the method of continuous variation.This method was applied to a series of solutions in which the total concentration of reactants (gallium + PAFH) was kept constant at 4 x The maximum occurred at a 0.32 molar fraction of gallium indicating that the composition of the complex is 1 2 (metal to ligand) so, according to the structure of the ligand and the co-ordination number of the gallium ion a charged complex is formed (I) in which the ligand acts as a tridentate planar chelating agent. M but the molar fractions were varied. I I Analytical Parameters Linear calibration graphs passing through the origin were obtained for two ranges of gal-lium(II1) concentrations covering a total range of 1-50 ngml-l with two concentrations of PAFH.Two series of 11 measurements on 36 and 10 ng ml-l of gallium(II1) gave relative errors of 1.96 and 2.6% and relative standard deviations of 2.95 and 3.91y0 respectively. The limit of detection defined as the concentration of gallium giving a signal to noise ratio of 2 1 was 0.8 ng ml-l of gallium. Effect of Foreign Ions The effect of various ions on the determination of gallium at the 30 ng ml-l level was investi-gated by first testing a 100-fold m/m ratio of interferent to gallium and if interference occurred, reducing the ratio progressively until interference ceased.The criterion for interference was a variation of fluorescence intensity of more than &4% from the value expected for gallium alone. The results are given in Table 111. Interferences in this method arise from two main chemical sources and also from an inter-action at the excited state level. The latter occurs with Mo(VI) V(V) Ni(I1) and Fe(III), because of their paramagnetic character. The chemical sources are In(II1) and Sc(III) which form fluorescent chelates with PAFH; EDTA and citrate give moderately stable chelates with gallium ions at pH 4.5. Higher ratios were not tested. TABLE I11 EFFECT OF FOREIGN IONS ON THE DETERMINATION OF 30 ng ml-l OF GALLIUM (ERROR 4%) Tolerance/ ng ml-' Foreign ion or species Tl(I) Ag Be Mg Sr Ba Pb Ca Se(VI) As(V) SCN- S20a2-Hg(II) La Th Zn Mn Tl(III) Sb(III) Y Cd .. 1600 . . 3000 Cu(II) Pd Al Ti Os Co Bi POq3- F- . . 300 Ni . . . . 150 Mo 60 v(v) iie(III);+In citrate . . . . 30 EDTA Sc . . . . 16 Applications The recommended procedure for the determination of gallium was applied in a variety of situations to evaluate its effectiveness. For this purpose synthetic mixtures of common metal ions that usually accompany gallium in natural and manufactured samples were prepared an 938 REQUENA LASERNA NAVAS AND GARCIIA SANCHEZ TABLE IV DETERMINATION OF GALLIUM IN THE PRESENCE OF SYNTHETIC MIXTURES OF FOREIGN IONS Mixture of foreign ions/ng ml-l 150 A1 + 750 Zn . . 150 Cu(I1) + 750 Zn . . 150 Cu(I1) + 1500 Cd . . 750 Zn + 1500 Cd + 15 In .. 150 A1 + 15 In + 760 Tl(II1) . . 150 Cu(I1) + 1500 Cd + 750 Zn 750 Zn + 1500 Cd + 750 Hg . . 150 A1 + 15 In + 750 Hg . . * . . . Ga added/ ng ml-l 30 30 30 30 30 30 30 30 Ga found*/ ng ml-1 29.45 28.00 27.50 32.66 35.00 28.00 29.84 35.00 * Results are the means of three determinations. analysed (Table IV). Also a series of recovery experiments were carried out by adding standard pure gallium solutions to aliquots of human urine samples treated as indicated in the experimental section (Table V). The results obtained indicate that the method would be effective for the analysis of samples of similar complexity. TABLE V DETERMINATION OF GALLIUM IN HUMAN URINE Gallium added/ng ml-l Gallium found/ng ml-l Recovery % 0 0.19* -1 .oo 1.22 103.0 2.00 2.03 92.0 4.00 4.01 95.5 6.00 5.77 93.0 8.00 8.25 100.8 * Triplicate determination.1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. References Katiyar S. S. and Tandon S. N. Talanta 1964 11 892. Vasilikiotis G. S. and Tossidis J. A. Microchem. J . 1969 14 380. Uno T. and Taniguchi H. Bunseki Kagaku 1971 20 997. Gallego M. Garcia Vargas M. and Valchrcel M. Analyst 1979 104 613. Silva M. and Valckrcel M. Analyst 1980 105 193. Taniguchi H. Tsuge K. and Nagano S. Yakugaku Zasshi 1974 94 759. Hart M. M. and Adamson R. H. Proc. Natl. Acad. Sci. USAI 1971 68 1623. Hoffer P. B. Bekerman C. and Henkin R. E. Editors “Gallium-67 Imaging,” John Wiley, Newman R. A. Brody A. R. and Krafoff I. H. Cancer 1979 44 1728. Kelsen D. P. Alcock N. Yeh S. Brown J. and Young C. Cancer 1980 46 2009. Nakamura K. Fujimori M. Tsuchiya H. and Orii H. Anal. Chim. Acta 1982 138 129. Pease B. F. and Williams M. D. Anal. Chem. 1959 31 1044. Dolgorev A. V. Sivak N. S. Pal’nikova T. I. and Gurevich L. M. Zh. Anal. Khim. 1979 34 106. Luque de Castro M. D. and Valckrcel M. Talanta 1980 27 645. Cejas M. A, G6mez Hens A. and Valckrcel M. Anal. Chim. Acta 1981 130 73. Lippert E. Angew. Chem. 1961 73 695. Becker R. S. “Theory and Interpretation of Fluorescence and Phosphorescence,” John Wiley New Eastman J. W. Spectrochim. Acta Part A 1970 26 1545. Schulman S. G. “Fluorescence and Phosphorescence Spectroscopy Physicochemical Principles and New York 1978. York 1969. Practice,” Pergamon Press Oxford 1977. Received January 6th 1983 Accepted March 17th 198