802 Analyst, November, 1973, Vol. 98, $y5. 802-810 Solvent Extraction of Copper(I1) and Zinc(I1) with 1,5=Diphenylcarbazone BY HISAHIKO EINAGA AND HA JIME ISHII” (National Institute joor Researches in Inorganic Materials, Niihari-gun, Ibaraki-hm, Japan) The extraction characteristics of 1,5-diphenylcarbazone and its com- plexes with the bivalent metal ions copper(I1) and zinc(I1) in an isobutyl methyl ketone - water system have been studied and the extraction curves of these metal complexes have also been obtained. The copper(I1) complex is extracted from a more acidic solution than is the zinc(I1) complex. The extraction equilibria have been examined and the extraction constants determined. The spectral properties of the complexes have also been deter- mined and the application of the reagent to the determination of copper and zinc is suggested. 1,5-DIPHENYLCARBAZONE has not been widely used as an analytical reagent for the separation and determination of trace amounts of metals in spite of the close similarity of its chemical structure to that of dithizone (1,5-diphenylthiocarbazone). It has been thought that 1,5-di- phenylcarbazone is “far inferior to dithizone for the purpose.”l This structural similarity between dithizone and 1 ,5-diphenylcarbazone, however, implies that the latter possesses considerable potentiality for the formation of complexes with certain of the metal ions that are known to form complexes with the former.Several studies have been made on the reaction between 1,5diphenylcarbazone and several metal ions, such as iron(II), iron(III), cobalt(II), copper(II), mercury(II), cadmium(II), lead(I1) and ~ i n c ( I I ) , ~ - ~ all of which, except iron(III), react with dithizone.It seemed, however, that relatively few details have been reported on the characteristics of solvent extraction of the 1,5--diphenylcarbazone complexes for analytical purposes, which led us to investigate them further. The present paper describes the extraction characteristics of copper( 11) and zinc(I1) with 1,5-diphenylcarbazone, together with those of the ligand itself, in the solvent system isobutyl methyl ketone - water. EXPERIMENTAL REAGENTS- Co@er(II) solution, 1.020 mg ml-1-This solution was prepared by dissolving recrystallised copper(I1) sulphate pentahydrate in water and acidifying the solution with a small amount of sulphuric acid so as to prevent precipitation of copper(I1) by hydrolysis.The solution was standardised by using a complexometric method with 1-(2-pyridylaz0)-2-naphthol as a metallochromic indicator.6 Zinc(1I) solution, 1.150 mg ml-l-This solution was prepared by dissolving pure zinc metal (99-99 per cent.) in hydrochloric acid, a slight excess of which was added in order to prevent precipitation of zinc(I1) by hydrolysis. This solution was also standardised by a complexo- metric method with Eriochrome black T as a metallochromic indicator.’ 1,5-DiphenylcarbazoneJ 0.020 per cent. (8.32 x lo-* M) solution in isobutyl methyl ketone- The 1,5-diphenylcarbazone used to prepare this solution was obtained from Kanto Chemical Co., Inc., Japan, and was purified before use by extraction with diethyl ether so as to remove any lJ5-diphenylcarbazide* and was finally recrystallised from ethanol.Adjustment of the pH and ionic strength of the aqueous phase was carried out by using 0.2 M sodium acetate - 0.2 M acetic acid or 0.05 M sodium borate - 0.1 M hydrochloric acid or sodium hydroxide solution for pH adjustment and 1 . 0 ~ ammonium chloride solution to adjust the ionic strength. * Present address : Department of Applied Chemistry, Faculty of Engineering, Tohoku University, Sendai, Japan. @ SAC and the authors.EINAGA AND ISHII 803 APPARATUS- Absorption spectra and absorbances at the specified wavelength were obtained with a recording spectrophotometer, Model EPS-3T (Hitachi Ltd., Japan); and a spectrophoto- meter, Model 139 (Hitachi Ltd., Japan), respectively, with matched 10-mm silica cuvettes. The pH of the equilibrated aqueous phase was measured with a glass electrode - saturated calomel electrode pair and pH meters, Model HM-5A (Toa Dempa Co. Ltd.) and Model F-5 (Horiba Ltd., Japan).A solution 0.09 M in ammonium chloride and 0.01 M in hydrochloric acid was defined as -log [Hf] = 2.00. Equilibration of both organic and aqueous phases was carried out with a universal shaker (Model KM, Iwaki Ltd., Japan) at the rate of 300 strokes per minute and at a temperature of 25 "C. PROCEDURE- Measurements of extractability (percentage extraction) and distribution ratios of the metal complexes were obtained by the following procedure. A mixture of n ml of the 1,5-diphenyl- carbazone solution and (20.0 - n) ml of isobutyl methyl ketone was equilibrated at various pH values with 20-0 ml of an aqueous solution that contained a definite amount of metal ions and was adjusted to an ionic strength of 0.10 M with ammonium chloride. After separation of the phases, the metal ions extracted into the organic phase were determined by measuring absorbances at a specified wavelength and by making use of the molar absorptivity of each metal ion (see under Extraction characteristics of the complexes of copper(I1) and zinc(I1) with 1,5-diphenylcarbazone). The separated aqueous phase was used for measurement of hydrogen-ion concentration.The concentration of metal ions in the aqueous phase was obtained as the difference between the initial concentration and that in the organic phase, and the distribution ratio, DA, the ratio of the concentration of the metal ions in the organic phase to that in the aqueous phase, was calculated from these results.The above procedure was also applied in the absence of metal ions to the determination of the extractability and distribution ratio of the ligand itself. In this instance the concen- tration of 1,5-diphenylcarbazone in the organic phase was determined by shaking the separated organic phase with an acetate buffer solution (pH 5) and then measuring absorbances at 560 nm (a calibration graph had been constructed by using purified 1,5-diphenylcarbazone). This shaking treatment was necessary in order to obtain reproducible and uniform absorption characteristics of 1 ,5-diphenylcarbazone, the concentration of which in the aqueous phase was obtained as described previously for the metal ions.RESULTS AND DISCUSSION EXTRACTION CHARACTERISTICS OF 1,5-DIPHENYLCARBAZONE- Consideration of the structure of 1,5-diphenylcarbazone suggests that its ketonic and enolic forms are present in tautomeric equilibrium in the solution. From its behaviour on neutralisation with sodium hydroxide it has been reported that 1,5-diphenylcarbazone is a monobasic acid.3 It is, however, reasonable to consider it to be a dibasic acid, just as the structurally closely similar dithizone is a dibasic acid (ha1 = about 2 x loB5 and ka2.< 10-15) .9 Studies were therefore first made on the spectral and extraction characteristics of 1,5-diphenylcarbazone in an isobutyl methyl ketone - water system, and these charac- teristics are shown in Figs.1 and 2, respectively. The absorption spectrum of this re- agent did not vary with change in pH of the equilibrated aqueous phase below 8, and had an absorption maximum at 460 nm, which showed, however, a gradual bathochromic shift when the pH was increased and reached a constant value above pH 11 (Amax. = 505 nm at pH 11-25). The distribution of the reagent from the organic to aqueous phase also became appreciable when the pH was above 8 and gradually increased with increasing pH. The reagent in the aqueous phase had an absorption maximum at 495nm, which was independent of the pH. These results can be qualitatively interpreted as follows : 1,5--diphenylcarbazone is present in the organic phase in ketonic (Amax.= 460 nm) and enolic forms (Amax, = 505 nm) in tautomeric equilibrium, which is, however, gradually shifted in favour of the enolic form by equilibration with an aqueous solution at pH above 8. The enolic form is then distributed into the aqueous phase in which it dissociates into a proton and the anion, HDN- (Amax, = 495 nm), where HzDN represents the undissociated 1,5-diphenyIcarbazone. A series of these equilibria can be represented as follows-804 where the subscript keto represents the ketonic form, enol the enolic form, org the organic phase and absence of subscript the aqueous phase. These equilibria can be exmessed as follows- 0.5 I / / ‘1 3’ / / / / / / I I I I 400 500 600 Wavelength/nm Fig. 1.Absorption characteristics of 1,5-diphenyl- carbazone in the isobutyl methyl ketone - water system. [1,5-Diphenylcarbazone] 4-16 x M ; and [NH4C1] 0.1 M. 1 to 4, absorption spectra of organic phase. pH value of equilibrated aqueous phase: 1, 5.25; 2, 8-80; 3, 10.20; and 4, 11.25. 3’, Absorption spectrum of aqueous phase; pH 10-20 Further, let us define PL as follows- PL = ( [ H a D N ] k e t O , O r g + [ H @ N ] e n o l , o r g ) / [ H 2 D N I e n o l = [H2DN]Org/ [H2DN]enOl - .. .. * (5) where [ H 2 D N ] o r g is the total concentration of 1,5-diphenylcarbazone in the organic phase. The terms Pa and PL can be correlated in the equation .. .. * (6) PL = (1 + Kke) P L ~ ..November, 19731 AND ZINC(I1) WITH 1,5-DIPHENYLCARBAZONE The distribution of 1,5-diphenylcarbazone can then be defined by 805 where the partition coefficient of the ketonic form was assumed to be so large that its effect in the aqueous phase can be neglected as compared with other species, which assumption is considered to be reasonable having regard to our results.By making use of equations (1) and (6), equation (7) can be rewritten in the following logarithmic form- Equation (8) implies that- Log DL = log J'L - log (1 + kal/[H+] + kalka2/[H+I2) . . - ' (8) (;) log D, should have no dependence on log [H+] if H2DN en01 is the principal species in the equilibrated aqueous phase; (G) log DL should have a linear relationship to log [Hf] with a slope of unity if HD,- is the principal species or with a slope of two if D,2- is the principal species; and (G) there should be a non-linear relationship between log DL and log [H+] with a tan- gential slope between 0 and 1 or 1 and 2, depending on whether appreciable concen- trations of both H2DN en01 and HDN- or HD,- and DN2-, respectively, are present in the aqueous phase.Fig. 2 shows the experimental results obtained for log DL and log [H-t]. A linear relationship exists between them with a slope of unity, indicating that HDN- is the principal species in the aqueous phase. Equation (8) can therefore be simplified as follows- Log D L = log P L - log + log [H+] . . .. . . (8a) The extraction constant of 1,5-diphenylcarbazone, Ken, which is expressed by the equilibrium &XL H~DN keto + eno1,org + H+ + HDN- can be defined as follows- .. * . (9) = PL/k,, .. .. .. The value of KexL was calculated by using the results in Fig.2 and was determined as log KeXL = 11.15. lbl - Log [H'] 100 - c; W W Q L $ 50- E .- c, 0 + UJ 0 - Fig. 2. M ; and [NH,Cl] 0.1 M. Extraction characteristics of 1,5-diphenylcarbazone. [ 1,5-Diphenylcarbazone] 4- 16 x (a), Extraction curve of 1,5-diphenylcarbazone; and ( b ) , dependence of distribution ratio on hydrogen-ion concentration GENERAL TREATMENT OF THE EXTRACTION EQUILIBRIA O F METAL(I1) - 1,5-DIPHENYLCARBAZONE COMPLEXES- The terms used below are defined as follows: Keq is the equilibrium constant of the extraction of the metal complex M(HNN)2(H2DN)n-.2, Kex the extraction constant and P, the806 EINAGA AND ISHII : SOLVENT EXTRACTION OF COPPER(II) [Analyst, Vol. 98 partition coefficient of the same metal complex, D, the distribution ratio of the metal, D M = [M(HDN)2(H2DN),-2]or,/[M2+], and D,’ the apparent distribution ratio of the metal, D,’ = [M(HD,)2(H2DN),-Jorg/ [M2+]total ( [M2+]total is the total concentration of the metal in the aqueous phase).I<%%, is the equilibrium constant in the aqueous phase of the metal complex M(DN)nH,c,2+n’-2ra, p series are formation constants of the side reaction of the metal ion with the species indicated, and acoeff the side-reaction coefficient of the metal ion. In the extraction equilibria of bivalent metal ions with 1,5-diphenylcarbazone, both ketonic and enolic forms can be considered to participate to an equal extent in the extraction reaction. It is, however, very difficult from equilibration studies, although unimportant in so far as it affects the extraction equilibria, to elucidate which of the forms is the principal participant in the reaction.Therefore, it was considered that both forms of 1,5-diphenyl- carbazone participate in the reaction involving the extraction of metal complexes. In addition to the assumption made above, it should be considered that the extracted metal complexes are electrically neutral. Under these conditions the principal extraction equilibrium can be represented as follows: Ke q M2+ + ~ H ~ D N org + M(HDN)~(H~DN),-~ org + 2H+ = D, [H+] 2 [H2DNIFn .. . . .. . . (10) The term D,’, which can easily be obtained experimentally, can be defined as follows- D,’ = n‘ n [M(HDN)2(H2DN)n-210Fg (W [h12+]+C C [M(D,),Hn,2+”’-2n ] + 5 [M (NH 3)m2+] + [ M (OAc) p2-”] + 5 [M (OH) a2- a ] 1 1 1 where the following equilibria were taken into consideration- K d M2+ + utHDN- + M(DN)nHn,2+n’--2n + (rt-%‘)H+ .... [ M ( DN) nHn,2+n’-2n] [H+] n--n’ [M2f] [HDN-]” P M Kn,/ = M(HDN)2(H2DN)7a-2 M(HDN)2(H2DN)?Z-2 Ore P, = [M (HDN)2(H2DN) n-2lorg . . . . .. [M(HDN)2(H2DN) 92-21 IN (NH,)m2+1 .. .. .. Bm’= [M2+] [NH,]“ .. .. .. .. .. .. P .. + X 1 /3i’[OAc-]” + 5 1 pql”[OH-]* . .November, 19731 AND ZINC(I1) WITH 1,5-DIPHENYLCARBAZOXE 807 Further, it should be appropriate to assume that the presence in the aqueous phase of lower and higher order complexes for the 1,5-diphenylcarbazone ligand than the electrically neutral complex M(HDN)2(H2DN)n-2 can be neglected and that the neutral complex has a partition coefficient that is high enough to be able to make l/PM < 1.With'these assumptions, equation (17) can be further simplified into logarithmic form- Equation (18) implies that a linear relationship should exist between log DM' + log xcoeff and log [H+] under constant conditions of log [HzDNlorg with a slope of 2 and log Dy' + log acoeff and log [HzDNIorg under constant conditions of log [H+] with a slope of n, which corresponds to the actual number of ligands in the extracted complex. Now let us define the term extraction constant, Kex, as follows: K e x M2+ + HDN- + M(HD,),(H2DN),2 org + (2 - n) H+ [M(HDN)2(H2DN)n-2]Org[H+]2-~ [M2+] [HD,-]" Kex = z= P, K , a(n-1) .. .. .. .. . . (19) Log Kex = log Ke, + n log Kex, .. .. . . (20) The following relationship can be obtained from equations (2), (3), (10) and (19)- The term Kex indicates how easily the complex can be formed in the aqueous phase and extracted into the organic phase, and can be obtained by using the value of Ke,, as determined in the preceding section.EXTRACTION CHARACTERISTICS OF THE COMPLEXES OF COPPER(I1) AND ZINC(I1) WITH 1,5-DI- PHENYLCARBAZONE- Preliminary experiments have shown that zinc(I1) cannot be extracted as its 1,ti-diphenyl- carbazone complex into the organic phase when ammonium salts are absent, although the extraction of the copper(I1) complex was not influenced by the presence or absence of ammonium salts. This difference in behaviour may be due to the formation of hydroxo species of zinc(II), especially the zincate ion, which retards or inhibits the formation of the zinc(I1) complex.It is therefore necessary to add ammonium salt to the aqueous phase i 0 I I I 2 3 4 5 6 7 8 9 10 11 PH Fig. 3. Extraction curves of copper(I1) and zinc(I1) as 1,5-diphenylcarbazone complexes. 1, Extraction curve of copper(I1) : [copper(II)] 8-03 x M ; [1,5-diphenylcarbazone] 2.08 x lo-* M; and [NH,C11 0.1 M. 2, Extraction curve of zinc(I1) : [zinc(II)] 1.76 x M- [i,Ei-diphenylcarbazone] 4-16 x M; and [NH,Cl] 0.1 M808 EINAGA AND ISHII: SOLVENT EXTRACTION OF COPPER(II) [Analyst, Vol. 98 and thus convert the zinc(I1) into the more labile ammine complexes. The presence of ammonium chloride at a concentration of 0.1 M was concluded to be sufficient for the extraction of zinc(I1) as its 1,5-diphenylcarbazone comdex. I 3 4 5 - Log [H'] Fig.4. Extraction characteristics of copper(I1) - 1,5-diphenylcarbazone complex. [Copper(II)] 8-03 x M ; and [NH,Cl] 0.1 M. (a), -Log [H+] 6.12; and ( b ) , -log [H2DN]org 3.70 Extraction curves for the copper(I1) and zinc(I1) complexes are presented in Fig. 3, which show that copper(I1) can be extracted into the organic phase from more acidic solution (pH,,, = 4-30) than is zinc(I1) (pH,,, = 7.02). Under the conditions specified in Fig. 3, copper(I1) was extracted quantitatively (more than 99 per cent.) and zinc(I1) almost quanti- tatively (98 per cent.) by a single extraction. It is therefore necessary to carry out a second extraction for the quantitative extraction of zinc(I1). 4 c $ 3 m - + H Q 2 0) -J 1 6 7 8 - Log [H'] Fig. 5. Extraction characteristics of zinc(I1) - 1,5-diphenylcarbazone com- plex.[Zinc(II)] 1-76 x M; and [NH,Cl] 0.1 M. (a), -Log [H+] 8.90; and (b), -log [H2DN]Org 3-40 The extraction characteristics of copper(I1) and zinc(I1) complexes are presented in Figs. 4 and 5 , respectively; in each instance acoeff was calculated by using the formation constants summarised in Table I with suitable modification for some of the results due to the change in ionic strength. It is evident from Figs. 4 and 5 that linear relationships exist for both copper(I1) and zinc(I1) between log DMf + log acoeff and log [H+], the slope of which is 2, as expected from equation (18). Linear relationships also occur for both metal ions between log D,' + log acoeff and log [H2DNIorg with a slope of 2 (n = 2). Therefore, the extraction equilibrium [equation (lo)] and extraction constant as defined in equation (19) can be simplified as follows- M2+ + 2H2DN org + M(HDN)z org + 2H+ KeqNovember, 19731 and AND ZINC(I1) WITH 1,5-DIPHENYLCARBAZONE 809 .... . . (19a) .. .. . . (20a) Values of Keq for both copper(I1) and zinc(I1) were calculated from the results presented in Figs. 4 and 5, respectively, and they are summarised in Table 11. Values of K e , were also calculated from the values of Keq and KexL, and they are also summarised in Table 11. TABLE I FORMATION CONSTANTS USED FOR THE CALCULATION OF Olcoeff" Ligand r L i Cation Term OAc- NH, OH- H+ pka 4.65 9-37 13.80 1-3 2.27 4.4 4.61 - 2.1 - 7.01 14.4 - 9-06 15.5 Zinc(I1) Log B1 Log B 2 Log Bs Log 8 4 Results were for p = 0.1 except with the hydroxo compIexes of copper(I1) and zinc(II), which were for p = 0 and were converted into p = 0.1 in the usual manner when calculations of ctcoeff were carried out.For acid dissociation of ligands, the acid- dissociation constant, pka, was cited instead of the formation constant. Copper(I1) and zinc(I1) complexes with 1,5-diphenylcarbazone [Cu(HD,), and Zn(HD,),, respectively] thus extracted into isobutyl methyl ketone have their absorption maxima at 530 and 520 nm, respectively, and the very large values of molar absorptivities of 7.6 x lo4 (530 nm) for the copper(I1) complex and 5.8 x 104 (520 nm) for the zinc(I1) complex suggest that the extraction into isobutyl methyl ketone of trace amounts of copper and zinc as their 1,5-diphenylcarbazone complexes can be utilised for their spectrophotometric determination.TABLE I1 EXTRACTION CHARACTERISTICS OF COPPER(II) AND ZINC(II) Composition Absorption maximum, Molar absorptivity, Metal of complex h,,X. /nm Emax. Log Keq Log Kex - 1*11* -7.10* -7.13t } 15*' -1.lOt } 21*2 cu CU ( HDN) 530 7-6 x 104 Zn Zn(HDN), 520 5.8 x 104 Log KexL = 11.15. * Results obtained from the relationship between log DM' + log acoeii and t Results obtained from the relationship between log DM' + log a c o e f f and log [€I+]. log [H,DN]org.810 EINAGA AND ISHII Balt and van Dalen5 reported that copper(I1) and zinc(I1) complexes with 1,5-diphenyl- carbazone have the following characteristics: ,a = 0.1 (NaC10,) at 20 to 22 “C; log Keq = 1.27 for copper(I1) and -6.76 for zinc(I1) ; Amax., 530 nm for copper(I1-) and zinc(I1); and EAmax., 6.8 x lo4 for copper(I1) and 3.7 x lo4 for zinc(II), which were determined by using a toluene - water system. By using these results for log Keq, and partition coefficient (PL = 39) and acid-dissociation constant (Kal = 2.9 x 10-9),5 the values of log K,, were calculated according to equations (9) and (ZOa) to be 21.5 for copper(I1) and 13.5 for zinc(I1).Fairly good agreement can be seen between the results found by Balt and van Dalen5 and those obtained in the present work for log Kex, although the partition systems used were different in the two instances. Some differences in these results, especially in the value for log Kex for zinc(II), and also in other characteristics (for example, molar absorptivity) might be partly due to the differences in the nature of the extraction systems toluene - water and isobutyl methyl ketone - water, which should, however, be elucidated in future from the standpoint of solution theory. The copper( 11) and zinc(I1) complexes showed no photochemical change under ordinary laboratory lighting conditions and are considered to be stable unless exposed to intense ultraviolet radiation. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. REFERENCES Sandell, E. B.. “Colorimetric Determination of Traces of Metals,” Third Edition, Wiley-Inter- Balt, S., and van Dalen, E., Analytica Chim. A&, 1961, 25, 507. science, New York, 1959, p. 178. f , Ibid., 1962, 27, 188. I Ibid., 1963, 29, 466. , Ibid., 1964, 30, 434. -- -- * -- Ueno, K., “Complexometric Titrations (Kireito Tekiteiho),” Nankodo, Tokyo, 1960, p. 260. Krumholz, P., and Krumholz, E., Mh. Chsm., 1937, 70, 431. Sandell, E. B., op. cit., p. 144. Ringbom, A., “ Complexation in Analytical Chemistry, ” Wiley-Interscience, New York, 1963, Received April 27th, 1973 Accepted Jwne 8th, 1973 -, 09. Cit.. p. 233. p. 293.