January, 19661 STOCK 27 The Amperometric Titration of Submillinormal Concentrations of Copperh) with Mercuryh) Perchlorate BY JOHN T. STOCK (Department of Chemistry, The University of Connecticut, Stows, Connecticut, 7J.S.A .) Although quite precise under rigidly controlled conditions, the mercury(1) amperometric titration, a t a rotating platinum electrode, of submillinormal concentrations of copper (11) in potassium thiocyanate - perchloric acid medium gives results that vary with the concentrations of thiocyanate and of copper(r1). When this variation is eliminated by the addition of potassium iodide, the titration of 1 0 - 5 ~ to ~ O - * N copper(I1) is precise and accurate to within & 1.5 per cent. The titration of N copper(r1) is precise to about 5 per cent. CENTIGRAM amounts of copper(1r) have been determined by titration in acidified thiocyanate medium with mercury(1) nitrate to a vi~ual~,~1~1* or a potentiometricl 9 2 9 4 end-point.Iron(II1) at a concentration as low as N can be determined by mercury(1) amperometric titration at a rotating platinum ele~trode.~ The present work concerns the titration by this ampero- metric method of submillinormal concentrations of copper(I1). VOLTAMMETRY- A deoxygenated 0.3 N potassium thiocyanatc - 0.02 N perchloric acid medium that had been freshly made by mixing solutions of these reagents gave a residual current of less than 0.1 pA at a new rotating electrode held at a potential of zero (unless otherwise stated, all potentials are with respect to the saturated calomel electrode, S.C.E.). The introduction of potassium iodide up to a concentration of approximately 0.06 N had no significant effect on this current.Both in the presence and absence of potassium iodide, the limiting current of copper(II), measured at zero potential, was found to be proportional to the concentration of this ion. The small anodic current given by mercury(1) in acid thiocyanate medium5 was also obtained in the presence of potassium iodide. Fig. 1 shows current - voltage curves obtained at various stages in the amperometric titration of 5-8 x N copper(I1) in acid thiocyanate - iodide medium with mercury(1) perchlorate solution. The anodic current due to an excess of titrant often increased markedly with time. However, when observations were completed within a few minutes, the anodic current was usually almost unaffected by increasing the mercury@) concentration. AMPEROMETRIC TITRATIONS- Initially, the platinum electrode was stored overnight in concentrated nitric acid and then pre-conditioned by immersing in de- oxygenated 0.1 PI; perchloric acid and short-circuiting to a saturated calomel electrode." After some weeks, a low residual current became difficult to attain.When the daily treatment of a new electrode with nitric acid was reduced to a brief immersion, the residual current remained low and was essentially unchanged after 4 months. to 1 0 - 4 ~ copper(I1) in acid thiocyanate medium by the pre-addition technique, similar to that used in the titration of iron(111),~ gave results that were precise to within about 3 per cent.However, the stoicheiometry of the titration reaction appeared to change when the concentration of either copper(I1) or thiocyanate was altered. Such a change was not observed when potassium iodide, up to a concentration of about 0.05 N, was introduced into the titration medium. METHOD REAGENTS- EXPERIMENTAL All titrations were carried out at zero potential. Preliminary titrations of Use analytical-grade reagents and distilled or demineralised water throughout. Mercury(1) perchlorate, approximately 0.01 N in N perchloric acid-Prepare this as des- Store over cribed by Berka, Vulterin and Z3ka,4 but dilute 10-fold with N perchloric acid.28 STOCK: AMPEROMETRIC TITRATION OF SUBMILLINORMAL [Analyst, Vol. 91 metallic mercury. Standardise by titration with copper(I1) sulphate solution as described in the Procedure.Prepare this from the metal by dissolution in the minimum amount of nitric acid, elimination of nitrogen oxides, and suitable dilution with approximately 0.1 N sulphuric acid. Copper(I1) sulphate, 0.01000 N (=0-01000 M) in 0.1 N sulphuric acid. Perchloric acid, approximately 0.6 N (Procedure A ) or N (Procedure B). Potassium thiocyanate, approximately 0.6 N (Procedure A) or saturated (Procedure B) . Potassium iodide, approximately N in boiled-out water. APPARATUS- Except for an agar - potassium nitrate salt bridge, use conventional apparatus for amperometric titration at a rotating platinum electr~de.~?~ At the end of each day, rinse the platinum electrode with water, rotate it in concentrated nitric acid for approximately 1 minute, rinse it again, then remove most of the water with a filter-paper strip and leave to dry.At the beginning of the day, pre-condition the electrode by running a preliminary titration of copper(I1) until the titrant is present in slight excess. I t then had a sensitivity of 146 pA per millimole of silver(1) per litre, measured at zero potential in deoxy- genated 0.1 N perchloric acid at 23" C. Titrations were carried out at room temperature (in the range 23" to 25" C). Electrical measurements were made with a Cambridge Voltamo- scope. Since all titrations are carried out at zero potential, this instrument can be replaced by a Cambridge Spot galvanometer or similar instrument. In the present work, the platinum electrode was rotated at 500 r.p.m.6 5 4 ; 3 c E 3 2 I C - +O +0.3 +0.2 + O . I 0 - 0.1 -0.2 -0.3 Po te t i t ial, volts Fig. 1 . Current - voltage curves a t stages in the titration of 58 p~ copper(I1) in 0.3 N potassium thiocy- anate - 0.02 N perchloric acid - 0.02 N potassium iodide. Percentage equivalent of mercury(1) pmchlorate added : curve A, 0; curve B, 50; curve C, approximately 100; curve D, 150 PROCEDURE- (A) Transfer 25 ml each of 0.04 N perchloric acid and 0-6 N potassium thiocyanate to the titration cell, then add 1 ml of N potassium iodide. Insert the platinum electrode and salt bridge, set the applied e.m.f. at zero, deoxygenate with a stream of nitrogen, then stop the gas stream. Inject 0.01000 N copper(I1) sulphate so that the amount of copper(I1) intro- duced is about 20 per cent.of that contained in the sample solution. After 2 minutes, note the current reading, P, then inject the sample solution. Read the current after aJanuary, 19661 CONCENTRATIONS OF COPPER(II) WITH MERCURY (I) PERCHLORATE 29 further 2 minutes, then titrate with 0.01 N mercury(1) perchlorate until the current has fallen nearly to zero. Allow an interval of 1 minute between a titrant addition and the reading of the current. Find the end-point graphically as the intersection of the linear portion of the titration curve and the line: current = P. (B) Transfer the sample solution to the titration cell and make it approximately 0.3 N in potassium thiocyanate and 0.02 N in perchloric acid by adding saturated and N reagent solutions respectively. Insert the platinum electrode and salt bridge, but do not close the circuit until the solution has been deoxygenated.Then stop the gas stream and at once make the solution 0.02 N in potassium iodide by adding to it a N solution of this reagent. Read the current after 2 minutes, then titrate as in (A). Determine the residual current, R, in a deoxygenated 0-3 N potassium thiocyanate - 0-02 N perchloric acid - 0.02 N potassium iodide medium freshly made from the concentrated solutions. Find the end-point graphically as the intersection of the linear portion of the titration curve and the line : current = R. RESULTS N copper(I1) ion in a given iodide-free potassium thiocyanate - perchloric acid medium showed that Procedure A gave results that were precise to within &2 per cent.Results obtained by a conventional method involving extrapolation of the arms of an L-shaped titration curve8 were less precise than, and signifi- cantly different from, those given by Procedure A. Since current readings in the post- equivalence region were time-dependent, all observations within this region were completed within a total time of about 4 minutes. Under conditions of small and essentially constant residual current, end-point determination, by producing the linear portion of the titration curve to cut the residual-current line,g gave results similar to, but slightly less precise than, those of Procedure A. Typical sets of 6 results, here expressed as the average apparent normality and standard deviation of the mercury(1) titrant solution, obtained at thiocyanate and perchloric acid concentrations of 0.3 and 0.02 N, respectively, were 0.0129 0.00014 by Procedure A, 0.0127 0.00019 by the L-curve method, and 0-0128 &- 0.00016 by the residual-current method. At this concentration of thiocyanate, titrations at various acidities up to 0.08 N gave similar results.Although the results were insensitive to changes in acidity, they were markedly dependent upon the concentration of thiocyanate. This is illustrated by the average results of the titra- tion of 5 x N copper(r1) that are listed in Table I. Titrations at high thiocyanate con- centration were run at low acidity to avoid rapid decomposition of the medium. As deter- A systematic examination of the titration of TABLE I EFFECT OF THIOCYANATE CONCENTRATION IN THE TITRATION OF 5 x 1 0 - 5 ~ COPPER(II) Apparent mercury (I j normality Thioc yanate concentration, 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.5 N Perchloric acid concentration, 0.02-0.1 ti 0-02-0*08 0-02-0-06 0.02-0-05 0.02-0-05 N 0*02-0.05 0*02-0.05 0.0 1-0.03 0.01-0.03 0.01 Number of runs 7 5 3 2 3 3 3 3 3 1 method A 0.0122 0.0123 0.0125 0.0127 0.0133 0.0135 0.0138 0.0150 0.0153 0.0 195 L-curve method 0.01 15 0.0118 0.01 19 0.0121 0.01 24 0.01 29 0.0132 0.0141 0.0143 0.01 80 1 Residual-current method 0.0120 0.0123 0-0123 0-0127 0.01 3 1 0.01 34 0.0137 0.0147 0.0149 0.01 87 mined by the dichromate - iodide method described by Berka, Vulterin and Zyka,4 the normality of this batch of titrant was 0.0124,.Sets of six titrations of 1 0 - 4 ~ copper(I1) in 0.3 N potassium thiocyanate - 0-02 N perchloric acid medium gave, as average apparent normalities of this same batch of titrant, 0.0129, 0.0127 and 0.0128, by Procedure A, L-curve and residual-current methods, respectively.Precise titration is therefore impossible unless the concentrations of both copper(r1) and thiocyanate are essentially constant.30 STOCK : AMPEROMETRIC TITRATION OF SUBMILLINORMAL [AlZdySt, VOl. 91 I I I I 0 0.1 0'2 0.3 0.4 Volume of titrant, ml Fig. 2. Titration of 50 ml of 50 ~ L N copper(r1) with 0.0131 N mercury(1) per- chlorate after arbitrary prc-addition of 0.5 microequivalcnts of copper(I1). P and R are yre-addition and residual-current lines, respectively The titrations were repeated and extended in acid thiocyanate media that also contained potassium iodide.Fig. 2 shows a typical composite titration curve. The results, all obtained at a perchloric acid concentration of 0.02 x, are summarised in Table 11. In each set of 3 experiments, the successive titrations were carried out at potassium iodide concentrations of 0.02 N, 0.04 N and 0.06 N, respectively. At least within these limits, the concentration of this halide is obviously unimportant. As determined by the method described by Berka, Vulterin and Z9ka,4 the normality of the titrant solution was 0.0131,. The results are essen- tially unaffected by 10-fold changes in the concentrations of either thiocyanate or copper(I1) ion, or by the presence of chloride or bromide. In the concentration range to 1 0 - 4 ~ copper(I1) and in the presence of iodide ion, end-point location by Procedure '4 is therefore precise and accurate to within &1-5 per cent.Although less precise, the residual-current method gives acceptable results. The conventional L-curve method8 is not to be recom- mended for this titration, nor is it applicable at the lower end of the copper(I1) concentration range. TABLE I1 TITRATION OF COPPER(II) IN THE PRESENCE OF IODIDE ION Thiocyanate Copper (11) concentration, concentration, N P 0.1 50 0.3 10 0.3 50 0.3 50 0.3 100 0.8 50 1.4 50 Average millinormality and standard deviation . . . . . . . . Apparent mercury(1) millinormality L -~ 7 7 Residual-current method A L-curve method method 13.0, 13.4, 13.2 12-3, 12.8, 12.6 12.6, 13.3, 13.1 13.1, 13.3, 13.3 Failed 13.3, 13.3, 13.1 13.1, 13.1, 13.0 12.9, 12.7, 12.5 13-1, 12.9, 12.9 13.3, 13.2, 13.3 13.1, 13-2, 12.9 13.3, 13.0, 13.3 13.5, 13-2, 13.0 13.3, 12.9, 12.6 13.5, 13.0, 12.8 13.3, 13.2, 13.2 13.0, 13.0, 13.1 13.5, 13.3, 13.3 13.2,* 13.0,t 13.2: - - 13.2, & 0-14 12.8, & 0.28 13.1, + 0.25 * 0.02 N chloride present.t 0.02 N bromide present. 3 0.02 N each chloride and bromide present.January, 19861 COXCENTRATIONS OF COPPER(I1) WITH MERCURY(1) PERCHLORATE 31 End-point location by Procedure A or, less satisfactorily, by the residual-current method, was found to be applicable in the titration of micronormal concentrations of copper(I1). Fig. 3 shows some typical titration curves obtained with the titrant diluted 10-fold with N perchloric acid. At a concentration of approximately p ~ " , copper(r1) can be titrated with a precision of about 5 per cent.An end-point is recognisable at a copper(I1) concentration of 0.6 p~ (curve D); in this titration, the residual current accounted for more than 60 per cent. of the initial current. Standardisation of the titrant against known amounts of copper(rI), at concentrations similar to those to be determined, is recommended for submillinormal titrations and is essential when the concentration of copper(I1) is in the micronormal range. Method B can be used when the copper sample is presented as a highly dilute solution. Further dilution is minimised by the use of concentrated reagents and the end-point is located by the residual-current method. Results obtained with a 10 p~ copper(I1) solution were precise to within 5 per cent.Volume of titrant, mI Fig. 3. Pre-addition titration of 50 ml of highly dilute copper(I1) solution with approximately 0.0013 N mercury(1) per- chlorate. The upper and lower bars across each titration curve are the pre-addition and residual-current lines, respectively. Copper(I1) concentration: curve A , 6 p ~ ; curve R, 2.4 p ~ ; curve C, 1.2 p ~ ; curve I>, 0.6 p~ D I sc u s SION CU(TI) + Hg(r) -+ CU(I) + Hg(I1). The marked depression of the formal potential of the mercury(I1) - mercury(1) couple by thiocyanate and the voltammetry of this couple in thiocyanate medium have been discussed el~ewhere.~ Initially, complexation by thiocyanate will tend to lower the formal potential of the copper(r1) - copper( I) couple. However, if the reaction proceeds normally, copper(1) will be produced as soon as the titration is started.Although the solubility productlo of copper(1) thiocyanate is only about 10-14, this compound is appreciably soluble when the excess of thiocyanate ion is large. On the assumption that the solubility equation given by Fridman and Sarbaevll holds approximately outside the stated range of 0.7 to 5 N thiocyanate, the solubilities of copper(1) thiocyanate at 25" C are calculated to be 4 x N, 2 X N and 8 X N, in 0.1 N, 0.3 x and N potassium thiocyanate solutions, In this titration, the underlying reaction is32 STOCK [Analyst, Vol. 91 respectively. Since the total copper concentration in the present work was always less than 2 x ~ O - * N , copper(1) will remain in solution.A consideration of the values of the formation constantslo of the various thiocyanate complexes of copper(I1) and copper(1) leads to the conclusion that the formal potential of the copper(I1) - copper(1) couple will be raised by the presence of an excess of thiocyanate ion. At 25” C, a potential shift of from +Ova7 to +0.49 volt is calculated for thiocyanate concentrations of 0.1 N to N. The approximate formal potential under the titration conditions should therefore be +0.63 volt (hydrogen scale) or $0.39 volt (with respect to the S.C.E.). The calculated formal potential (25.C) of the mercury(I1) - mercury(1) couple in 0.1 N to N thiocyanate ranges from -0.09 to -0.35 volt (hydrogen scale). Since the formal potential of the copper couple is a t least 0.7 volt more positive than the formal potential of the mercury couple, the titration reaction should go to analytical completion unless the kinetics are unfavourable.Although quick and quite precise under rigidly controlled conditions, the titration exhibits stoicheiometry that is dependent upon the concentrations of thiocyanate and of copper(I1) ions. High titrant normalities, obtained at high concentrations of thiocyanate, are almost certainly due to the partial reduction of copper(I1) by the medium- 2Cu2+ + (2% + 2)SCN- + ~CU(SCK)(F’~ + (SCT\jI2 . . * - (1) 3(SCX), + 4H20 -+ 5SCN- + SO,2- + 7H+ + HCiK , . - (2) where n has integral values of 1 to 4. This reduction has been recently studied by Kolthoff and Okinaka,12 who observed a large hydrogen wave even in freshly prepared 0.001 K copper(I1) in N potassium thiocyanate solution. No copper(I1) remained in the solution after overnight standing.However, the reduction is much slower in 0.1 N potassium thiocyanate; about 55 per cent. of the original copper(I1) remained after 2 days. Since thiocyanogen oxidises iodide,13 the addition of an excess of this ion suppresses reaction (2) and may suppress reaction (1). Any iodine produced will be titrated4 with mercury(1) along with the remaining copper(II), so that the apparent normality of the titrant should become independent of the concentrations of both thiocyanate and copper( 11) ions. Although iodide can form complexes with both copper(I1) and copper(1) ions,14 these effects, and the resulting change in the formal potential of the copper(I1) - copper(1) couple, are likely to be small in the presence of an excess of thiocyanate ion.However, iodide com- plexes mercury(I1) ion very strongly,15 and should cause the formal potential of the mercury(I1)- mercury(1) couple to become more negative than in an iodide-free thiocyanate medium. This work was carried out with the partial support of the United States Atomic Energy Commission (Contract AT(30-1) - 1977) and was completed at the Imperial College of Science and Technology, London. The facilities afforded by the College authorities, in particular by Professors R. M. Barrer and T. S. West, are gratefully acknowledged. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. REFERENCES Tarayan, V. M., and Arutyunyan, A. A., Zavod. Lab., 1953, 19, 900. Tarayan, V. M., “Merkuroreduktrometriya,” Izdatel’stro Erevanskogo Universiteta, Erevan, Matsuo, T., J . Chem. SOG. Japan, Ind. Chem. Sect., 1955, 58, 962. Berka, L4., Vulterin, J., and Zgka, J., Chemisl-Analyst, 1963, 52, 122. Stock, J. T., and Heath, P., Analyst, 1965, 90, 403. Kolthoff, I. M., and Tanaka, N., Analyt. Chem., 1954, 26, 632. Stock, J . T., “hmperomctric Titrations,” Interscience Publishers, a division of John \;Z’iley and Stock, J . T., op. cit., chapter 1 . -, op. cit., pp. 520 and 558. SillCn, L. G., and Martell, A. E., “Stability Constants of Metal-Ion Complexes,” The Chemical Fridman, Ya. D., and Sarbaev, Dzh. S., Russian J . Inorg. Chem., 1959, 4, 835. Kolthoff, I. M., and Okinaka, Y . , Rec. Trav. Chim. Pays-Bas., 1960, 79, 551. Kaufmann, H. P., and Gaertncr, P., Ber. dtsch. chem. Ges., 1924, 57, 928. Sillen, L. G., and Martell, A. E., op. cit., p. 338. U.S.S.K., 1958, p. 144. Sons Tnc., New York, 1965, chapters 7, 9 and 10. Society, London, 1964, p. 121. 3 , op. cit., p. 341. -- Received June 24th. 1965