Journal of Research of the National Bureau of Standards-C. Engineering and Instrumentation Vol. 63C, No.1, July-September 1959 A Coulometric-Titration Coulometer Stanley W. Smith! and John K. Taylor
(April 28, 1959) .. A highly precise coulometer is described which permits time integration of currents totaling 100 coulombs or more with a precision of about 1 part in 100,000. The current to be integrated oxidizes hydroquinone in an electrolysis cell, producing quinone and acid. The quinone is then reduced by constant-current co ulometric titration, the end point being indicated by hydrogen-ion concentration measurements. 1. Introduction Since hydrogen ions are involved in the reaction , it is possible to use a hydrogen-ion concentration I" The increased interest in coulometric methods of (pH) measurement as the means for detection of chemical analysis has created a need for a current the end point, that is, when the reverse reaction is integrator of higher precision than the mechanical, complete. This has the distinct advantage over electrical, or electrochemical devices now available. other possible systems in that an indicating electrodc Such an integrator should be simple to operate and system and related measuring equipment are readily give precise results for small quantities of electricity. available which are very stable and of a high degree Overall considerations indicated that it should be of refinement, making additional investigation in possible to integrate a total charge of 100 coulombs, this area unnecessary. correspondi.ng to 1 milliequivalent of chemical reac ti.on, with a precision of about 1 part in 100,000, 2 . Reagents and Apparatus using an electrochemical coulometer. The purpose of this investigation was to develop such a device. The supporting electrolyte is made up from rea The simplest arrangement \>;Thich might be ex gent-grade sodium chloride, precipi Lated from the pected to fulfill the requirements is an acidimetric sa turated solution by hydrogen chloride gas, and type of electrochemical cell in which the current to reery tallized from water. be integrated generates strong acid in a neutral The hydro quinone is recl'Y Lalli "ed once from air electrolyte. The amount of acid so generated is free water and dried under reduced pressure. The then determined precisely by constant-current coulo product of the recrystallizaLion consists of white metric titration.2 A titration coulometer of the needles appearing slightly violet in large quantities. acidimetric type has been reported in the literature,3 The electrical system used is shown in figure l. but preliminary investigations showed that the The voltage i upplied by a 48-v lead storage bat electrode reactions involved were not efficient to tery isolated from ground. Variable resistor R1, a the degree required for this application. voltage dropping resistor, suppli es several ranges of As a result of the experiments which showed the current. Switch 8 2 is a double-pole, double-throw defects in this method, it was decided to use an knife switch of the Leeds and N orthl'Up potentiom oxidation-reduction type of reaction. Upon in eter type. When in position (1), the current i vestigation of several possibilities, the hydro quinone passed through a dummy resistor Rz which is ad- coulometer was selected as best fitting the require ments placed on the method. The reaction involves the oxidation of hydro quinone to quinone by the I current to be determined, and the subsequent reduc 62 t tion back to hydro quinone by constant-current (2) .....--.I ----~I--+ coulometry. The reactions may be represented as s, TO "TIM" follows:
Hydro quinone + in tegra ted curren t ~ q uinone + FJ+( 1)
J-l++quinone+constant curr ent ~hydroquinon e (2) Reaction (2) is the reverse of (1), so that the quantity of electricity required to return the cell to the start ing point, reaction (2), is equivalent to that respon (I) sible for reaction (1). ~x ':'rY:' 1 Present address, Delco·Remy Division, General Motors Corporation, N------" Anderson, Ind. TO CELL 'J. K. T aylor and S. W. Smith, J. Research NBS (in press). ' J. A. P age and J. J . Lingane, Anal. Chim. Acta 16 175 (1957). FIGU RE 1. Schematic diagram of electrical circuit.
65 justed to have the same resistance as the cell. When WA SHING TOWERS ASCARITE switch S 2 is thrown in position (2), the current flows through the cell and simultaneously the quartz crystal controlled time-interval meter, TIM, B erke ley Model 7250 CD, is triggered by the pulse from battery B 2• Switch S3 is a reversing switch, allowing the cur rent to flow through the cell in either direction. When switch 8 4 is in position (1), the voltage drop across standard resistor R3 is measured by the pre cision potentiometer Pl' When switch S4 is in position (2), t he voltage across R3 opposes the stand ard cell, S.O., and the difference is measured by Pl' The resistor network R4, Rs, and R6is for adjustment of the current. The coarse adjustmen t, R 4, is com posed of resistance decade boxes. Fine regulation is obtained by means of a fixed resistor and a carbon compression resistor, Rs. The resistance range of the network, R5- R6, is approximately 18 ohms, and COMPARTMENT A COMPA RTMENT C the range of R4 is 2,000 ohms in 0.1 ohm steps. The resistance is adjusted during a determination to FIGU RE 2. Coulometric cell and associated apparatus. maintain a null deflection of the galvanometer of potentiometer Pl' Switch Ss, when in position (1), connects the cell the cell through a water trap. A hole is provided to the source of current to be integrated accordillg for the introduction of the glass electrode. A to r eaction (1). Position (2) is used for the eoulo cylindrical platinum gauze electrode, 5 by 1.5 cm in metric titration according to reaction (2). diam, completes the asscmbly. The end point is determincd by a commercial pH Compar tment 0 is separated from the rest of the meter equipped with standard glass and reference cell by a 5-percent agar plug as described above. electrodes. In addition, a precision potentiometer It contains the silver- silver chloride working elec is connected across the recorder terminals of the pH trode, constructed from a piece of fin e silver 5 by 10 meter, allowing a precision of 0.002 pH units in de cm, and the calomel reference electrode. termining the end point.2 To hasten the rate of attainment of equilibrium at The stability of the pH meter is of prime impor th e end point, it is desirable to immerse the cell in a tance, since th e reproducibility of the end point is water bath maintained at 45° C. A satisfactory directly related to this factor. A commercial meter arrangement is to place the cell in an aluminum pan providing automatic zero standardization is very sitting on a combination hotplate-magnetic stirrer satisfactm-y in this respect. to provide both stirring and temperature control. The cell and associated equipment are shown in figure 2. Water-pumped nitrogen is passed suc 3. Experimental Procedure cessively over copper at 500° C to remove oxygen, through ascarite to remove carbon dioxide, through Before assembly, th e cell is cleaned in hot sulfuric a 1-N sulfuric acid wash-solution to entrap any dust acid-chromic acid solution and rinsed by drawing from the ascarite, and finally through two wash large quantities of hot distilled water through the solu tions of distilled water to remove traces of acid sintered glass disks to ensure removal of all traces of and saturate the gas with water vapor. The func the cleaning solution. tion of the stopcock arrangement will be explained The agar plug is formed in the side arm of the under experimental procedure. clean cell as follows. An agar solution, made by The cell consists of two electrode compartments, dissolving 5 g of agar-agar in 100 ml of a 2-jVf 4.5 by 10 cm, connected by a bridge containing 3 sodium chloride solution and heating to near boiling, sintered glass disks, of which (1) and (2) are of is poured into the side arm. The gel sets on cooling medium porosity and (3) is of fin e porosity. All and will not soften unless the temperature is raised agar plug at the entrance port of compartment 0 above 60° C. effectively prevents transfer of solution to or from The supporting electrolyte is prepared by adding this compartment. The bridge is provided with 10 g of sodium chloride to each compartment, plus tubes for blowing the solutioll from its compartments sufficient distilled water (approximately 85 ml) to by means of nitrogen pressure. cover the platinum electrode when the bridge is The stopper assembly which closes working com filled. After the stirring bar is added, the stopper partment A is filled with three glass tubes: one for assembly is fitted to the cell and it is placed in bubbling nitrogen through the solution for the position in the water bath. The nitrogen manifold purpose of removing carbon dioxide and oxygen; one connections are then made as shown in figure 2. for flowing nitrogen over the solution during the After the pH electrodes have been standardized determination; and one for exit of the nitrogen from 111 a buffer solution at the operating temperatme, 66 45° 0, the glass elec trode is insertec) in the s~opp er TABLE 1. Performance data assembly. The r e f er en ~e el ec tr~d e IS J?lace d 111 ~h e isolated compartment wIth the sIlver- SLIver chlonde 1958 Reco very 1958 electrode. Nitrogen is then bubbled through Lh e cell and the bath Lempera Lur e is raise d Lo 45 ° . Jail . 14 ______% Jail. 30 ______% Jan. 15 ______100. 000 100.001 After purging for 20 to 30 min , the stopcock i~ 100. 001 Feb.2i ______99.999 Jan. 16 ______100.002 Oct. 1L ______99.999 turned to allow the gas to flow over the surface. of Jail . 16 ______99. 999 Oct. 14 ______• ______100.002 the solution. The bridge is then fill ed by applymg Jan.29 ______• ______JOO . OOO Oct . 14 ______100. 001 suction to the center compartments. The current is adjusted to 10 ma and passe.d through the cell until the pH reaches 7. At tIllS 5. Discussion point, t he solution in both central .compartments lS The method does not involve any precision chem pushcd into compartment A ?y mtrog e~ pressure. ical operations which would require the services of After mixing, it is drawn back mto the br~dg e. The a highly skilled chemist. pH is again brought ~ o 7 by: electrolys1s . and the The precision and accuracy of this method depend process is repeated unt11 t~ er e IS no change 111 pH. on the control maintained over a number of factors, Hydroquinone (0.50 g) IS now added to compart including: Accidental loss of electrolyte, diffusion ment A and allowed to dissolve. The equilibrium and migration of ions, efficiency of electrode reac pH (approximately 5.6) is recorded and used as th e tions, slop e of the pH-time curve, stability and end point. calibration of the circuitry, and freedom from ground For accurate results, the cell is preconditioned in loops. the following manner . With the platinum el cctro~ e Accidental loss of electroly te could occur through as anode, the current is adjusted to 100 ma and aCld the gas exit tub e, around the stopper assembly, or is genel"ated for approximately ~ ,2 00 sec. The through the bl'idge into the isolated .compal~tment . '-, polarity is then reversed and ?ase IS generated foJ' I If the nitro O"en is flowed over the solutIOn durmg the about 1 150 sec. The curren t IS then changed to ] 0 enLire run band the stopper assembly is carefully ma and'the generation of base is continued to the made and fi LLed, th ere is no appreciable loss from end point. The bridge is emptied, the~ r e fill ~ d , this source. The possibility of loss through the and the solu tion is brough t to the end pomt agam . bridge is negligibl e if th e agar plug is fi,:m and does This rinsin g process is continued un til no change no t soften a t Lh e tempera ture of operatLOn. It has occurs. The cell is now ready to be use d as a been demonstrated expel'imentally Lh at only a mi coulometer. v- nuLe amount of acid rcache the second bl'idge com Performance testin g is carried out in the followin g pal' tment, mainly by convection through the fri ts I manner. The current is adjusted through Lh e (1 and 2). None is transferred Lhrou gh an agftl" dummy r esistor to 100 mao ,Vith the plaLinum plug in good condiLion. . elec tro·de as anode, acid is generated for about 1,000 The use of fl, rather concentrated salt solu tlOn to sec, while the CUlTent is monitored conLinuously Lo carTY the current effectively minimizes e~ ectromi g.r a maintain a con s tan t and known value. Lion of hydrogen ions across the bndge dunng The acid so produced is Lhen titrated by reversin g reaction (1). the polarity, and passin g a cons tan t and.accuratel.y The reaction s at the platinum elec trode are 100 measured curren t of about 100 ma unt11 appw:\'"l percent efficien t as long a the silver- silver chloride mately 95 percent of the acid is ncutralizcd. rr:he electrode is maintained in a separate compartment. current is then redu ced to 10 rna and thc end pom t In experiments where this electrode was installed in (original equilibrium pH) is. ~·e a ~h e d by the same the same compartment with the generating electrode, process as described for condltlOmng tb e cell. the efficiency of the reduction reaction ( ~ ) was fo.upd The procedure is the s am ~ for integrating a.n un to be less than 100 percent due to the fimte solubIlity known eurrent. However, If the total quantIty of of silver chloride in the electrolyte and the subsequent charge is not known approxima~ el y , the pH must reduction (plating out) of silver on the cathode. '{l be observed occasionally to aVOId generatll1g past It is necessary, of course, to exclude all traces of the end point. When the solution is back. to .the foreign acid or alkali from the: cell. :U:or example, end point, the cell is ready for another detenmnatlOn. the leaching of 10- 4 ml of cleanll1g solutIOn from the walls of the cell would result in an error of about 3 4. Results parts in 10,000. . The precision of the detection of the end POll1t When following the procedure described in section depends on the reproducibility of. the pH-meter rea~ 3, a precision of approximately 1 part in 100,000 J?l-ay ings and on the slope of the pH-tIme curve. ExperI be obtained. The res ul ts of a number of determma ence has shown that the experimental arrangement tions made over a period of 10 m?nths ~r e given i!1 is of sufficient stability and sensitivity to mal~e the table 1. These data were all obtall1ed W1.th apprOXI pH measurements reli able to about ± 0.002 umt. mately 100 coulombs input and are given as percent The slope depends on the concentration apd vo~ recovery; namely, (coulombs, oxida tion/coulombs, ume of the electrolyte in the cell. Hydro.q':lll1o~e IS reduction) X 100. a very weak acid, K 1= 1.1 X 10- 1O , and exhIbIts shght ;0
67 buffering action at the end point. Since the slope 6.8 increases as the solution is diluted, it is desirable to use the most dilute solution practicable. At the 6.6 - same time, the volume of the working compartment of the cell must be kept small to insure significant 6 .4 changes in the pH for small amounts of reduction z 0 near the end point. § 6.2 --' Theoretically, 55 mg of hydroquinone is oxidized 0 by 100 coulombs of electricity. To minimize con 68