IH ( lHlWll‘lll HIM \ \ I —1_‘_. M 10000 (no-m1 ”.111; mcm LYTEC sawnm in mm ZN A BUF‘FE‘RED MCKEL SULFfiTE bOLfiMN Thesis {'3 1 im 3.3;;an cf ML 5. {VLECZ‘E‘LGML :HTB CSLL ESE 3'“ “lam be sub“ ‘f'z’flh 55' {H 'tl‘ 5&6J’L -5. _ 4 . -.... flea—62:2-— Hm ._,_ This is to certify that the thesis entitled The Electrolytic Behavior of Iron in a Buffered Nickel Sulfate Solution. presented by Richard L. Nyquist has been accepted towards fulfillment of the requirements for M degree in Chemi stry (Physical) Azémq Major profesfor Date March 9, 1951 0-169 'x' I Van; ' “thyrfkt; l 3 ‘1. Jr‘- .59 ”lit"; 344 “he"; a ' A) ‘ 1“ v i"? ..' ‘V :l . '9 ',"- -”\.Z- . "5“" :zfvsiffa;‘?tih‘fé r‘itafilt ". ‘92; L 't um? fétfid» “' l.» «mg 54% Y' "/ “1‘, . e, ' fiwk-fi' .. gffiémw. g}! a 4?? *qslitfii .’ '2‘ on .‘kltfla «if; =._j‘%”¥1 {1’34 T" 7 ya, in. ' ,e‘ l flag/.3 ‘ £1} ' n. V l t 1984‘“! gt.” s‘- 33 I’ll '3’ ' w.-'_ 5‘. ' - I. fir . a. . A THE ELECTROLYTIC BEHAVIOR OF IRON IN A BUFFERED NICKEL SULFATE SOLUTION BY Richard LeRoy Nyquist A THESIS Submitted to the School of Graduate Studies of.Miehigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemistry 1951 CHEMISTRY DEPT. 'T 2-3 4 \ .37 '. "‘7 TV (\J't l Q. .ufi/g/ AC KNOE’ILEDGMENT The author wishes to express his sincere appreciation 'to Dr. D. T. Ewing, Professor of Physical Chemistry, for his guidance and assistance throughout the course of this investi- gation. 904 ' n 0 ~ H LI 0] TABLE OF CONTENTS I. INTRODUCTION . . . . . . . . . . . . . . . . II. PROCEDURE . . . . . . . . . . . . . . . . . Preparation of the nickel stock solution . Purification of the stock solution . . . . Preparation of the cathode . . . . . . .. Analytical methods . . . . . . . . . . . Total iron . . . . . . . . . . . . . . . Ferrous iron . . . . . . . . . . . . . . Ferric iron . . . . . . . . . . . . . . III. RESULTS . . . . . . . . . . . . . . . . . . Stability of ferrous and ferric ions in the nickel stock solution . . . . . . . . . . Behavior of ferrous ions in the nickel solution . . . . . . . . . . . . . . . . . Behavior of ferric ions in the nickel solution . . . . . . . . . . . . . . . . . Behavior of a mixture of ferrous and ferric ions in the nickel solution . . . . . . . Behavior of ferric ions when anolyte and catholyte are separated . . . . . . . . . IV. CONCLUSIONS. . . . . . . . . . . . . . . . . REFERENCES . . . . . . . . . . . . . . . . . . . comb-PPWNNNH 13 19 '19 2h 27 INTRODUCTION This investigation was carried out to examine the electrochemical behavior of ferrous and ferric ions in a buffered nickel sulfate solution. Iron is one of the more common impurities found in nickel solutions, and consequently there have been previous studies made on the effect of iron in such solu- tions. W. D. Gordon has compiled a number of references to these studies (2). H. J. Wiesner found that ferrous iron was fairly soluble even at a pH as high as 6.0 (electrometric), but the solubility of ferric iron decreased considerably at a pH of 2.8 and 3.0 and at a pH of h.0 or higher it was insoluble (8). Ewing, Brouwer and Werner have completed a study of the effects and removal of iron with respect to several types of nickel solutions from which nickel is electrodeposited (1). Throughout this investigation a buffered nickel sulfate-chloride solution wasused at pH 2.2. A low pH was necessitated by the fact that at higher pH's ferrous ions are oxidized to the ferric state by aid. Furthermore, the solubility of ferric ions is decreased as the pH is increased (8). Tests were made to assure that no chemical oxidation of the ferrous or reduction of ferric ions took place in the nickel solution. PROCEDURE Preparation of the Nickel Stock Solution The composition of the nickel stock solution which was used throughout the experimental work is as follows: Ni SO#.7H20 300 g./l. 40 oz./gal. Ni 012.6H20 60 g./l. 8 oz./gal. Total Ni 82 g./l/ ll oz./ga1. pH 2.2 Temperature #000 lhOOF Current Density h.3 amp./dm? 40 a.s.f. The anodes were electrolytic nickel. Except for the case where a porous cup was used all the runs were made with one liter of stock solution. The pH was measured throughout this work with a Beckman Laboratory Model C pH meter, which was checked with 0.05 M potassium acid phtholate, pH 3.97. Purification of the Stock Solution Iron, as an impurity, was removed by raising the pH of the solution to 5.3 - 5.5 with nickelous carbonate, bubbling purified air through for twenty-four hours, and then filter- ing. The iron in the precipitate was in the form of ferric hydroxide. The pH of the solution was then lowered to 2.2 with c.p. grade sulfuric acid. The removal of iron was deemed necessary so that when a run was started the iron could be added as the ferrous or ferric salt without having a mixture of the two. Preparation of the Cathode A strip of sheet steel, 5.08 cm. by 9.24 cm. (2 in. by 3.6 in.), was used as the cathode. It was first cleaned with carbon tetrachloride and then electrocleaned for five to ten minutes in an alkaline cleaner of the following com- position: Sodium hydroxide 21 g./l. 2.8 oz/gal. Sodium metasilicate 15 g./l. 2.0 oz./ga1. Trisodium phosphate 18 g./l. 2.4 oz./gal. Sodium carbonate 6 g./l 0.8 oz./gal. Sodium acetate 7 g./l. 0.9 oz./gal. Sodium lauryl alcohol sulfate 0.05 g./l. 0.0067 oz./gal. Temperature 90°C l9h°F Current Density 10.77 amp./dm? 100 a.s.f. The electrocleaner was made up fresh as needed and a steel plate was used as the anode. After electrocleaning the steel cathode was immersed in a 20% hydrochloric acid solution for one minute, rinsed with iron-free, redistilled water, connected to the negative line from the switchboard with the e.m.f. applied, and placed in the solution midway between the two nickel anodes. It was important to have the electrical potential applied when the cathode was immersed in the solution so that deposition of the nickel would take place instantane- ously and prevent any of the steel from dissolving into the solution. Anodes were placed at each end of the solution with the cathode equidistant from each to avoid having any low current density areas on the cathode. Analytical Methods ,Tatal Iron. - The method of analysis for small amounts of iron in nickel solutions was devised by E. J. Serfass, et a1. (7). The iron from a 1 ml. sample of the nickel solution was com- plexed with cupferron, extracted from the nickel solution by amyl acetate, and then in turn extracted from the amyl acetate with nitric acid. After reducing the iron in the nitric acid to the ferrous state and adjusting the pH, o-phenanthroline was added to form a complex iron ion, which is a reddish color. A sample of this solution was placed in a Klett-Sommersan calorimeter with green filter No. 54 and the calorimeter reading was recorded. This same procedure was repeated without the addition of the sample of the nickel solution to obtain a blank reading. This .blank reading was then subtracted from the sample reading. A calibration curve, Fig. l, was obtained by analyzing nickel solution samples with known amounts of iron in them and plotting the iron concentration versus the calorimeter reading. Ferrous Iron. - Two ml. of concentrated sulfuric acid (c.p. grade) were pipetted into a 100 ml. beaker and diluted with approximately the same amount of water (iron-free, redistilled). Five ml. of the solution to be analyzed were then pipetted into the beaker and more redistilled water was added to bring the volume up to 60 - 70 ml. This solution was then titrated potentiometrically with 0.001 N potassium dichramate (analytical) using a Beckman Laboratory Model C pH meter, with platinum and calomel electrodes, to record the e.m.f. values (3). The solution in the beaker was stirred during the titration. The e.m.f. values recorded after the addition of each increment of potassium dichramate were plotted versus the volume in milliliters of dichramate. The paint on the curve at which the rate of change of the slope was zero was the end point, and the corresponding volume of dichramate solu- tion was recorded. To avoid repeated calculations, a graph, Fig 2, of milligrams of ferrous iron per liter of nickel solution versus the volume in milliliters of 0.001 N potassium dichramate was drawn. One ml. of the dichramate solution is equivalent to 0.056 mg. of ferrous iron (6). Although a 5 ml. sample was analyzed, it was desired to express the results on a liter basis. multiplying 0.056 by 1000/5 a conversion factor of 11.2 was obtained. That is, for every milliliter of 0.001 N potassium dichramate used to’reach the end point there were 11.2 mg. of ferrous iron per liter of nickel solu- tion. Standard solutions with various concentrations of ferrous ions were made up and analyzed by this method. The analysis was found to be accurate to within 10.5 mg./liter. Ferric Iron. - The concentration of ferric iron was obtained by the difference between the total and ferrous iron analyses Hilligrams iron per liter 120 100 80 ha 20 4/ 50 100 150 200 Calorimeter units Figure 1. Calibration curve for the determination of iron in nickel solutions. (1.0 ml. sample.) 250 300 Iilligrams ferrous iron per liter 100 ,/ 123L56789 Hilliliters 0.001 N potassium dichramate Fig. 2. Calibration curve for the determination of forrgus iron in the nickel solution. (5.0 ml. sample. RESULTS Stability of Ferrous and Ferric Ions in the Nickel Stock Solution To be sure that there would be no chemical oxidation of ferrous ions or reduction of ferric ions in the nickel solution two tests were made. First, 100 mg. of ferrous iron, as FeSOh.7H20, were added to a liter of the purified stock solution. The solu- tion was analyzed for total and ferrous iron, then allowed to stand for twenty-four hours at 40°C with mechanical agitationzis in an actual run except that there was no anode or cathode and no current was flowing. At the end of the prescribed time a sample of the solution was analyzed for total and ferrous iron again. The results from the two sets of analyses were the same and in neither case was there any ferric iron. A similar test was set up to determine the stability of the ferric iron, added as F92(SO .XH20 to the stock :33 solution. As in the first test, the results of the two sets of analyses were the same, showing that no reduction of the ferric ions took place. The results of these two tests proved that there was no detectable chemical oxidation of ferrous ions or reduc- tion of ferric ions in the nickel solution. The fact that ferric ions were not reduced is logical since the test was carried out in an oxidizing atmosphere. In the case of the ferrous ions, however, there might be a tendency to expect oxygen in the air to cause oxidation, and such is the case .at higher pH's. It should be noted that if a solution of ferrous sulfate be made sufficiently acid it can be, and is, used as a titrating agent which is so slowly oxidized that it need be restandardized only every few days. Behavior of Ferrous Ions in the Nickel Solution About 100 mg. of ferrous iron were added to one liter .of the purified stock solution and allowed to dissolve. After a sample of the solution had been removed for analysis ' the current was passed and a properly prepared cathode was immersed in the solution. Thereafter, a sample of nickel solution was withdrawn from the solution every three hours wand analyzed for total and ferrous iron. When the concen- tration of the total iron was below 10 mg. per liter the current was discontinued and the run ended. Table I shows the data obtained from this run. The concentration of ferrous ions decreased at a faster rate than the total iron, indicating that some ferrous ions were oxidized to the ferric state. The total iron depletion curve obtained from this run was compared in fig. 3a totshe depletion curve obtained by Ewing, et al. from a nickel solution containing ferrous sulfate (1). The curves were nearly identical. TABLE I ELECTRO-CHEMICAL REACTIONS OF FERROUS IONS IN THE NICKEL SOLUTION Time Amp. Hrs. Fe, total Fe+2 Fe+3 hrs. per Liter mg./l. mg./l. mg./l. 0.00 0 103 103 0 3.00 12 48 Q5 3 6.00 2h 25 19 6 9.00 36 15 6 9 12.00 48 10 2 8 13.75 55 ’ 6 1 5 milligrams iron per liter 100 80 Figure 3. solution. total iron, u. 6 8 10 12 Time, hours The electrochemical reactions 1‘ ferrous iron in the nickel Current density of 1;..3 am ./ . pH 2.2 and at LLOOC. (l) (2) ferrous iron, and (3 ferric iron. 0 - expcr; ~.tal; O - calculated. Iilligrems iron per liter 100 80 \o 0“ 10 - 20 30 h0 50 Ampere hours per liter Figure 3a. A comparison of the iron depletion curve from Figure 3 with that obtained by living, at al. (1). O - values obtained in this investigation; A - values obtained by skiing, et a1. Behavior of Ferric Ions in the Nickel Solution When work was being initiated in this phase of the investigation the ferric iron was sometimes added to the solution several hours ahead of the time that the current was passed and the cathode put in place. In the analysis of the sample of the solution which was taken just before the run was started it was noted that some of the ferric ions had already been reduced to the ferrous state. Since the only difference between this solution and the solu- tion used in the stability test previously mentioned was that the nickel anodes were immersed in this solution, it appeared that there was an electrochemical reaction between ferric ions and metallic nickel. To confirm this possibility a purified nickel solup tion with ferric iron added was allowed to stand with nickel anodes, which were connected externally by a copper wire, but no external e.m.f. was applied. During the run the temperature of the solution was A000 and agitation was used. Analyses for ferrous and total iron were made on samples taken every three hours, and the results are shown in Table II. The results proved that there was a reaction between metallic nickel and ferric ions,.as would be expected from the following equations: 1h 1/2 Ni° - e . 1/2 wi+2 BO . + 0.250 v. Fe*3+ e I Fe+2 E0 I + 0.771 V. 1/2 Ni° . Fe+3- 1/2 Ni+2 + Fe+2 E° . + 1.021 v The E0 values are from Oxidation Potentials and are for acid solutions (5). Since the sign of the E° value for the total reaction is positive the reaction will proceed spontaneously as written. Furthermore, the equilibrium constant, K, is equal to the antilog of (E0) (nF)/(2.3 RT) or, 1.021/0.059, which is equal to 1.93 x 1017. This means that for all practical purposes the reaction proceeds to completion. The above values for E and K pertain to an aqueous acid solution, and since in this investigation the reaction was carried out in an almost saturated nickel salt solution the activities of the nickel, ferrous and ferric ions were changed and it is obvious that the e.m.f. and K values were different, but the sign of the e.m.f. value for the reaction remained the same so that the reaction proceeded spontaneously. Also, the reaction went to completion after 9 to 12 hours. To determine the effect of the above reaction when current was floWing through the nickel solution ferric iron was added and dissolved in the purified stock solution just before the current was passed and the cathode put in place. Results are shown in Table III. From this it can be seen that the ferric ions were reduced rapidly to the ferrous state. TABLE II ELECTROLYTIC REACTION OF FERRIC IONS AND NICKEL ANODES IN A NICKEL SOLUTION WITH NO EXTERNAL E. M. F. APPLIED Time Re, total. Fe+2 Fe+3 mg./1. mg./1/ mg./1. 0 min. 63 0 63 10 min. 63 3 6O 30 min 63 14 #9 1 hr. 63 18 A5 3 hr. 63 #9 1h 6 hr. 63 59 h 9 hr. 63 62 12 hr. 63 63 O Milligrams iron per liter n \/ /\ 0 / J i. 2 L 6 8 10 12 Time , hours Figure ’11; The electrol 1c reaction of ferric ions and nicks: anodes the nickel so ution with no external e.m.f. applied. pH 2.2 and at 140°C. (1) total iron, (2) ferrous iron, and (3) ferric iron. 0 - experimental; O - calculated. TABLE III ELECTRO-CHEMICAL REACTIONS OF FERRIC IONS IN A NICKEL SOLUTION Time Fe, total Fe+2 Fe+3 mg./1; mg./1. mg./1. 0 min. 70 6 64 15 min 65 22 #3 1 hr. 50 23 27 3 hr. 28 21 6 hr. 9 8 1 9hr. 3 ' 3 o 12 hr. 2 2 O Hilligrems iron per liter 2 I... 6 8 10 12 Time, hours Figure 5. The electrochemical reacti ns of ferri ions in e. r ':el solution. Current density of .3 snms./ . pH 2.2 and at l . 3. (1) total iron, (2) ferrous iron, and (3) ferric iron. 0 - experimental; O - calculated. l9 Behavior of a Mixture of Ferrous and Ferric Ions in the Nickel Solution A mixture of approfiimately equal amounts of ferric and ferrous iron was added to the purified stock solution. .Samples were-taken every two hours in this run and the results are shown in Table IV. The concentration of the ferrous iron dropped much more slowly than that of the ferric iron, since the ferric iron was continually being reduced to the ferrous state at the nickel electrodes. Behavior of Ferric Ions When Anolyte and Catholyte are Separated A porous cup was used to separate the anolyte and catholyte. Ferric iron was added to the catholyte, with results shown in Table V. There was some ferric iron in the bath which had not been removed when the bath was purified. Ferric iron in the anolyte was reduced to the ferrous state, while very little of the ferric iron in the catholyte was reduced, since it did not come in contact with the nickel anodes. Agitation was used in the catholyte, but because of lack of space it was not used in the anolyte. Consequently, the ions in the anolyte were not distributed evenly and when samples were withdrawn for analysis they did not necessarily contain a representative amount of iron ions. This was shown by the inconsistent results. TABLE IV REACTIONS OF A MIXTURE OF FERROUS AND FERRIC IONS IN A NICKEL SOLUTION gig? Fe, total Fe'2 Fe""3 mg./1, mg./l. mg./l. o 110 51 ' 59 2 79 a8 31 1+ 51+ 39 15 6 36 28 8 8 28 20 8 IO 20 1h 6 12 13 7 6 Milligrams iron per liter L P L N I 1“ n I a Tit a 5% Figure 6. The electrochemical reactions of ferric and ferro ions in the nickel solution. Current density of lt.3 awn/din. pH 2.2 and at 110°C. (1) total iron, (2) ferrous iron, and (3) ferric iron. 0 - experimental; 0 - calculated. TABLE V REACTIONS OF FERRIC IONS IN A NICKEL SOLUTION 'NHEN ANOLYTE AND CATHOLYTE ARE SEPARATED BY A POROUS CUP Time Fe, total Fel'2 Fe+3 hrs. mg./1. mg./l{ mg./1. Anolyte 0 8 1 7 3 6 l 6 8 l 9 10 10 O Catholyte 0 1+9 0 1+9 3 33 3 30 6 23 t. 19 9 16 A 12 lilligrams iron per liter 2 it 6 8 Time, hours Figure 7. The electrolytic reactions of ferric ions in the nickel solu ion when anolyte and cath- olyte are separated y a porous cup. Cgrrent den- sity of 11.3 smps./ H 2. 2 and at 11.068: 0. £1), (1..) total iron,o (2), (5 ferrous iron, (3) ferric iron. 0, n - experimental; . I - calculated. 0 - catholyte; a - anal e. 21.. CONCLUSIONS In the first run where iron was added as the ferrous salt there was a small amount of ferric ions formed. This was due to ferrous ions being oxidized at the anode and then swept away before they could be reduced again to the ferrous state by the nickel. Ferric ions in contact with nickel electrodes immersed in the nickel solution while no current was passing were reduced to the ferrous state. This proved that there was galvanic action between metallic nickel and ferric ions. By passing current through the solution of ferric ions the reduction of the ferric ions was accelerated considerably. By adding a mixture of ferrous and ferric iron to the nickel solution it was possible to compare the behavior of the two forms. The concentration of the ferrous ions de- creased slowly, while the concentration of the ferric ions dropped at a rapid rate. Both oxidation states were being electro-deposited at the cathode, but the ferric ions were being reduced to ferrous ions at the nickel anode, thereby depleting thewsupply of ferric ions and replenishing the ferrous ions. The use of a porous cup to separate the anolyte and catholyte allowed the reactions of ferric ions in the catho- 1yte to be observed without the interference of the ferric ion nickel reaction. Some of the ferric ions, instead of being reduced completely to the metallic state, were reduced to the ferrous state and then swept away from the cathode. Most of the ferric ions were, however, reduced directly to the metallic state and electro-deposited on the cathode. The most important result of this investigation was the observation of the ferric ion - nickel galvanic action. There was some ferrous iron oxidizediat the anode to ferric iron and the reverse process took place at the cathode, but these reactions were minor as compared to the galvanic cell reaction. After a nickel solution had been electro- lyzed for a few hours there was little or no ferric iron present. It had been hoped to determine whether the ferric ion- nickel reaction were complete or if there were always a few ferric ions present as long as there was iron in the solu- tion. However, it was found that if the concentration of ‘ total iron were less than ten milligrams per liter the total iron analysis was not accurate. In some cases there appearai to be more ferric iron than ferrous iron, while in other cases the ferrous iron analysis gave a concentration slighthr higher than the total iron analysis. This situation is obviously impossible. A clue to the answer may be found in the fact that ferric iron was precipitated easily by cupferron, while it was somewhat more difficult to precipitate ferrous iron. It is possible that there was always enough' unprecipitated ferrous iron to give a total iron concentration a few milli- grams per liter lower than the actual concentration. A study of the total iron analysis at concentrations less than ten milligrams per liter using ferrous iron would be‘ enlightening. It should be kept in mind that the results of this investigation pertain only to a pH 2.2 nickel solution. At the higher pH's (above 3.0) there is a tendency for the ferrous ions to be oxidized by air, and the ferric ion concentration becomes limited. (l) (2) (3) (4) (5) (6) (7) ‘(8) REFERENCES Ewing, D. T., Brouwer, A. A., and Werner, J. K., "Effects of Impurities and Purification of Electro- plating Solutions: I. Nickel Solutions: 6. The Effects and Removal of Iron," Plating (in press). Gordon, W. D., "The Effects of Impurities and Their Removal from Nickel Plating Solutions." Master's thgsis, Michigan State College, East Lansing, 19h6. 22 pp. Kolthoff, I. M. and Furman, N. H., Potentiometric Titrations. First Edition; New York: John Wiley and Sons, Inc., 1926. 333 pp. Kolthoff, I. M. and Laitinen, H. A”, pH and Electro Titrations. Second Edition; NeW' orE: UEEE‘WIley and Sons, Inc., l9hl. 190 pp. Latimer, W. M., Oxidation Potentials. New York: Prentice-Hall, Inc., 1933. 352 pp. Scott, W. W., Standard.Methods of Chemical Anal sis. . Fifth Edition, VoI. I; NeW'YErE: D. Van Nostrand Company, Inc., 1939., 123h pp. Serfass, E. J., et al., "Determination of Impurities in Electroplating Solutions: IV. Iron in Nickel Plating Baths," ghg’Monthly Review, 33:1189-97, November, 19h6. Wiesner, H. J., ”The'Colloidal State and Precipitation of Certain Metallic Hydroxides in Concentrated Solutions of Nickel Sulfate." Ph.D. thesis, Michigan State College, East Lansing, l9h3. 2h pp. II‘- .. ‘1‘“ All fair LBW! " " . /. T 41037 25 J N997 Nyquist \ 5904 (I (I . WM \\ x\xuww\\\\\\\\\\\\\\\\\\\\W 93 02446 79