Scholars' Mine
Masters Theses Student Theses and Dissertations
1972
A study of some complexometric titrations in nonaqueous solvents
Ngo The Hung
Follow this and additional works at: https://scholarsmine.mst.edu/masters_theses
Part of the Chemistry Commons Department:
Recommended Citation Hung, Ngo The, "A study of some complexometric titrations in nonaqueous solvents" (1972). Masters Theses. 5071. https://scholarsmine.mst.edu/masters_theses/5071
This thesis is brought to you by Scholars' Mine, a service of the Missouri S&T Library and Learning Resources. This work is protected by U. S. Copyright Law. Unauthorized use including reproduction for redistribution requires the permission of the copyright holder. For more information, please contact [email protected]. A STUDY OF SOME COMPLEXOMETRIC TITRATIONS IN
NONAQUEOUS SOLVENTS
BY
NGO THE HUNG, 1940-
A THESIS
Presented to the Faculty of the Graduate School of the
UNIVERSITY OF MISSOURI-ROLLA
In Partial Fulfillment of the Requirements for the Degree
MASTER OF SCIENCE IN CHEMISTRY T2732 92 pages 1972 c.l
Approved by To my Parents ii
ABSTRACT
Complexometric titrations of calcium, zinc and lead with polyamino carboxylic acids: ethylenediaminetetraacetic acid (EDTA), 1,2-diamino cyclohexanetetraacetic acid (DCTA), ethyleneglycol-bis(2-aminoethylether) tetraacetic acid (EGTA) and tetraethylenepentamine (tetren) have been investigated and compared in the following organic solvents: methanol, dimethyl formamide, dimethyl sulfoxide and methyl ethyl ketone.
Various end point detection methods have been used: direct visual titration with metallochromic indicators and instrumental detection by photometry, potentiometry (mercury electrode and lead ion selective electrode) and amperometry.
Calcium, zinc and lead can be determined up to trace levels (ppm) under specific conditions.
A concrete application of this thesis is the determination of zinc or calcium in a sample of polystyrene within a few minutes while the classical methods will necessitate several hours. iii
ACKNOWLEDGMENT
The author wishes to extend his appreciation to his advisor, Dr.
William R. Carroll, Professor of Analytical Chemistry, for suggesting the topic of this research and for introducing him to the fascinating field of titrations in nonaqueous media. Without his expert guidance this thesis could not have been led to completion.
He also wishes to thank Mr. Edward W. Hennecke for diligent help in some glass blowing situations.
He is particularly grateful to the United States Government (Agency for International Development) and to the Vietnamese Government for financing his stay in the United States. iv
TABLE OF CONTENTS
Page
ABSTRACT . •.•.•...•....•...••.•..•.•.•..•...•..•..•...•...•..••..... ii
ACI TABLE OF CONTENTS • ••.•••••••..•••••••..•••.•••.•.•.••••..•.•..•••• iv LIST OF FIGURES ...... vii LIST OF TABLES ...... viii ABBREVIATIONS. . • . • • • . • • . • . . • . . . • . . . . . • • . . • . • ...... • ...... ix I. INTRODUCTION • •••••••••••••••••••••••••••••••.••••••••••• 1 II. REVIEW OF LITERATURE • ...... 2 A. Studies with a Mixed Solvent System .•••••.•.• 2 B. Studies with Pure Anhydrous Organic Solvents ••••••.•• 5 c. Conclusion ...... 6 III. EXPERIMENTAL SECTION ••••••••••••.••••••••••••••••••••.•.• 8 A. Apparatus ...... 8 B. Reagents and Solutions •••••••••••.••••••••••.••.••... 10 1. Chelan So 1 utions ...... 10 2. Standard Metal Solutions •••.••••••.•••...•••..•.• 13 3. Indicator Solutions •••••••...•.•.••••..•.•.•...•• 14 c. Procedure ...... 16 1. Direct visual Titrations with Metal1ochromic Indica tors ...... 16 a} Titrations of zinc with EDTA ••••••••••..•••• 16 b) Titrations of zinc with DCTA ••••••••••••••••• 17 c) Titrations of zinc with EGTA •••••.•••••••.••• 17 d) Titrations of zinc with tetren ••••••••••••••• 17 Table of Contents (continued) Page e) Titrations of calcium with EDTA11 ••••• - ••••••••• 18 f) Titrations of calcium with DCTA ...... 18 g) Titrations of calcium with EGTA • •.••••••....•.• 18 h) Titrations of calcium + zinc with DCTA •..•..... 19 i) Titrations of calcium + zinc with EGTA • ••••.••. 19 j) Analysis of calcium and zinc in a sample of polystyrene...... 19 2. Photometric Titrations...... 20 3. Potentiometric Titrations...... 22 a) Titration of mercury with the mercury elec trade...... 22 b) Titrations of zinc with the mercury electrode...... 22 c) Titration of calcium with the mercury elect rode ...... 22 d) Titration of lead with the lead elec trade .•...• 23 e) Titration of zinc with the lead electrode •..... 23 4. Anlperometric Titrations .••••••.•.•••...••••..•.••.•. 24 IV. RESULTS AND DISCUSSION.. . • • • • • . . . . • • • . . . • • . . . • • • . • . . • . • • • . • 25 A. Direct Visual Titrations with Metallochromic In- die a tors...... 25 1. General connn.ents ...... 25 2. Titrations of Zinc ...... 26 3. Titrations of Calcium ...... 39 4. Titrations of Calcium + Zinc ...... 45 5. Analysis of Calcium and Zinc in a Sample of Polystyrene...... 48 vi Table of Contents (continued) Page 6. Conclusion...... 48 B. Photometric Titrations ••••••••••••••••••••••••••••• 51 c. Potentiometric Titrations •••••••••••••••••••••••••• 59 D. Amperometric Titrations •••••••••••••••••••••••••••• 73 v. CONCLUSION • •••••••••••••••••••••••••••••••••••••••••.•. 75 REFERENCES. • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • . • • • • • . 77 VITA. • • • • • • • • . • • . . • • . • . . . • • . . • . . • . • . • . • . • • • . . . . • • . • • • . • . • • . . • . . . . 82 vii LIST OF FIGURES Figure Page 1. Absorbance Curves in Dimethyl Sulfoxide ...... •...... •. 52 2. Photometric Titrations of Zinc in Dimethyl Sulfoxide .•...... 54 3. Photometric Titration of Zinc in Dimethyl Formamide ...•..... 56 4. Potentiometric Titrations of Calcium and Zinc with EDTA ..•.. 61 5. Potentiometric Titrations of Lead with EDTA and EGTA .....•.. 66 6. Potentiometric Titrations of Zinc with EDTA, DCTA and EGTA .. 69 viii LIST OF TABLES Table Page I. Results of Titrations of Zinc with EDTA Using Metallochromic Indicators ••••••••••••••••••••••••••••• 27 II. Results of Titrations of Zinc with DCTA Using Metallochromic Indicators ••••••••••••••••••••••••••••• 31 III. Results of Titrations of Zinc with EGTA Using Metallochromic Indicators ••••••••••••••••••••••••••••• 34 IV. Results of Titrations of Zinc with Tetren Using Metallochromic Indicators ••••••••••••••••••••••••••••• 36 V. Results of Titrations of Calcium with EDTA in Methanol Using Methylthymol Blue ••••••••••••••••••••••••• 40 VI. Results of Titrations of Calcium with DCTA Using Metallochromic Indicators ••••••••••••••••••••••••••••• 41 VII. Results of Titrations of Calcium with EGTA Using the System Zincon + Zinc-EGTA ••••••••••••••••••••••••• 43 VIII. Results of Titrations of Calcium + Zinc with DCTA Using Methylthymol Blue ••••••••••••••••••••••••••••••••••••• 46 IX. Results of Titrations of Calcium + Zinc with EGTA Using Metallochromic Indicators ••••••••••••••••••••••••••••• 47 X. Results of Titration of Zinc and Calcium in Polystyrene...... 4 9 XI. Results of Photometric Titrations of Zinc •••••••••••.••••••• 58 XII. Results of Potentiometric Titrations of Calcium and Zinc with EDTA ••••••••••••••••.••••••••••••••••• 60 XIII. Results of Potentiometric Titrations of Lead with EDTA arid EGTA...... • ...... 65 XIV. Results of Potentiometric Titrations of Zinc With EDTA, DCTA and EGTA. • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • 68 ix ABBREVIATIONS DCTA: 1,2-diamino£Yclohexanetetr~cetic acid DMSO: Di..!!!ethyl _!Ulfoxide EDTA: Ethylenediamine.E_etra.!_cetic acid EGTA: Ethylenealycol-bis(2-aminoethylether)-tetra~cetic acid S.C.E.: saturated calomel electrode Tetren: Tetraethylenepentamine 1 I. INTRODUCTION Complexometric titrations of metal ions with chelating agents like polyamines or polyaminocarboxylic acids have been extensively studied for two decades.(l-6) All these studies, though, deal with aqueous titrations and the field of complexation in non-aqueous media has been relatively uninvestigated. When metals in organic products are determined by classical aqueous complexometry, it is required to destroy the organic matrix by an ashing procedure followed by a dissolution of the residue. This method is time comsuming and the ashing procedure can lead to some loss of material. If the sample can be dissolved in an organic solvent and a direct complexometric titration be performed in this medium, any loss of material inherent to the ashing procedure will be prevented and the time required for the analysis will be tremendously shortened. It is the purpose of this thesis to investigate some complexometric titrations in various organic s6lvents and to find a rapid method of determination of calcium and zinc in polystyrene. 2 II. REVIEW OF LITERATURE Studies of complexometric titrations in non-aqueous media using polyaminocarboxylic acids as chelating agents are met with two main difficulties: 1) the insolubility of the chelating agent in the organic solvents and 2) the lack of knowledge concerning the forma tion of the metal-chelate in these media. These studies can be classified into two divisions: those which involve a mixed solvent system (organic solvent + water) and those which deal with pure organic anhydrous solvents. A. Studies with a Mixed Solvent System The metals to be titrated are dissolved in a mixed solvent system and the chelating agent is always dissolved in water. The role of the organic solvent is to dissolve the sample and/or to prevent the blocking of the metallochromic indicator. Schwarzenbach and Ackerman (7) found that the stability constants of some EDTA-alkaline earth chelates are 100 times greater in a 48% alcoholic medium than in water. Gotschalk (8) discovered the same property in titrating aluminum with EDTA in a 50% ethanolic medium and using dithizone as indicator. Potentiometric titrations of several metal ions with EDTA, nitrilo triacetic acid (NTA) and N,N-Di(-hydroxyethyl)glycine in a water + pyridine mixture have been investigated by Siggia (9) who used various 3 electrode systems as Pt vs. calomel, Hg on Pt vs. calomel or Ag vs. calomel. A lead ion-selective membrane electrode was used by Rechnitz and Kenny (10) to determine lead by a differential measuring technique in various concentrations of water and methanol, dimethyl sulfoxide, 1,4- dioxane and acetonitrile. The determination of metal additives in petroleum products has been often investigated in various water + organic solvent mixtures. Gerhardt and Hartmann (11) determined calcium or zinc in lubricating oils and concentrates by dissolving the samples in acetone, complexing the metals with an excess of aqueous standard solution of EDTA and then back titrating the excess EDTA with an a~ueous solution of magnesium chloride using Eriochrome Black T as indicator. Fisher (12) used an extraction procedure to separate barium, calcium and zinc in oil samples but this method could not avoid excessive emulsion formation and sometimes incomplete extraction. A toluene + isopropyl alcohol mixture was used by Crump (13) to dissolve lubricating oils. The additive metals (barium, calcium, lead and zinc) were complexed with an excess of standard EDTA which was then back titrated with a standard magnesium iodate solution using Eriochrome Black T as indicator. The method was rapid but the solution was cloudy due to the presence of a suspension which tended to obscure some end-points. Instead of using EDTA Hooks and Noar (14) used 4 diethylenetriaminepentaacetic acid (DTPA) and dithizone as indicator to determine zinc directly while the other metal additives were complexed with an excess of DTPA. This excess DTPA was back titrated with a standard magnesium solution. Combining their methods Crump and Noar (15) found them having an accuracy comparable to classical "wet" methods of the ASTI1 and the Institute of Petroleum. An analysis took 30 minutes. A double extraction procedure was used by Hypta (16) to determine barium and calcium in lubricating oils. An analysis took 3 hours and calcium could be determined down to 0.06% and barium down to 0.02%. Kiss (17) investigated various water + organic solvent mixtures to avoid the blocking of Eriochrome Black T and found that ethanol was the best organic solvent. Methylthyrnol Blue and Methylthyrnol Blue + Fluorexone have been used by Kiss, Gaal, Suranyi and Zsigrai (18) to determine consecutively calcium and magnesium with EDTA in a 60% ethanolic medium. Micro amounts of magnesium (0.0243 to 12.16 mg) have been titrated by Kiss and Suranyi (19) with EDTA and Methylthyrnol Blue in a 60% ethanolic medium. Magnesium sterate dissolved readily in 96% ethanolic medium and magnesium could be titrated with EDTA and Methylthyrnol Blue by Kiss and Knezevic (20). A high frequency technique to determine bivalent metal ions in 5 water + ethanol and water + ethanol + benzene mixtures was investigated by Casassas and Fernandex (21) who used 1,2-diaminocyclohexanetetraacetic acid (DCTA) as chelating agent. They found that the DCTA-metal complexes are more stable than the EDTA-metal complexes and the high frequency titration could be used in non-buffered media. B. Studies with Pure Anhydrous Organic Solvents Yoshimura and Tamura {22) used a conductometric method· to analyze antimony (III) and (V) in dimethyl formamide with EDTA. They found that the end point corresponds to molar ratio EDTA:Sb equal to 4:1. An amperometric end point detection was used by Arthur and Hunt (23) to analyze lead, cadmium and zinc in petroleum products. They dissolved the sample in 2-propanol and the chelating agent was 1,2- diaminocyclohexanetetraacetic acid (DCTA) dissolved in 95% ethanol made 0.33 M in ethanolamine. Micro titrations being possible with a high frequency technique have been successfully investigated by Fernandez, Casassas and Garcia -5 MOntelongo (24) who studied the complexation between 4.10 M EDTA in ethanol and cadmium, mercury, zinc, copper and lead solutions in -4 -5 concentration range from 3.3 x 10 M to 2.5 x 10 M. A complexation reaction is always followed by a release of protons, which can be titrated by a high frequency technique. Fernandez and Casassas (25) applied this method when they studied the reaction be- tween 1,2-diaminocyclohexanetetraacetic acid (DCTA) and some bivalent 6 Metals in methanol +benzene (1:4 and 4:1) mixtures. Acetone, acetonitrile and methanol were the media used by Rechnitz and Kenny (26) when they titrated potentiometrically cupric ions with EDTA, ethylenediamine, tetraethylenepentamine and 5,6-dimethyl-1,10- phenanthroline and a solid membrane cupric ion electrode. Direct visual titrations of zinc, cadmium and magnesium in water insoluble stearates with EDTA, nitrilotriacetic acid (NTA), 1,2-diamino cyclohexanetetraacetic acid (DCTA) and N-(2-hydroxyethyl)-ethylene diamine-NNN'-triacetic acid (HEDTA) have been performed by Kiss, Zsigrai, Cirin and Krizsan (27) in formamide and dtmethyl sulfoxide with Eriochrome Black T and PAN as indicators. Magnesium and zinc in micro amounts from 1 to 12 pg in a 0.5 ml sample could be successfully deter mined. A thermometric end point detection was used by Kiss (28) in the direct titration of EDTA, 1,2-diaminocyclohexanetetraacetic acid (DCTA) and nitrilotriacetic acid (NTA) with manganese (II) nitrate in dimethyl sulfoxide and in the indirect titration of some bivalent and tervalent metals with EDTA or DCTA in dimethyl sulfoxide. C. Conclusion From the survey of literature the following conclusions can be drawn: 1. Complexometric titrations in non-aqueous media are possible. 7 2. The theory of complexometric titrations in aqueous medium is mainly centered around the notion of pH. In a non-aqueous medium the pH is not well defined and this explains why all the investigations up to now have been highly empirical and are still in a stage of trial and error experiments. 8 III. EXPERIMENTAL SECTION A. Apparatus 1. Potentiometric studies were performed with a Sargent recording titrator, model D; a Beckman Zeromatic SS 3 pH meter and a Beckman Ex- pandomatic pH meter. The following electrodes have been used: a) Glass electrodes: Beckman No. 39301 on the Expandomatic pH meter, and Beckman No. 41263 on the SS 3 pH meter. b) Modified calomel reference electrode made from a Beckman quartz fiber standard calomel electrode with saturated KCl - methanol electrolyte, and a Standard calomel electrode, Beckman No. 39402. c) Silver billet indicator electrode, Bec~n No. 39261. The electrode was cleaned periodically by light buffing with steel wool. d) Mercury electrode, J - type, Sargent-Welch No. S-30438. Before use the mercury electrode cup was cleaned with 5 N nitric acid, rinsed with distilled water followed by deionized water and dried in an oven at 1050 C. e) Lead specific ion electrode, Orion model 94-82. Before use the electrode was polished with the abrasive plastic supplied with the electrode by the manufacturer and soaked in the same lead standar- dizing solution for five minutes. 2. Photometric instruments included a Beckman DK-2A ratio re- cording spectrophotometer used for running absorption curves and a Bausch and Lomb Spectronic 20 spectrophotometer equipped with a 55 9 milliliter, one inch diameter test tube containing a micro Teflon coated stirring bar used for measuring absorbance during titrations. The test tube was surrounded by a tube opaque to light and closed by a rubber stopper with a hole in the middle to let the tip of the buret dip into the tube. A magnetic stirrer was placed under the instrument, under the bottom of the test tube. The stirrer was stopped for 15 seconds before any absorbance reading and set into motion for 30 seconds after each increment of titrant. 3. Amperometric studies were performed with a Sargent Polarograph model XV. Three different electrolysis vessels have been used: - polarographic, H form, with stopcock, Sargent No. S-29401 - polarographic, Heyrovsky, 10 ml volume, Sargent No. S-29370. - 50 ml Pyrex glass beaker. The reference electrodes were a mercury pool used with the Heyrovsky polarographic electrolysis vessel, or a standard calomel reference electrode prepared according to Hume and Harris (29) with a flexible salt bridge containing a saturated KCl solution and terminated by a 6 mm diameter glass tubing containing an agar KCl gel dipping into a 10 mm diameter glass fritted tube containing a minimum amount of solvent and an appropriate electrolyte for contact. The glass fritted tube was fitted in a rubber stopper and plunged into the measuring solution the level of which was 1 or 2 millimeters higher than the level of the solution in the tube.(30) 10 The working electrodes used were: - a capillary dropping mercury electrode, 6 em length, Sar- gent No. S-29417. - a silver-silver chloride electrode, Beckman No. 39261. 4. All glassware used was Pyrex glass. 10 milliliter burets with straight or curved delivery tips were graduated in 0.02 milliliter divisions. B. Reagents and Solutions 1. Chelan Solutions. a) Ethylenediaminetetraacetic acid (EDTA)(Matheson Coleman & Bell) 99% purity was used as received. Solutions were prepared by weighting an approximate amount required for 0.01 M solution and after dissolution in the solvent were standardized against a 0.01 M calcium chloride standard aqueous solution using Calcon as metallochromic in- dicator. The pH of the solution was maintained around 13 by the use of diethyl amine. EDTA dissolved in methanol made 1 M in ethanolamine. EDTA dissolved in dimethyl formamide upon heating near 148°C. The so- lution was stable for less than six hours and a voluminous white pre- cipitate appeared after that period of time. This precipitate could be redissolved by heating the solution again but the normality had to be redetermined. EDTA dissolved in dimethyl sulfoxide upon heating near 1060 C. The solution was very stable over months. 11 EDTA did not dissolve in the following organic solvents: methanol, iso propanol, acetonitrile, propylene carbonate, 1,4-dioxane, benzene, methyl ethyl ketone, methyl isobutyl ketone. Aqueous 0.01 M standard EDTA solutions used to determine the nor mality of metal solutions were prepared by dissolving an approximate amount of the disodium salt (Fisher Scientific Co.) in water and the titre was checked against the 0.01 M primary standard calcium solution. b) 1,2-diaminocyclohexanetetranaacetic acid (DCTA) (Ciba Geigy, Chel CD) was used as received. A 0.01 M solution of DCTA in methanol, made 0.5 M in ethanolamine was prepared by weighing an approximate amount of DCTA. The titre was determined against the 0.01 M primary standard calcium solution using Calcon as metal indicator and diethyl amine to adjust the pH to 13. A standard zinc nitrate solution can also be used with Eriochrome Black T as indicator at a pH of 10 obtained with a buffer recommended by Schwarzenbach (1) (dissolve 70 g ammonium chloride in 570 ml ammonia of density 0.90, dilute to one liter with demineralized water.) DCTA solutions in dimethyl formamide or dimethyl sulfoxide were obtained by dissolving the appropriate amount of reagent in the organic solvent under heating. Standardization was done with calcium or zinc standard solutions as with the case of methanolic solutions. c) Ethylene glycol-bis(2-aminoethylether)-tetraacetic acid (EGTA)(Ciba-Geigy, Chel DE) was used as received. Solutions were 12 prepared by dissolving an approximate amount of reagent in the organic solvent and were standardized against a standard zinc aqueous solution with Zincon as indicator at pH around 9.5 according to Ringbom (31) or with Eriochrome Black T as indicator at a pH of 10. EGTA dissolved with heating in dimethyl sulfoxide but dissolved in methanol made 0.25 M in ethanolamine at room temperature. d) Tetraethylenepentamine (tetren)(Eastman Kodak, technical) was purified as the pentahydronitrate according to Reilley and Holloway (32): dissolve 75 g of tetren in 150 ml of ethanol (solution A), cool 0 near 0 C. Prepare a mixture of 150 ml of water and 150 m1 of ethanol and 2.5 moles (approximately 140 ml of nitric acid, density 1.42) of nitric acid (solution B) and cool near 0°C. Add slowly with stirring solution B to solution A, maintaining the temperature under 100 C. Filter the precipitate by suction. Recrystallize five times from a- queous 5% (by volume) nitric acid solution. The white precipitate is washed with acetone and dried at 50°C. To prepare a 0.01 M solution of tetren, weigh approximatly 0.6 g of the salt and dissolve in 100 ml of organic solvent. Tetren being a primary amine reacts with ketones to form condensation products (Schiff bases) (33), therefore it was not prepared in these solvents. A 0.0129 M tetren solution in methyl ethyl ketone was found to have a concentration of 0.0041 M one month later. Tetren salt dissolves in methanol made 1 M in ethanolamine and in dimethyl sulfoxide without 13 ethanolamine with slight heating. Due to the fact that nitrates are decomposed in dimethyl fomamide (34) no experiments have been attempted with tetren solutions in this solvent. Tetren solutions were standardized according to Reilley and Schmid (35) against a standard zinc nitrate solution with Eriochrome Black T as indicator and triethanolamine maintaining a pH of 7.8. 2. Standard metal solutions. a) A one liter 0.01 M calcium standard solution was prepared by dissolving 1.000 g of calcium carbonate (primary standard, Thorn Smith Co.) in 20 m1 water and 2 ml concentrated nitric acid, followed by dilution with demineralized distilled water to volume. b) A 0.01 M zinc standard solution was prepared by dissolving 0.8137 g of zinc oxide (Mallinckrodt, analytical reagent) in 20 ml water and 2 ml concentrated nitric acid, followed by dilution to one liter with demineralized distilled water. c) Calcium, zinc and lead perchlorates were prepared by dissolving calcium carbonate (primary standard), zinc oxide and lead carbonate (Mallinckrodt, analytical reagent) in the minimum amount of 0.2 ~ perchloric acid solution under heating to expel the carbon dioxide and heating to dryness on the sand bath. To prepare 100 ml of 0.01 ~ solutions dissolve 0.3, 0.4 and 0.5 g of calcium, zinc or lead perchlorate respectively in the appropriate solvent. Completely anhydrous solutions of these metals have been prepared according to 14 Erley (36) and Starke (37) by dissolving the above mentioned amounts of salts in 20 m1 of organic solvent and 2 m1 of 2,2-dimethoxypropane followed by 30 minutes of stirring on a combination magnetic stirrer - hot plate. Toward the end of that period the solution was heated to expel the last trace of methanol and acetone and diluted to 100 m1 with the organic solvent. No better results have been obtained with these completely anhydrous metal solutions. 3. Indicator solutions. a) Eriochrome Black T (Fisher Scientific Co.) was prepared in 0.1% solution by dissolving 20 mg in 15 ml triethanolamine and 5 m1 absolute alcohol or 25 mg in 25 ml dimethyl formamide or dimethyl sul foxide. The solutions were stable for months. b) Calcon (Fisher Scientific Co.) solutions were prepared in 0.05% solution in methanol (12.5 mg in 25 ml methanol) and 0.1% solution in dimethyl sulfoxide (25 mg in 25 ml dimethyl sulfoxide). The solutions were stable for months. c) Zincon (Aldrich Chemical Co.) was prepared 0.1% in methanol or dimethyl sulfoxide and 0.065% in dimethyl formamide. The indicator dissolves in methyl ethyl ketone only when tetrabutylammonium hydroxide (25% in methanol) (Eastman Kodak) is present: (13 mg of Zincon in 20 ml of methyl ethyl ketone+ 0.5 m1 of tetrabutylammonium hydroxide). All Zincon solutions were stable for one week only. d) Methylthymol Blue (pentasodium salt) (Eastman Kodak) is 15 not soluble in alcohol (38), dimethyl sulfoxide or dimethyl formamide. A 0.1% solution of Methylthymol Blue was prepared by dissolving 20 mg of the indicator in 2 ml of water followed by dilution with 18 m1 of methanol. e) Aqueous Mercury-EDTA indicator solution (Anderson Labora tories, Inc.) was used as received. The concentration of the complex was determined according to Starostin (39) by back titration of an ex cess of standard bismuth nitrate solution with 0.01 M EDTA using Xylenol Orange as indicator and found to be 0.020 M. 4. 1,4-Dioxane (City Chemical Co., New York) was purified according to Huber (40) by filtering through a chromatographic column containing a strong acid cation exchanger, Amberlite IR-120 (Mallinckrodt). Other re agents, analytical grade, were used as received. Methanol and methyl isobutyl ketone were from Matheson Coleman & Bell, dimethyl formamide dimethyl sulfoxide and methyl ethyl ketone from Fisher Scientific Co. To adjust the "pH" of the solution during titrations, 0.1 M, 1 M or concentrated solution of monoethanolamine, diethylamine, triethyl amine, triethanolamine, tetrabutylammonium hydroxide, tetramethylam monium hydroxide and perchloric acid were prepared in the solvent used for titration. Perchloric acid, giving an explosive reaction with di methyl sulfoxide (40), was first diluted with 1,4-dioxane. MOnoethanol amine being a primary amine reacts with ketones to form condensation pro ducts (Schiff bases) (33), therefore it was not used with these solvents. 16 C. Procedure 1. Direct visual titrations with metallochromic indicators. Titrations were performed in a 50 ml beaker, at room temperature, without using a dry box. The metal solution, one to five milliliters, was pipeted into 25 ml of solvent followed by two to five drops of metallochromic indicator. The acidity of the solution was brought to the desired value, read on the millivolt scale of the pH meter equiped with glass electrode and modified S.C.E. A magnetic stirrer was used continuously during the analysis. a) Titrations of zinc with EDTA Zinc was titrated in methanol with Eriochrome Black T, Methylthy mol Blue or Zincon as indicator. The acidity was maintained around -300 mV with ethanolamine (around -200 mV for the last indicator). The color changes were respectively from pink to blue for Eriochrome Black T, blue to yellow for Methylthymol Blue and blue to pink for Zincon. In dimethyl formamide the potential was maintained around -250 mV by use of diethylamine (Eriochrome Black T) or ethanolamine (Zincon). The color changes were respectively from pink to blue and blue to pink. In dimethyl sulfoxide the same indicators were used at a potential around -350 mV. When methyl ethyl ketone was used as solvent, the titrant was a solution of EDTA first dissolved in dimethyl sulfoxide then diluted with methyl ethyl ketone in a 1 to 3 volume ratio. Tetrabutyl ammonium hydroxide (20% in methanol) diluted 1:10 with the solvent was used to bring the potential around -400 mV with the occasional addition of perchloric acid (0.1 Min the solvent). The indicator Eriochrome Black 17 T changed color from pink-violet to blue. b) Titrations of zinc with DCTA In methanol as solvent, Eriochrome Black T, Methylthymol Blue and Zincon required respectively the potential values: -350 mV, -300 mV and -275 mV, which were obtained with 1 ml of ethanolamine, 1 ml of diethylamine and 0.25 ml of ethanolamine in each particular case, for a volume of 25 m1 of solution. They changed color respectively from pink to blue, violet to canary yellow and blue to pink orange. The reaction with Eriochrome Black T was very slow at room temperature but increased noticeably when the solution was heated to 60°C. Micro titrations of zinc in dimethyl sulfoxide were performed with Zincon as indicator at a potential around -350 mV, obtained with the addition of ethanolamine. The color change was from blue to pink orange. When methyl ethyl ketone was used, the solution had to be heated at 60 0 C, and Methylthymol Blue changed from blue to yellow at the end of the reaction. Two milliliters of diethylamine added to 25 ml of solvent gave a potential around -300 mV. c) Titrations of zinc with EGTA Zincon used in methanolic medium buffered with ethanolamine around -275 mV gave a very sharp color change from blue to pink orange. In dimethyl sulfoxide it reacted slowly with the potential set around -350 mV with the same buffer. Eriochrome Black T, on the contrary, gave a very sharp change from violet pink to blue in the same experi- mental conditions. d) Titrations of zinc with tetren The potential of the titrating solution was around -350 mV when 18 dimethyl sulfoxide served as solvent. Zincon changed color as in the other media. In methyl ethyl ketone the solution was buffered with triethanolamine, one molar in the same solvent. The ionic strength was kept at 0.1 by the addition of lithium or sodium perchlorate. Zincon was used as indicator and gave a very sharp end point. e) Titrations of calcium with EDTA In methanol calcium was determined with EDTA using Methylthymol Blue. The potential was around -275 mV and was attained by the use of 1 ml of ethanolamine for 25 ml of solvent. The indicator changed rapidly from blue to yellow with a brief transition of grey before the end point. f) Titrations of calcium with DCTA Calcon was used for the titration of calcium in methanol. Two milliliters of diethylamine added to the solution gave a potential around -300 mV. The indicator changed from pink to violet. With the same experimental conditions, Methylthymol Blue gave a sharper end point from blue to yellow in passing through a grey tran sition just before the appearance of the yellow color. In methyl ethyl ketone, with the same experimental procedure as above, the potential was -330 mV and Methylthymol Blue changed from blue to canary yellow with a transition of greenish yellow which was rather persistent. g) Titrations of calcium with EGTA In methanol calcium was titrated with the indicator system Zincon + Zn-EGTA. To 25 ml of methanol were added 1 ml of ethanolamine, 5 drops of Zincon, 2 drops of 0.005 M Zn-EGTA solution prepared by mixing equal parts of 0.01 M zinc perchlorate and 0.01 M EGTA in 19 methanol, and the calcium solution. The solution was titrated slowly and changed color from blue to orange. In dimethyl sulfoxide the same procedure was used but the Zn-EGTA solution was prepared in this solvent. The color change was from blue to pink. The potential was around -400 mV. If, instead of Zincon, Eriochrome Black Twas used, the color change was even sharper from pink to blue. Quantitative results were obtained if the ionic strength of the solution was maintained at 0.1 with potas sium perchlorate. h) Titrations of calcium + zinc with DCTA In methyl ethyl ketone with Methylthymol Blue as indicator and DCTA in methanol as titrant, the sum of calcium + zinc was determined with the solution heated to 60° C and buffered around -330 mV with 2 ml of diethylamine. The color change was from blue to yellow. i) Titrations of calcium + zinc with EGTA The sum calcium + zinc was determined in dimethyl sulfoxide, using Eriochrome Black T as indicator and ethanolamine as buffer at -300 mV. The ionic strength was maintained at 0.1 with potassium perchlorate. In methanol the indicator used was Zincon and the procedure was iden tical with the ones concerning the titration of calcium or zinc separately, but no need for the addition of the zinc-EGTA complex. j) Analysis of calcium and zinc in a sample of polystyrene One gram of polystyrene was dissolved in 25 ml of methyl ethyl ketone and 10 drops of 0.1 M perchloric acid in the same solvent. In the hot solution were added 2 ml of diethylamine and 5 drops of Methyl thymol Blue. The solution assumed a blue color and was titrated 20 immediately and slowly with a 0.001 M solution of DCTA in methanol until a yellow color was reached. The volume of titrant gave the sum of calcium + zinc. To titrate zinc, another aliquot was dissolved in the same way with the addition of 0.25 g of lithium perchlorate. To the resulting solution cooled to room temperature, were added 1 ml of 1 M triethanolamine in methyl ethyl ketone, and 5 drops of zincon, (0.1 %) in dimethyl sulfoxide. The solution assumed a blue color, with the potential around -110 mV. Zinc was then titrated with a 0.001 M solution of tetren in dimethyl sulfoxide.until the appearance of the pink color. 2. Photometric titrations. a) The absorbance curves of the indicators and of the metal- indicator complexes were recorded on the Beckman DK-2A instrument. The concentration of the indicator was around 10-6 M, the metal concentration about one hundred times larger. Although Eriochrome Black T, Calcon, Methylthymol Blue were also studied, only Zincon appeared to give the best resolved curves. Ac- cording to Rush and Yoe (41) a 0.13 % solution of Zincon has a con- centration of 0.002 M, therefore a 0.065 % or 0.1 % solution of the indicator has a concentration of 0.001 M or 0.0015 M respectively. The 0.1% soltuion in dimethyl sulfoxide was diluted 50 times with the solvent and 0.50 ml of the resulting solution was further diluted to -6 25 m1 with the same solvent, giving a 6 x 10 molar solution. One tenth milliliter of ethanolamine was added to provide a basic medium having a concentration of about 0.06 M in ethanolamine. The newly pre- pared solution was run on the Beckman DK-2A with dimethyl sulfoxide 21 serving as the blank solution. The zinc-Zincon solution was prepared by the same procedure but one milliliter of 0.01195 M solution of zinc perchlorate in dimethyl sulfoxide was added before diluting to 25 m1 with the solvent. The concentration of zinc was calculated to be -4 4.78 x 10 M. b) Photometric titrations were performed only for the deter- mination of zinc with Zincon as indicator. To 25 m1 of dimethyl sulfoxide were added 1 ml of 0.00119 M zinc perchlorate and 5 drops of 0.1 % Zincon in the same solvent. A paten- tial of the glass electrode vs. the modified S.C.E. around -350 mV was obtained by adjusting the acidity with ethanolamine diluted ten times with the solvent. The absorbance was measured at 650 nm while titrating with a 0.00103 M solution of DCTA in dimethyl sulfoxide. The concentration of Zincon was calculated to be 1.44 x 10-5 M in the titrating solution. With other titrants the same experimental conditions were followed except that 20 ml of solvent were used instead of 25, thus giving a -5 more concentrated Zincon solution (1.78 x 10 M). One milliliter aliquots of 0.00119 M, 0.00179 M and 0.00171 ~ zinc perchlorate solution in dimethyl sulfoxide were titrated respectively with 0.00103 M EGTA, 0.00104 M EDTA and 0.00102 M tetren, all dissolved in the same solvent. For the titration of zinc in dimethyl formamide, 1 m1 of 0.01085 M zinc perchlorate in dimethyl formamide was added to 25 ml of the solvent followed by 1 m1 of 0.065 % Zincon solution in the same solvent. The concentration of Zincon was calculated to be 3.7 x 10-5 M. The titrant was 0.0070 M EDTA in dimethyl formamide. When the absorbance curves of 22 Zincon and zinc-Zincon in dimethyl formamide were recorded they appeared to have the same shape as in the case of dimethyl sulfoxide but the maximum difference in the absorbances was at 625 nm instead of 650 nm. During the photometric titration the instrument was set at this wave length of 625 nm. 3. Potentiometric Titrations Two kindsofmetal indicator electrodes have been successfully used: the mercury electrode and the lead ion selective electrode. a) Titration of mercury with the mercury electrode To 40 ml of dimethyl sulfoxide were added 4 m1 of 1 M potassium perchlorate, 5 drops of ethanolamine diluted ten times with the same solvent and 5 drops of 0.01 M mercuric acetate, all dissolved in the same solvent. The potential indicated by the glass electrode was around -380 mV. The solution was titrated with 0.01046 M EDTA in di methyl sulfoxide and the potential of the mercury electrode recorded. b) Titration of zinc with the mercury electrode The former titration was stopped after the addition of 0.50 ml of titrant, during which period the reaction was already completed, 1 m1 of 0.01364 M zinc perchlorate in dimethyl sulfoxide was then added and the titration started again until a potential break was obtained. c) Titration of calcium with the mercury electrode To 40 ml of dimethyl sulfoxide were added 4 m1 of 1 M potassium per chlorate, one drop of 0.001 M mercuric acetate and 1 m1 of 0.00914 M calcium perchlorate, all dissolved in the same solvent. Ethanolamine diluted ten times with dimethyl sulfoxide was used to bring the potential of the glass electrode around -400 mV. The titration was performed with 23 0.01055 M EDTA in dimethyl sulfoxide. When methanol was used as solvent, to 25 ml of solution were added two drops of 0.020 M aqueous mercury-EDTA indicator solution and one milliliter of 0.01344 M calcium perchlorate in methanol. The titration was performed with 0.0100 M EDTA in methanol. d) Titration of lead with the lead electrode To 25 ml of dimethyl sulfoxide, made 0.1 ~with potassium per chlorate, were added ethanolamine, diluted ten times with the same solvent in order to have a potential at the glass electrode around -400 mV, and one milliliter of 0.0125 M lead perchlorate in dimethyl sulfoxide. The potential at the lead electrode was recorded during the titration with 0.01055 M EDTA in dimethyl sulfoxide. e) Titration of zinc with the lead electrode To 25 ml of dimethyl sulfoxide made 0.1 M with potassium perchlorate were added ethanolamine, diluted ten times with the same solvent in order to have a potential at the glass electrode around -400 mV., one drop of 0.0125 M lead perchlorate and one milliliter of 0.01364 M zinc perchlorate, both in dimethyl sulfoxide. The titration was performed with 0.01055 M EDTA in the same solvent. When 0.00845 M DCTA in dimethyl sulfoxide was used as titrant the same experimental conditions were followed except for the drop of lead perchlorate which had a concentration of 0.00125 M. With 0.0101 M EGTA in dimethyl sulfoxide as titrant the zinc solu tion had a concentration of 0.0171 M and the same experimental procedure was used with one drop of 0.001 M lead perchlorate. 24 4. Amperometric Titrations Before attempting an amperometric titration, different polarograms for dilute solutions of zinc or calcium (10-4 M) were run in order to determine the potentials of the plateaus. The amperometric titration was then performed with the voltage set to the value of the beginning of the plateau. A wide variety of experimental comditions were tried but no quantitative results were obtained. 25 IV. RESULTS AND DISCUSSION A. Direct Visual Titrations with Metallochromic Indicators 1. General comments. The results of titrations of zinc, calcium and their sum are listed in Tables I - IX. In performing these titrations the main problem was the search for suitable "buffer substances" which could be used in the solvents studied. Since there is no universal pH scale in organic media the only logical way of recording the acidity of the solution was to use the millivolt scale on the pH meter, equiped with the electrode assembly: glass electrode vs. modified S.C.E. This system has been investigated by Rorabacher et al. (42) in methanol, McClure and Reddy (43) in dimethyl formamide and by Reddy (44) in di~thyl sulfoxide who found it to be reliable. A comparison of the relative scales of acidity of methanol, dimethyl formamide, dimethyl sulfoxide and methyl ethyl ketone published by Bykova and Petrov (45) and by Kreshkov (46) show that the pK values for ionization of these solvents are respective ly 16.7, 24, 31 and 31. These figures prove that the "pH range" in these media vary widely but according to Petrov and Bykova (47) these values are valid only for pure anhydrous solvents and if traces of water are present, which is mostly the case in practical analytical determinations, then these figures change drastically. Bates (48) and Reddy (44) have studied some buffer mixtures in methanol and dimethyl sulfoxide respectively. Some of these mxitures (benzoic acid + sodium benzoate) have been used ( in dimethyl sulfoxide ) but did not give any good results. 26 The three metallochromic indicators mostly used, Eriochrome Black T, Methylthymol Blue and Zincon change color in basic aqueous media. Var ious amines have been tried to control the basicity of the solution during the titrations. The potential values listed in each determination were determined by two factors: 1) they must keep the basicity of the solution within a reasonable narrow range, 2) the indicator must change color at these conditions. The first factor was determined experimental ly by titrating a small amount of amine intended to be used as buffer, with a standard acid solution dissolved in the same solvent. The titration curve, titrant added vs. potential at the glass electrode, was drawn and the potential at the half neutralization point was re corded. As in the case of aqueous acidimetric titrations the neutral ization curve shows at the half neutralization point an almost flat portion in which the potentials do not vary very much. The possibility of a reaction between the amine buffer and the solvent was another problem, eg. ethanolamine cannot be used in methyl ethyl ketone (33). 2 Titrations of zinc. The results are condensed in Tables I - IV. With EDTA as titrant, Eriochrome Black T seemed to be the best in dicator in all the media studied. When methyl ethyl ketone was the solvent, EDTA was first dissolved in dimethyl sulfoxide then diluted (1:3) with the solvent. This experiment was performed in the hope to use it for the determination of zinc in polystyrene, since polystyrene does not dissolve in dimethyl sulfoxide but dissolves in methyl ethyl ketone or in a mixture of this latter solvent with a small amount of dimethyl sulfoxide added. Attempts to titrate zinc in dimethyl 27 Table I. Results of Titrations of Zinc with EDTA Using Metallochromic Indicators A. Solvent: Methanol 1. Indicator: Eriochrome Black T (Buffer: ethanolamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.635 0. 647 4 2.68 +1.90 1.270 1.273 4 0.57 +0.24 1.905 1.906 4 0.43 +0.05 2.540 2.520 4 0.69 -0.80 3. 085 3.085 4 0.00 0.00 2. Indicator: Zincon (Buffer: ethanolamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.635 0.645 4 2.31 +1.60 1.270 1.262 4 0.86 -0.63 1905 1.901 4 0.41 -0.20 2.540 2.518 4 1.02 -0.86 28 Table I (continued) 3. Indicator: Methylthymol Blue (Buffer: ethanolamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.588 0. 540 3 10.00 -8.16 2. 940 2.964 3 0.97 +0.80 3.085 3.087 3 0.16 +0.06 B. Solvent: Dimethyl formamide 1. Indicator: Eriochrome Black T (Buffer: diethylamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.709 0.723 1 +1.97 1.418 1.395 1 -1.62 2.127 2.094 1 -1.55 2.836 2.796 1 -1.41 29 Table I (continued) 2. Indicator: Zincon (Buffer: ethanolamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.709 0.753 1 +6.20 1.418 1.477 1 +4.16 2.127 2.109 1 -0.85 2.836 2.823 1 -0.46 C. Solvent: Dimethyl sulfoxide 1. Indicator: Eriochrome Black T (Buffer: ethanolamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.781 o. 779 3 0.32 -0.26 1.562 1.561 3 0.32 -0.06 2.343 2.338 3 0.37 -0.20 3.124 3.127 3 0.19 +0.96 3.905 3.913 3 0.28 +0. 20 30 Table I (continued) 2. Indica tor: Zincon (Buffer: ethanolamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.078 0.078 3 1. 74 0.00 0.156 0.157 3 0.86 +0.64 0.234 0.235 3 1. 60 +0.43 0.312 0.311 3 1.38 +0.32 0.391 0.380 3 3.54 -2.81 0. 781 0.784 3 1.12 +0.38 1.562 1.568 3 0.56 +0.38 2.343 2.336 3 0.39 -0.30 3.124 3.102 3 0.92 -0.70 3.905 3.865 3 1. 25 -1.02 D. Solvent: Methyl ethyl ketone Titrant: EDTA dissolved in DMSO +methyl ethyl ketone (1:3) Indicator: Eriochrome Black T (Buffer: tetrabutylammonium hydroxide + perchloric acid) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.891 0.926 1 +3.93 1. 782 1.819 1 +2.07 2.673 2. 729 1 +2.09 31 Table II. Results of Titrations of Zinc with DCTA Using Metallochromic Indicators A. Solvent: Methanol 1. Indicator: Eriochrome Black T (Buffer: ethanolamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.635 o. 634 3 1.03 -0.16 1.270 1.292 3 3.23 +1. 73 1.905 1.920 3 1.53 +0.80 2.540 2.525 3 1.62 -0.60 3.175 3.105 3 2. 74 -2.20 2. Indicator: Methylthymol Blue (Buffer: diethylamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.634 0.641 3 1.35 +1.10 1.268 1. 266 3 0.33 -0.16 1.902 1.902 3 0.61 0.00 2.536 2.512 3 1.16 -0.95 32 Table II (continued) 3. Indicator: Zincon (Buffer: ethanolamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviations 0.634 0.629 3 1.30 -0.79 1.268 1.255 3 1.45 -1.02 1.902 1.873 3 1.77 -1.52 2.538 2.509 3 1.26 -1.14 B. Solvent: Dimethyl sulfoxide Indicator: Zincon (Buffer: ethanolamine) .Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.078 0.078 2 o.oo 0.00 0.156 0.158 2 1.43 +1.28 0.234 0.235 2 0.85 +0.49 0.312 0.314 2 1.28 +0.64 0.391 0.391 2 0.25 0.00 33 Table II (continued) C. Solvent: Methyl ethyl ketone Titrant: DCTA in methanol Indicator: Methylthymol Blue (Buffer: diethylamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.065 0.064 3 1.08 -1.54 0.130 . 0.129 3 0.54 -0.77 0.195 0.191 3 0.36 -2.05 0.260 0.257 3 0.62 -1.15 34 Table III. Results of Titrations of Zinc with EGTA Using Metallochromic Indicators A. Solvent: Methanol Indicator: Zincon (Buffer: ethanolamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.635 0.639 4 1.23 +0.63 1.270 1.268 4 0.86 -0.16 1.905 1.914 4 0.92 +0.47 2.540 2.543 4 0.69 +0.12 B. Solvent: Dimethyl sulfoxide 1. Indicator: Eriochrome Black T (Buffer: ethanolamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.892 0.922 2 5.0 +3.36 1.784 1.844 2 5.0 +3.36 2. 676 2.783 2 5.7 +4.00 3.568 3.701 2 5.4 +3.72 4.460 4.646 2 5.9 +4.17 35 Table III (continued) 2. Indicator: Zinc on (Buffer ethanolamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.892 0.878 4 2.21 -1.57 1.784 1. 742 4 2.77 -2.35 2.676 2.592 4 3. 71 -3.14 3.568 3.559 4 1.50 -0.25 4.460 4.447 4 1. 72 -0.29 36 Table IV. Results of Titrations of Zinc with Tetren Using Matallochromic Indicators. A. Solvent: Dimethyl sulfoxide 1. Indicator: Zincon (Buffer: ethanolamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.112 0.115 1 +2.68 0.224 0.230 1 +2.68 0.335 0.345 1 +2.98 0.447 0.467 1 +4 .47 0.559 0.573 1 +2.50 1.118 1.127 2 1.44 +0.81 2.236 2. 241 2 1. 73 +0.22 3.354 3.368 2 1.52 +0.41 4.472 4.501 2 1.38 +0.65 5.590 5.595 2 0.20 +0.09 37 Table IV (continued) 2. Indicator: Zincon (tetraethylammonium perchlorate + ethanolamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.112 0.113 2 1. 78 +0.89 0.224 0.233 2 5.35 +4.01 0.335 0.342 2 2.82 +2.09 0.447 0.453 2 1.89 +1.34 0.559 0.568 2 2.15 +1.61 1.118 1.134 1 +1.43 2.236 2. 240 1 +0.18 3.354 3.347 1 -0.21 4.472 4.461 1 -0.25 5.590 5.574 1 -0.29 38 Table IV (continued) B. Solvent: Methyl ethyl ketone Titrant: Tetren in DMSO 1. Indicator: Zincon (sodium perchlorate+ triethanolamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.784 0.800 2 2.97 +2.04 1.568 1.562 2 1.46 +0.38 2.352 2.336 2 0.96 -0.68 3.136 3.113 2 1.05 -0.73 2. Indicator: Zincon (lithium perchlorate + triethanolamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.784 o. 777 2 1.45 -0.89 1.568 1.555 2 1.39 -0.83 2.352 2.328 2 1. 52 -1.02 3.136 3.109 2 1. 22 -0.86 39 formamide with Eriochrome Black T buffered by ethanolamine or triethanol amine, with Methylthymol Blue and ethanolamine, with Calcon and diethyl amine have failed. In methyl ethyl ketone, Zincon buffered with tri ethanolamine, Calcon with triethylamine, Eriochrome Black T with trieth anolamine or the mixture tetramethylammonium hydroxide + tetraethyl ammonium perchlorate did not give any result. In dioxane, which dissolves polystyrene, Eriochrome Black T buffered by ethanolamine or triethylamine did not give any quantitative result when EDTA dissolved in dimethyl sulfoxide was used as titrant. With DCTA as titrant (Table II), Methylthymol Blue appeared to be the best indicator, particularly in the solvent methyl ethyl ketone and diethylamine as buffer the figures suggest a possibility of determination of zinc in polystyrene. With EGTA {Table III), Zincon proved to be superior to Eriochrome Black T. Tetren dissolved in dimethyl sulfoxide could be used to titrate zinc in methyl: ethyl ketone. It was noticed that lithium perchlorate used to maintain the ionic strength at 0.1 was much better than sodium perchlorate since the color change was very rapid and sharp. 3. Titrations of calcium. Direct titrations of calcium have been less successful than the case of zinc. To the indicators used for zinc was also added Calcon. With EDTA (Table V), only Methylthymol Blue gave some results in methanol. Calcon did not behave quantitatively in all the solvents and buffers {even tetrabutylammonium hydroxide, one of the most basic buffers) tried. DCTA appeared to be more effective than EDTA since results were 40 Table V. Results of Titrations of Calcium with EDTA in Methanol Using Methylthymol Blue (Buffer: ethanolamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.397 0.404 3 2.36 +1.76 0. 794 0.806 3 2.04 +1.51 1.191 1.206 3 1.80 +1.26 1.588 1.608 3 1.68 +1.26 41 Table VI. Results of Titrations of Calcium with DCTA Using Metallochromic Indicators A. Solvent: Methanol 1. Indicator: Calcon {Buffer: diethylamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.842 0.850 4 1.55 +0.95 1.263 1.276 4 1.40 +1.02 1.684 1.695 4 1.12 +0.65 2.105 2.120 4 0.90 +0.71 2. Indicator: Methylthymol Blue (Buffer: diethylamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.421 0.427 4 1.90 +1.42 0.842 0.852 4 1.45 +1.18 1.263 1.280 4 1.58 +1.34 1.684 1. 707 4 1.67 +1.36 2.105 2.128 4 1.28 +1.09 42 Table VI (continued) B. Solvent: Methyl ethyl ketone Titrant: DCTA in methanol Indicator: Methylthymol Blue (Buffer: diethylamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.057 0.058 3 2.14 +1.75 0.114 0.115 3 1.52 +0.87 0.171 0.172 3 1.01 +0.58 0.228 0.228 3 0.70 0.00 0.285 0.285 3 0.43 0.00 0.421 0.429 3 2.85 +1.90 0.842 0.851 3 1.39 +1.06 1.263 1. 276 3 1.31 +1.02 1.684 1. 714 3 2.16 +1. 78 2.105 2.135 3 1. 74 +1.42 43 Table VII. Results of Titrations of Calcium with EGTA Using the System Zincon + Zinc-EGTA A. Solvent: Methanol Indicator: Zincon + Zn-EGTA (Buffer: ethanolamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.397 0.401 3 1. 22 +1.00 0. 794 0.805 3 1. 74 +1.38 1.191 1. 212 3 2.20 +1. 76 1.588 1.618 3 2.39 +1.88 B. Solvent: Dimethyl sulfoxide 1. Indicator: Zincon + Zn-EGTA (Buffer: ethanolamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.366 0.367 4 2.24 +0.27 0.040 0.044 1 +10.00 0.080 0.089 1 +11.25 0.120 0.134 1 +11.66 0.160 0.173 1 +8.13 44 Table VII. (continued) 2. Indicator: Eriochrome Black T + Zn-EGTA ~uffer: ethanolamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.040 0.042 1 +5.00 0.080 0.083 1 +3 0 75 0.120 0.124 1 +3.34 0.160 0.164 1 +2.50 3. Indicator: Eriochrome Black T + Zn-EGTA (+KC104) ~uffer: ethanolamine) Taken Found Number of % Relative Standard % Error mg. mg. Determinations Deviation 0.040 0. 040 1 0.00 0.080 0.080 1 0.00 0.120 0.120 1 0.00 45 obtained with Calcon and Methylthymol Blue in methanol and with Methyl thymol Blue in methyl ethyl ketone. This latter case suggested a pos sibility of determination of calcium in polystyrene with zinc. Calcon was once more inefficient in all the other media tried. With EGTA (Table VII.) none of the indicators used gave quantitative results but if trace quantities of zinc-EGTA complex were added then a substitution reaction took place and the titration was possible. This kind of reaction was investigated by Ringbom, Pensar and·Wanninen (31), who used the system Zincon + zinc-EGTA. In methanol only this latter reaction was studied, but in dimethyl sulfoxide the system Eriochrome Black T + zinc-EGTA was also tried and proved to be superior to the system with Zincon, in particular if the ionic strength was kept at 0.1 with potassium perchlorate, then quantitative results seemed to be attained. To sum up, in comparison with EDTA and EGTA, DCTA proved to be the best titrant for calcium. This result does not contradict the findings of Holloway and Reilley (49) when they studied the metal stability constants of various aminopolycarboxylate ligands in water with their mercury electrode. 4. Titrations of calcium+ zinc. Calcium + zinc have been determined in methyl ethyl ketone with DCTA dissolved in methanol and Methylthymol Blue as indicator in an anticipated try to use the method for the analysis of polystyrene. The sum of these two metals has also been determined with EGTA, but the addition of the complex zinc-EGTA was not necessary. EGTA titrant was much better than DCTA but unfortunately EGTA could not be used 46 Table VIII. Results of Titrations of Calcium + Zinc with DCTA Using Methylthymol Blue Solvent: Methyl ethyl ketone Titrant: DCTA in methanol (Buffer: diethylamine) Taken: J.l-mole Ca • ••••••••••••••••• 1.01 1.01 1.01 ll-mole Zn . ••.••.••.••.•.... 0.99 1.98 2.97 Total JJ-mole Ca+Zn taken ...... 2.00 2.99 3.98 Total ).!-mole Ca+Zn found ...... 2.00 3.00 4.01 Number of determinations •••••••.••• 4 4 4 % Relative standard deviation ••.... 1.04 1.43 1.65 % Error ...... 0.00 +0.33 +0.75 47 Table IX. Results of Titrations of Calcium + Zinc With EGTA Using Metallochromic Indicators A. Solvent: Dimethyl sulfoxide Indicator: Eriochrome Black T (Buffer: ethanolamine) Taken: ~-mole Ca •••••••••••••• 0.91 p-mole Zn ...... 1.71 Total ~-mole Ca + Zn taken ••••• 2.62 Total ~-mole Ca + Zn found ••••• 2.63 Number of determinations •••.••• 5 %Relative Standard Deviation •• 0.02 % Error ...... +0.38 B. Solvent: Methanol Indicator: Zincon (Buffer: ethanolamine) Taken: ~-mole Ca • ••••••••••••• 9.9 19.8 ~-mole Zn • ••••••••••••• 9.7 9.7 Total ~-mole Ca + Zn taken ••••• 19.6 29.5 30.0 Total ~-mole Ca + Zn found ••••• 19.9 Number of determinations ••••.•• 3 3 % Relative Standard deviation •• 0.16 0.21 % Error ...... •...... · · · · · · · • · +1.53 +1.69 48 with methyl ethyl ketone since an immediate precipitate formed when they were brought together. 5. Analysis of calcium and zinc in a sample of polystyrene. The polystyrene sample was dissolved by pouring it into the solution already stirred on the magnetic stirrer-hot plate; this prevented it from forming a big coagulate which sticked to the stirring bar and stopped its motion. The dissolution was complete within ten minutes and the titra tion performed with the hot solution required less than three minutes thus the total analysis time was less than 15 minutes. Two samples were analyzed; one with an assay labled Zn = 100-115 ppm, Ca = 5 ppm, the other Ca = 100 ppm, Zn ~ 1 ppm,(?), phosphate (?). The sample with out phosphate gave after analysis of zinc a concentration of calcium equal to 18 ppm. The sample with phosphate gave a very low result for calcium and the color of the complex was not well developed. This interference due to the phosphate has been reported by Garcia Montelongo, Herrera and Arias (50) who studied the colorimetric determination of calcium with Methylthymol Blue in aqueous medium. It appeared that this interference was also valid in methyl ethyl ketone. 6. Conclusion Direct visual titrations of zinc and calcium with metallochromic indicators in non aqueous media are possible, although more limited than in aqeuous solution. Eriochrome Black T and Zincon gave good re sults for the analysis of zinc, Methylthymol Blue for the analysis of calcium. This latter indicator has been extensively studied in aqueous medium by Korbl and Pribil (38), Buben, Korbl and Pribil (51), Korbl and Kakac (52), Janousek and Studlar (53), Chalmers and Miller (54) and 49 Table X. Results of Titration of Zinc and Calcium in Polystyrene Origin of sample: Dow expandable polystyrene (Bendix Co.) Product: SD - 572 Lot No.: OH - 1701 Assay: Zinc = 100 to 115 ppm Calcium 5 ppm A. Titration of Zinc Solvent methyl ethyl ketone Titrant 0.001 M tetren in dtmethyl sulfoxide Indicator Zincon Buffer Triethanolamine Weight of sample Weight of zinc found 1.000 g 109 ppm 112 106 98 114 106 Average • •••••..•••• • • • • • • • • • • • • • • 108 ppm Standard deviation...... 6 %Relative standard deviation.... 6 Confidence ltmit at 90 % Probability level •••••••••••••••• + S Result: 108 ± 5 ppm Zinc 50 Table X. (continued) B. Microtitration of zinc Same experimental conditions as in A except titrant is 0.0001 M and weight of sample is 0.100 g. Weight of zinc found 100 ppm Average ...... •...... 103 ppm 110 Standard deviation. . • • • • • • • • • • • • 10 110 %Relative standard deviation •••• 10 90 Confidence limit at 90 % 100 probability level. • • • • • • • • • • • • •. + 8 110 Result: 103 ± 8 ppm Zinc C. Titration of calcium+ zinc. Solvent: methyl ethyl ketone Titrant: 0.001 M DCTA in methanol Indicator: Methylthymol Blue Buffer: diethylamine Weight of DCTA Zinc * Calcium Calcium sample }l-mole p-mole J..l-mole ppm 1.000' g 2.10 1.65 0.45 18 2.08 0.43 17 2.16 0.51 20 2.07 0.42 17 2.12 0.47 19 2.12 0.47 19 *Value taken from A Average for calcium •••• 18 ppm % Relative standard deviation ...... 7 Standard deviation ••••• 1 Confidence limit at 90% probability level •.••••• ± 1 Result: 18 + 1 ppm Calcium 51 Yoshino et al. (55). Since it shows two ranges of pH color changes it could be used to titrate zinc buffered by urotropin at pH = 6.5 and the sum zinc + calcium buffered by annnonia. Its behavior in "acidic" medium in methyl ethyl ketone for the analysis of zinc should be an interesting investigation. B. Photometric Titrations 1. The absorbance curves in Figure 1 show a maximum difference be tween the indicator and the metal-indicator complex at 650 nm. Due to the fact that the total volume of titrant added represented less than 8 % of the total volume of the solution, a dilution factor was not necessary in constructing the titration curve. 2. The titration curves in dimethyl sulfoxide as solvent are shown on Figure 2. DCTA appears to be the best titrant for zinc: the curve drops sharply and the relative error is +0.84 %. No extrapolation is needed to locate the end point which lies on the abscissa axis. For the case of EGTA, EDTA and tetren, the end point has to be extra polated by the interesection of two straight lines. Concerning their effectiveness the titrants can be ordered in the following decreasing order according to Table XI: DCTA > Tetren > EDTA > EGTA This compares favorably with the direct visual titration which gives the following order: DCTA > EDTA > Tetren > EGTA 52 Figure 1. Absorbance curves in dimethyl sulfoxide -6 -4 (A) 6.0 x 10 M Zincon, 4.78 x 10 M Zinc perchlorate 0.06 M ethanolamine in dimethyl sulfoxide -6 (B) 6.0 x 10 M Zincon, 0.06 M ethanolamine in dimethyl sulfoxide 53 0 0 U) E c: 0 .. I() :I: &() ..._ <.!) z w ...J . 0 w r-l Q) 0 .... &() ;j ~ 00 ;: ..-I ~ &() "'"v 0 &()v &() ~------~------~------~------~----~N0 0 0 0 v v ro N 0 0 0 0 3:>Ntt8~0S8tt 54 Figure 2. Photometric titrations of zinc in dimethyl sulfoxide Wavelength: 650 nm; Indicator: Zincon (A) Titrations of 1.19 ~-mole of zinc perchlorate in DMSO Titrant: 0.00103 M DCTA in DMSO Zincon: 1.44 x 10-5 M in DMSO (B) Titration of 1.19 ~-mole of zinc perchlorate in DMSO Titrant: 0.00102 M EGTA in DMSO Zincon: 1.78 x 10 -S Min DMSO (C) Titration of 1.79 v-mole of zinc perchlorate in DMSO Titrant: 0.00104 M EDTA in DMSO Zincon: 1.78 x 10-5 M in DMSO (D) Titration of 1.71 ~-mole of zinc perchlorate in DMSO Titrant: 0.00102 M tetren in DMSO Zincon: 1.78 x 10-S Min DMSO 55 0.30 r---T1 ----.,.1----,1---,--1 ---. 0.20 ~ - A 0.10- - I I I I lJJu 0.20 z ~ CD a: 0.10 0 (/) CD 0.30 0.20 D 0.10 0 0.5 1.0 1.5 2.0 TITRANT ADDED (MICROMOLES) Figure 2. 56 Figure 3. Photometric titration of zinc in dimethyl formamide Wavelength: 625 nm; Indicator: Zincon Titration of 10.85 ~mole of zinc perchlorate in dimethyl formamide Titrant: 0.0070 ~ EDTA in dimethyl formamide -5 Zincon: 3.7 x 10 Min dimethyl formamide 57 . 0.20~------ L&J 0.15 uz <( m a:: 0 0.10 (/) m <( 0.05 0 I 2 3 4 5 6 7 8 9 10 II 12 13 TITRANT ADDED (MICROMOLES) Figure 3. 58 Table XI. Results of Photometric Titrations of Zinc A. Solvent: Dimethyl sulfoxide 1. Indicator: Zincon (Buffer: ethanolamine) Zinc Taken Zinc Found Number of % Error Titrant ll-mole ll-mole Determination 1.19 1.20 1 +0.84 DCTA 1.19 1.23 1 +3.36 EGTA 1. 79 1.84 1 +2.79 EDTA 1. 71 1.68 1 -1.75 Tetren B. Solvent: Dimethyl formamide 1. Indicator: Zincon (Buffer: ethanolamine) Zinc Taken Zinc Found Number of % Error Titrant v-mole ll-mole Determinations 10.85 10.95 1 +0.92 EDTA 59 In fact the relative errors in the two methods are of the same order of magnitude and the photometric titration does not give better results in comparison with the direct visual titration. One further disadvan tage of the photometric method is the fact that it is time consuming and the end point must be determined on a graph. A direct visual titration takes about five minutes while a photometric titration requires twenty to thirty minutes. The photometric method will only be useful in the case where impurities in the solution might block the indicator or when a recording automatic phototitrator is available. 3. In dimethyl formamide the titration curve of zinc at 625 nm in Figure 3. shows a sharp end point. The relative error of less than 1 % is much better than the direct visual titration which has a error of 6 %. Due to the lack of time no investigation have been made in methanol as solvent. 4. The absorbance curves run for different indicators Calcon, Methyl- thymol Blue, Eriochrome Black T and Zinconc and their complexes with calcium or zinc showed that Zincon (Figure 1), gave two best resolved curves in which the maximum absorbance of one curve occured at a wave length corresponding to the minimum absorbance of the other, and vice versa. For this reason, only some photometric titrations of zinc with Zincon as indicator have been studied. Photometric titrations being an extension of the direct visual method of titration should of course give positive results for cases which work in the latter method. C. Potentiometric Titrations 1. Table XII and Figure 4 show the results of potentiometric titrations of calcium and zinc with EDTA in dimethyl sulfoxide and in Table XII. Results of Potentiometric Titrations of Calcium and Zinc with EDTA Metal Taken Found % Error Electrode Solvent Amplitude of Curve in ll-rnole ll-mole Potential Break Figure 4. rnV Hg 2.50 2.55 +2.00 Mercury DMSO 25 A Zn 13.64 13.90 +1.90 Mercury DMSO 75 B Ca 9.14 9.35 +2.30 Mercury DMSO 25 c Ca 18.28 17.65 -2.90 Mercury DMSO 60 Not Drawn Ca 13.44 13.80 +2.68 Mercury Methanol 90 D Ca 33.60 33.50 -0.30 Mercury Methanol 96 Not Drawn Zn 13.64 13.37 -1.98 Lead DMSO 55 E Zn 27.28 27.37 +0.33 Lead DMSO 130 Not Drawn 0\ 0 61 Figure 4. Potentiometric titrations of calcium and zinc with EDTA (A) Titration of 2.5 ~-moles of mercuric acetate in DMSO Titrant: 0.01046 M EDTA in DMSO Mercury electrode vs. modified S.C.E. (B) Titration of 13.64 ~-moles of zinc perchlorate in DMSO Titrant: 0.01046 M EDTA in DMSO Mercury electrode vs. modified S.C.E. (C) Titration of 9.14 ~-moles of calcium perchlorate in DMSO Titrant: 0.01055 M EDTA in DMSO Mercury electrode vs. modified S.C.E. (D) Titration of 13.44 ~-moles of calcium perchlorate in DMSO Titrant: 0.0100 M EDTA in methanol Mercury electrode vs. modified S.C.E. (E) Titration of 13.64 ~-moles of zinc perchlorate in DMSO Titrant: 0.01055 M EDTA in DMSO Lead ion selective electrode vs. modified S.C.E. 62 000 . IJJ. 0 . C/) ~ -3.00 IJ.. 0- 0 :E . -400 tn > > E .. -500 ...J -<( 1-z IJJ -600 1- 0 a.. E -700 0 5 10 15 20 TITRANT ADDED (MICROMOLES) Figure 4. 63 methanol. a) Volume determinations. Except for curve A and D where no volume correction is needed since in A mercury was titrated directly with the mercury electrode and in D 2 drops of Hg-EDTA indicator solution were used, all the volumes read on the end point of other curves must be corrected for the indicator blank. These blank volumes are respectively 2.5 ~-moles for curve B, 0.05 ml x 0.001 M = 0.00005 mM = 0.05 ~-mole for curve C, and 0.05 ml x 0.00125 M = 0.000625 mM = 0.63 ~-mole for curve E, since one drop of the indicator has a volume of 0.05 ml. b) Comments on the results. The amplitude of the potential break will be defined by the dif ference in potential between the two points on the curve, one on each side of the end point, separated by an increment of one micromole. The titration of zinc in dimethyl sulfoxide with the mercury elec trode gives a sharper potential break than with the lead electrode (75 mV vs. 55 mV). The volume of indicator ion used was respectively 2.5 ~-moles of mercuric ions and 0.63 ~-mole of lead (II) ions. Figure 6 shows that when the lead electrode is used the amplitude of the poten tial break increases with an increase of the concentration of zinc. The experiments with the mercury electrode show the opposite. It has not been possible to determine calcium with the lead elec trode. When the mercury electrode was used the volume of indicator was respectively 2 x 0.05 ml x 0.020 M = 0.002 mM = 2 ~-moles of Hg (II) ions in methanol and 0.05 ~-mole of Hg (II) ions in dimethyl sulfoxide (see part a) above). In methanol the mercuric ions were introduced 64 in the form of an aqueous solution. The presence of these 2 drops of water in 45 ml of solution gives a water concentration of 0.22 % and might explain why the amplitude of the potential break is larger in methanol than in dimethyl sulfoxide (90 mV vs. 25 mV) (curves C and D). Other experiments performed at different concentrations of calcium show that in methanol the amplitude of the potential break increases slightly with an increase of concentration of calcium while in dimethyl sulfoxide this trend is more accentuated. Figure 1 indicates also that when the mercury electrode is used, the titration curve for zinc appears in a higher potential region than the curve for calcium. This result in dimethyl sulfoxide is also found in aqueous medium (56). A comparison between curves B and C shows that in the titrations of zinc and calcium with the mercury electrode the potential break is larger for zinc than for calcium (75 mV vs. 25 mV). This leads to the assumption that in dimethyl sulfoxide, as in water, the stability constant of the zinc-EDTA complex is larger than the stability of the calcium-EDTA com plex. 2. Table XIII and Figure 5 give the results of the potentiometric titration of lead in dimethyl sulfoxide with EDTA and EGTA using the lead ion selective electrode. Since the lead ion is determined with the lead electrode no volume correction is necessary. EDTA appears to be a better titrant for lead than EGTA. For almost the same quantity of lead the potential break with EDTA is 140 mV, with 65 Table XIII. Results of Potentiometric Titrations of Lead with EDTA and EGTA Lead ion selective electrode vs. modified S.C.E. Solvent: Dimethyl sulfoxide Taken Found % Error Titrant Amplitude of Curve in 11-mole 11-mole Potential Break Figure 5 mV 12.46 12.66 +1.60 EDTA 140 A 24.92 25.32 +1.60 EDTA 100 B 12.83 12.73 -0.78 EGTA 44 c 25.66 25.76 +0.40 EGTA 26 D 66 Figure 5. Potentiometric titrations of lead with EDTA and EGTA Lead ion selective electrode vs. modified S.C.E. (A) Titration of 12.5 u-moles of lead perchlorate in DMSO Titrant: 0.01055 M EDTA in DMSO (B) Titration of 25.0 umoles of lead perchlorate in DMSO Titrant: 0.01055 M EDTA in DMSO (C) Titration of 12.8 u-moles of lead perchlorate in DMSO Titrant: 0.0100 ~ EGTA in DMSO (D) Titration of 25.6 u-moles of lead perchlorate in DMSO Titrant: 0.0100 M EGTA in DMSO 67 0 (/) ~ - UJ__. 0 :E It) 0 (\J a:: u :E- - . 0 1.() (\J 0 UJ al ~ 0 &, 0 ...... 0 It),.._ I ·3·~·5 031-:IIOOW "St\ AW '1\111N3l0d 68 Table XIV. Results of Potentiometric Titrations of Zinc with EDTA, DCTA and EGTA. Lead ion selective electrode vs. modified S.C.E. Solvent: Dimethyl sulfoxide Taken Found % Error Titrant Amplitude of Curve in Jl-mole Jl-mole Potential Break Figure 6 mV 13.64 13.37 -1.98 EDTA 55 A 27.28 27.37 +0.33 EDTA 130 B 13.64 13.35 -2.12 DCTA 120 c 27.28 26.99 -1.06 DCTA 90 not drawn 17.10 17.80 +4 .09 EGTA 60 D 34.20 25.80 +4.67 EGTA 35 E 69 Figure 6. Potentiometric Titrations of Zinc with EDTA, DCTA and EGTA Lead ion selective electrode vs. modified S.C.E. Solvent: Dimethyl sulfoxide (A) Titration of 13.64 ~-moles of zinc perchlorate in DMSO Titrant: 0.01055 M EDTA in DMSO (B) Titration of 27.28 u-moles of zinc perchlorate in DMSO Titrant: 0.01055 M EDTA in DMSO (C) Titration of 13.64 ~-moles of zinc perchlorate in DMSO Titrant: 0.00845 M DCTA in DMSO (D) Titration of 17.10 ~-moles of zinc perchlorate in DMSO Titrant: 0.0101 M EGTA in DMSO (E) Titration of 34.20u-moles of zinc perchlorate in DMSO Titrant: 0.0101 M EGTA in DMSO 70 0 CD !'() -(/) LLJ It) _J (\J 0 :E 0 a: 0 u (\J - ~ - . 0 \0 LLJ Q) ~ It) 0 ;::l bO 0 •.-l It) 0 0 0 0 0 0 0 0 0 0 0 0 10 U) .-... (X) en ~ I I I I I I ·3·:> ·s 031:JIOOW ·sA J\W 'l"llN3lOd 71 EGTA it is only 44 mV. This might tell that the stability of the lead EDTA complex in dimethyl sulfoxide, as in water, is greater than the stability of the lead-EGTA complex. Here again the potential break decreases noticeably with an increase in concentration of lead ions. This phenomenon might be explained by the fact that at higher lead ion concentrations other reactions, possibly ion association, compete with the complexation reaction. 3. Figure 6 and Table XIV give the results of potentiometric deter minations of zinc with the lead ion selective electrode in dimethyl sulfoxide using different titrants. All the volumes read on the curves must be corrected for the blank volume due to the drop of lead perchlorate added before the titration: the quantity of lead introduced as indicator was respectively 0.63, 0.06 and 0.05 ~-mole in the case of EDTA, DCTA and EGTA used as titrant. EDTA in contrast with DCTA and EGTA gives a sharper break with an increase in concentration of zinc. The three titrants can be listed in the following decreasing effectiveness in their reaction with zinc: EDTA > DCTA > EGTA This result differs slightly with the one obtained in the photometric titration (with zinc solutions ten times more dilute): DCTA > EDTA > EGTA One possible explanation of this result may be the fact that the mole cules of titrant are more complex from EDTA ·to DCTA and EGTA. This complexity of the molecule is expected to give a slower chelation 72 reaction with zinc ions. In the photometric method the solution was stirred for 30 seconds after each addition of titrant and let stand for 15 seconds before any absorbance reading, thus the complex formation has more time to be effective in contrast with the potentiometric method where the potentials were read continuously with the addition of titrant. According to the manufacturer (57) the following solvents attack the epoxy used to seal the sensiting crystal into the body of the elec trode: acetonitrile, methyl ethyl ketone, dimethyl formamide and methyl isobutyl ketone. For this reason no attempts have been made to use the lead electrode in these media. The following potentiometric titrations have been attempted but did not give any quantitative results: Potentiometry with the mer~ury·electrode: - Titration of zinc with EDTA in m~thanol and ethanolamine as buffer. The potential at the mercury electrode was not stable. - Titration of zinc with DCTA in methanol and ethanolamine as buffer: The potential break was too small. - Titration of zinc with EDTA in a solvent mixture methanol +methyl ethyl ketone (1:1), and ethanolamine as buffer: No potential break. - Titration of mercury with EDTA, both dissolved in dimethyl sulfoxide, in methyl ethyl ketone with a mixture of tetramethyl ammonium hydroxide and tetraethylammonium perchlorate as buffer: The potential at the mercury electrode was not stable. - Titration of mercury with DCTA in dimethyl formamide: Erratic poten- tials. Potentiometry with the lead ion selective electrode: 73 - Titration of zinc with EDTA in methanol: No potential break. - Titration of calcium with EDTA in dimethyl sulfoxide with ethanolamine as buffer: No potential break. - Same titration of calcium but use tetrabutylammonium hydroxide as basic medium: No potential break. - Titration of calcium with EGTA in dimethyl sulfoxide with ethanolamine as buffer: Potential not stable. Potentiometry with the silver electrode. The silver electrode has been extensively used by Strafelda (58-60) and by Fritz and Garralda (61) in aqueous medium buffered with borax. This substance was tried in the titration of calcium with DCTA in di methyl formamide but did not dissolve in the solvent. Zinc and calcium could not be determined with EDTA in methanol buffered with ethanolamine. One explanation might be that the buffer complexes all the silver ions added to serve as indicator. To avoid the blocking of these indicator ions it might be interesting to investi gate the behavior of the ion selective electrodes of the first kind in organic media, particularly the calcium ion electrode since it has been successfully: used in many cases in aqueous solution (62-63). D. Amperometric Titrations Since polarography in organic solvents, particularly in dimethyl formamide and dimethyl sulfoxide, has been extensively studied within the last decade (44, 64-76). It was thought that amperometric titrations in these two media could be performed as in alcohol (23). All the at tempts have been unsuccessful despite the fact that almost all the 74 combinations of electrode systems have been tried: mercury pool vs. DME, S.C.E. vs. DME, Rotating platinum electrode vs. S.C.E. Polarographic curves for zinc in particular have been recorded but the amperometric titration curves did not give quantitative results. 75 V. CONCLUSION Some complexometric titrations in nonaqueous media have been studied. The principal difficulty encountered was the problem of dissolving the chelating agents in these solvents. Polyamine carboxylic acids dissolved in alcohols with the presence of ethanolamine, in dimethyl sulfoxide and dimethyl formamide upon heating and without any addition of other sub stances. Their solutions in alcohols and dimethyl sulfoxide were very stable, in dimethyl formamide they reprecipitated after a period of several hours. Direct visual titrations of calcium and zinc have been investigated in the solvents methanol, dimethyl formamide, dimethyl sulfoxide and methyl ethyl ketone. The metallochromic indicators which were used appeared to work well in these media: Eriochrome Black T, Methylthymol Blue and Zincon proved to be satisfactory for the determination of zinc, Calcon, Methylthymol Blue, the systems Zincon + zinc-EGTA and Eriochrome Black T + zinc-EGTA for the determination of calcium. Photometric titrations of zinc were also investigated with Zincon as indicator. Microgram quantities of zinc could be determined by this method. Potentiometric analysis with the mercury and lead electrodes gave some results, particularly for the determination of zinc. The use of the lead ion selective electrode in nonaqueous media was restricted by the nature of the solvent which must be chosen to be inert to the resin which composed the body of the electrode. Amperometric titrations have been investigated without success. 76 A practical application of these experiments was found in the analysis of trace quantities of zinc and calcium in a sample of poly styrene dissolved in methyl ethyl ketone. Provided no phosphate is present in the sample the sum calcium + zinc can be determined with DCTA dissolved in methanol and Methylthymol Blue as indicator. Zinc is determined separately with tetren, dissolved in dimethyl sulfoxide, and Zincon as indicator. A large number of experiments which did not work do not prove though that they are impossible under other experimental conditions. This research has been performed in a one year period, therefore it does not pretend to be exhaustive. The following topics of investigation would bring more light to the field of complexometry in nonaqueous media: research on the problem of dissolving the chelating agents, the determin ation of metal-chelate stability constants, a comparative study of metal chelates positively, negatively or not charged, a systematic study on buffer compositions in nonaqueous media, research on stepwise titrations by the use of masking agents, a wide investigation to devise ion selec tive electrodes with materials compatible with organic solvents. If this last problem could be solved, complexometry in nonaqueous media would be an extremely powerful, versatile and simple method in the hands of the analytical chemist. 77 REFERENCES 1. Schwarzenbach, G. and H. Flaschka, "Die Komplexometrische Titrations," Enke, Stuttgart, 1965. 2. Pribil, R., "Komplexone in der chemischen Analyse," VEB Deutscher Verlag der Wissenschaften, Berlin, 1961. 3. Ringbom, A., "Complexation in analytical chemistry," Wiley-Inter science, New York, 1963. 4. Sillen, L. G. and A. E. Martell, "Stability Constants of Metal-IOn Complexes," Chemical Society, London, 1964. 5. Flaschka, H. and A. J. Barnard, Jr., "Chelates in Analytical Chemis try," Marcel Dekker, New York, 1972. 6. Reilley, C. N., A. J. Barnard, Jr. and R. Puschel, "Complexometric Titrations," in "Handbook of Analytical Chemistry," L. Meites ed., MCGraw Hill Book Company, New York, p. 3-76 (1963). 7. Schwarzenbach, G. and H. Ackerman, Helv. Chim. Acta, 31, 1029 (1948). 8. Gotschalk, G., Z. Anal. Chern., 172, 192 (1960). 9. Siggia, s., Anal. Chem., ~, 1745 (1955). 10. Rechnitz, G. A. and N. C. Kenny, Anal. Letters, 2, 259 (1970). 11. Gerhardt, P. B. and E. R. Hartmann, Anal. Chern., 29, 1223 (1957). 12. Fischer, w. J., J. Inst. Petrol., 48, 290 (1962). 13. Crump, G. B., Analyst, 94, 465 (1969). 14. Hooks, R. w. and J. w. Noar, Analyst, 94, 473 (1969). 15. Crump, G. B. and J. w. Noar, J. Inst. Petrol, 56, 180 (1970). 16. Hypta, s. Chemia analit. (Poland). 15, 285 (1970). 17. Kiss, T. A., z. Anal. Chem., 208, 334 (1965). 78 References (continued) 18. Kiss, T. A., F. F. Gaa1, T. M. Suranyi and I. J. Zsigrai, Anal. Chirn. Acta, 43, 340 (1968). 19. Kiss, T. A. and T. M. Suranyi, Mikrochirn. Acta, 1968, 1102. 20. Kiss, T. A. and o. I. Knezevic, Microchem. J., 13' 459 (1968). 21. Casassas, E. and T. Fernandez, An. R. Soc. esp Fis. Quirn, !_, 66 45 (1970). 22. Yoshimura, C. and K. Tamura, J. Chem. Soc. Japan, pure Chern. Sect., 87, 625 (1966). 23. Arthur, P. and B. R. Hunt, Anal. Chern., 39, 95 (1967). 24. Fernandez, T., E. Casassas, F. Garcia MOntelongo, An. R. Soc. esp. Fis. Quim, B, 64, 315 (1968). 25. Fernandez, T. and E. Casassas, An. R. Soc. esp. Fis. quim, !_, 66, 61 (1970). 26. Rechnitz, G. A. and N. C. Kenny, Anal. Letters, 1, 395 (1969). 27. Kiss, T. A., I. J. Zsigrai, V. C. Cirin and R. L. Krizsan, Micro chern. J., 14, 573 (1969). 28. Kiss, T. A., Z. Anal. Chern.., 252,12 (1970). 29. Hume, D. N. and W. E. Harris, Ind. Eng. Chern., Anal. Ed., _!2, 465 (1943). 30. Kolthoff, I. M. and J. F. Coetzee, J. Am. Chern. Soc., 22, 870 (1957). 31. Ringbom, A., G. Pensar and E. Wanninen, Anal. Chim. Acta, 19, 525 (1958). 32. Reilley, C. N. and J. H. Holloway, J. Am. Chern. Soc., 80, 2917 (1958). 33. Van Meurs, N. and E.A.M.F. Dahmen, Anal. Chirn. Acta,~' 193 (1959). Brown G. H. and R. Al-Urfali, J. Am. Chern. Soc., 80, 2113 (1958). 34. ' - 79 References (continued) 35. Reilley, C. N. and R. W. Schmid, Anal. Chern., 31, 887 (1959). 36. Erley, D. S., Anal. Chern., 29, 1564 (1957). 37. Starke, K., Inorg. Nucl. Chem., 11, 77 (1959). 38. Korbl, J. and R. Pribil, Colln. Czech. Chern. Commun., ~' 873 (1958). 39. Starostin, V. V., Zh. Anal. Khim., 16, 620 (1961). 40. Huber, W., "Titrations in Nonaqueous Solvents," Academic Press, New York, pp. 84, 81 (1967). 41. Rush, R. M. and J. H. Yoe, Anal. Chern., 26, 1345 (1954). 42. Rorabacher, D. B., W. J. MacKellar, R. R. Shu and Sister M. Bonavita, Anal. Chern., 43,561 (1971). 43. McClure, J. E. and J. B. Reddy, Anal. Chem., 40, 2065 (1968). 44. Reddy, J. B., Ph.D. Thesis, University of Minnesota (1961). Univer sity Microfilms, L. C. Card No. Mic. 60.6140. 45. Bykova, L. N. and S. I. Petrov, Zh. Anal. Khim., 25, 5 (1970). 46. Kreshkov, A. P., Talanta, 17, 1029 (1970). 47. Petrov, S. I. and L. N. Bykova, Zh. Anal. Khim., 25, 1686 (1970). 48. Bates, R. G., "Determination of pH, Theory and Practice," John Wiley, New York, p. 226 (1964). 49. Holloway, J. H. and C. N. Reilley, Anal. Chem., 11, 249 (1960). 50. Garcia Montelongo, F., C. Herrera and J. J. Arias, Infcion. Quim. analit. pura. apl. Ind., ~ 26 (1971). 51. Buben F., J. Korbl and R. Pribil, Colln. Czech. Chem. Commun. Q, 881 (1958). 52. Korbl, J. and B. Kakac, Colln. Czech. Chern. Commun., 23, 889 (1958). 53. Janousek, I. and K. Studlar, Colln. Czech. Chern. Commun., ~. 3799 (1959). 54. Chalmers, R. A. and F. I. Miller, Analyst, 96, 97 (1971). 55. Yoshino, R., H. Imada, T. Kuwano and K. Iwasa, Talanta, 16, 151 (1969). 80 References (continued) 56. Reilley, C. N. and R. W. Schmid, Anal. Chern., 30, 947 (1958). 57. Orion Research Inc., Cambridge, Massassuchetts, Technical Services, Personal communication. 58. Strafelda, F., Colln. Czech. Chern. Commun., 27, 343 (1962) 59. Strafelda, F., Colln. Czech. Chern. Commun., 28, 3345 (1963). 60. Strafelda, F., Colln. Czech. Chern. Commun., 30, 2320 (1965). 61. Fritz, J. S. and B. B. Garralda, Anal. Chem., 36, 737 (1964). 62. Durst, R. A., "Ion-selective electrodes, R. A. Durst, Ed., US Nat. Bur. Stand., Spec. Publ. 314." p. 394. (1969). 63. Orion Research Inc., Bibliography on Orion specific ion electrodes. Form Bib/1712, December 1 (1971). 64. Gutmann, V. and G. Schoeber, Z. Anal. Chern., 171, 339 (1959). 65. Schoeber, G. and V. Gutmann, Advan. Polarog., Proc. Intern. Congr.: 2nd, Cambridge, Eng., 1959, l, 940 (1960). 66. Kalthoff, I. M. and T. B. Reddy, J. Electrochem. Soc. 108, 980 (1961). 67 .. Schaap, W. B. and D. L. McMasters, J. Am. Chern. Soc., 83, 4699 (1961). 68. Burrus, R. T., Ph.D. Thesis, University of Tennessee (1962), Univer sity Microfilms, L. C. Card No. Mic. 63.4101. 69. Le Guillanton, G., Bull. Soc. Chim. France, 2359 (1963). 70. McMasters, D. L., R. B. Dunlap, J. R. Kuempel, L. W. Kreider and T. R. Shearer, Anal. Chem., 39, 103 (1967). 71. Butler, J. N., J. Electroanal. Chern., 14 89 (1967). 72. Headridge, J. B. , M. Ashraf and H. L. H. Dodds, J. Electroanal. Chern., 16, 114 (1968). 73. Tsuji, K. and P. J. Elving, Anal. Chem., 41, 216 (1969). 74. Coetzee, F. F., J. M. Simon and R. J. Bertozzi, Anal. Chern., 41. 766 (1969). 81 References (continued) 75. Reddy, T. B., Pure Appl. Chem., ~, 459 (1971). 76. Nicholson, R. S., Anal. Chern., 44, 478R (1972). 82 VITA Ngo The Hung was born on November 18, 1940 in Hanoi, Vietnam. He received his primary, elementary and almost all his secondary education in Paris, France. After finishing high school at the lycee Jean-Jacques Rousseau in Saigon he studied physics and chemistry at the faculty of science of the university of Saigon. He obtained the licence in Physical Sciences in 1965. Since 1965 he has been a member of the faculty at the National Technical Center in Saigon, the only university level engineer ing college in his country. In August 1970 his government sent him under the auspices of the Agency for International Development to the United States for advanced studies in analytical chemistry at the University of Missouri - Rolla