The Coordination Chemistry and Electrochemistry of Chromium (III) Complexes of Thiobis (Ethylenenitrilo) Tetraacetic Acid Citation Peerce, Pamela J . (1978) The Coordination Chemistry and Electrochemistry of Chromium (III) Complexes of Thiobis (Ethylenenitrilo) Tetraacetic Acid. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/8ceh-nc51. https://resolver.caltech.edu/CaltechTHESIS:02072026-103724305 Abstract The product of the reaction between Cr(Cl0 4 ) 3 and the ligand, thiobis(ethylenenitrilo)tetraacetic acid (TEDTA), was very dependent on the pH of the reaction mixture. Two monomeric isomers and several dimeric species were identified. Characterization of the various Cr(III) complexes of TEDTA included an examination of their pH behavior, visible and ultraviolet absorption spectra, substitutional !ability with excess azide ion, behavior on a Selectacel anion exchange column, reaction with heavy metal cations and methyl mercury cation, infra-red spectra and distinctive electrochemistry. Comparisons of the behavior of the Cr(III)-TEDTA complexes with those of the Cr(III) complexes of related ligands, including EDTA, oxybis(ethylenenitrilo)tetraacetic acid (EEDTA) and pentamethylenedinitrilotetraacetic acid (PMDTA), were very helpful. Based on these data, it was concluded that both cis and trans isomers (with respect to the nitrogens) were formed. In the trans isomer, the sulfur is forced to be close to the Cr(III) center and coordination occurs. At pH 5-9, solutions of the pentadentate cis isomer are a mixture of the monomeric complex and an acetate bridged dimer. Confirmation of dimer formation and a value for the dimerization constant were obtained from potential step experiments. The structure of the dimer was strongly supported by the cyclic voltammetry which indicated the presence of two nonequivalent metal centers in the complex. The other dimeric species were highly charged and their IR spectra suggested that they were oxo-bridged. The electrochemistry of both the cis and trans isomers and the acetate bridged dimer was investigated in detail. Cyclic voltammetry and electrocapillary measurements indicated that trans-CrTEDTA was much more strongly adsorbed on mercury electrodes than was cis-CrTEDTA(OH 2 ), cis-CrTEDTA(OH 2 ) or the acetate bridged dimer, and that free TEDTA was not adsorbed. Experiments with CrEEDTA showed that it was not adsorbed, indicating that the Cr(III)-TEDTA complexes were attached to the electrode through the sulfur. Quantitative measurements of the adsorption confirmed all of this. The adsorption of trans-CrTEDTA increased abruptly over a very narrow range of bulk concentrations and was strongest at positively charged electrodes. The potential dependence was corroborated by the fact that the charge on the electrode in the presence of the complex was more positive. The concentration and potential dependences are attributed to back bonding from Cr(III) to the thioether sulfur which a-donates electron density to the electrode. The abrupt change in the surface coverage over a narrow charge of bulk concentration indicates that the adsorption of trans-CrTEDTA is cooperative. Chronocoulometric measurements of the adsorption of the cis isomer at pH 4 and 7 demonstrated that the neutral complex was more strongly adsorbed than the anions especially at more negative potentials. The adsorption appeared to increase as the electrode potential was made more negative. However, the electrocapillary curves were discrepant. Electrode charge measurements supported the electrocapillary data. The adsorption of the cis complex depressed the double layer capacitance at all potentials. This behavior is similar to that of organic adsorbates. Although the non-adsorption of the free ligand was attributed to steric difficulty 1n exposing the sulfur atom to the electrode, the fact that the La(III) and Zn(II) complexes of TEDTA were also not adsorbed shows that the role of Cr(III) in inducing the adsorption of TEDTA in the cis isomers is complicated. Item Type: Thesis (Dissertation (Ph.D.)) Subject Keywords: (Chemistry) Degree Grantor: California Institute of Technology Division: Chemistry and Chemical Engineering Major Option: Chemistry Thesis Availability: Public (worldwide access) Research Advisor(s): Anson, Fred C. (advisor) Gray, Harry B. (advisor) Thesis Committee: Unknown, Unknown Defense Date: 22 May 1978 Additional Information: Thesis is missing pg. 26. Funders: Funding Agency Grant Number NSF UNSPECIFIED California Institute of Technology UNSPECIFIED Record Number: CaltechTHESIS:02072026-103724305 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:02072026-103724305 DOI: 10.7907/8ceh-nc51 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 17871 Collection: CaltechTHESIS Deposited By: Benjamin Perez Deposited On: 12 Feb 2026 22:18 Last Modified: 12 Feb 2026 22:53 Thesis Files PDF - Final Version See Usage Policy. 32MB Repository Staff Only: item control page

THE COORDINATION CHEMISTRY AND ELECTROCHEMISTRY OF CHROMIUM(III) COMPLEXES OF THIOBIS(ETHYLENENITRILO)TETRAACETIC ACID Thesis by PAMELA J. PEERCE In Partial Fulfillment of the Requirements for the Degree of DOCTOR OF PHILOSOPHY California Institute of Technology Pasadena, California 1978 (Submitted May 22, 1978) ii To my Dad iii ACKNOWLEDGMENTS I would sincerely like to thank Fred Anson for being easily accessible and always willing and interested in discussing any aspect of this work. His patient instruction has contributed greatly to my education and his many suggestions have been helpful without being stifling. His confidence, encouragement and dependable enthusiasm were greatly appreciated. I am also sincerely grateful to Harry Gray for his support and for allowing me the freedom to pursue this work. I am very grateful to John Turner for his willing cooperation in maintaining and repairing the chronocoulometer. Roger Baar's enthusiastic and competent assistance last summer 1s also acknowledged. Special thanks goes to Beth Cooper for her expert typing of this thesis and especially for her patience with my continual revisions. I acknowledge the financial support of the National Science Foundation and the California Institute of Technology. iv ABSTRACT The product of the reaction between Cr(Cl0 4 ) 3 and the ligand, thiobis(ethylenenitrilo)tetraacetic acid (TEDTA), was very dependent on the pH of the reaction mixture. Two monomeric isomers and several dimeric species were identified. Characterization of the various Cr(III) complexes of TEDTA included an examination of their pH behavior, visible and ultraviolet absorption spect r a, substitutional !ability with excess azide ion, behavior on a Selectacel anion exchange column, reaction with heavy metal cations and methyl mercury cation, infra-red suectra and distinctive electrochemistry. Comparisons of th~ behavior of the Cr(III)-TEDTA complexes with those of the Cr(III) complexes of related ligands, including EDTA, oxybis(ethylenenitrilo)tetraacetic acid (EEDTA) and pentamethylenedinitrilotetraacetic acid (PMDTA), were very helpful. Based on these data, it was concluded that both cis and trans isomers (with respect to the nitrogens) were formed. In the trans isomer, the sulfur is forced to be close to the Cr(III) c e nter and coordination occurs. At pH 5-9, solutions of the pentadentate cis isomer are a mixture of the monomeric complex and an acetate bridged dimer. Confirmation of dimer formation and a value for the dimerization constant were obtained from potential step experiments. The structure V of the dimer was strongly supported by the cyclic voltammetry which indicated the presence of two nonequivalent metal centers in the complex. The other dimeric species were highly charged and th e ir IR spectra suggested that they were oxo-bridged. The el;ctrochemistry of both the cis and tra n s isomers and the acetate bridged dimer was investigated in detail. Cyclic voltammetry and electrocapillary measurements indicated that trans-CrTEDTA was much more strongly adsorbed on mercury electrodes than was cisCrTEDTA(OH 2 ), cis-CrTEDTA(OH 2 ) or the acetate bridged dimer,and that free TEDTA was not adsorbed. Experiments with CrEEDTA showed that it was not adsorbed, indicatin g that the Cr(III) -TEDTA complex e s were attached to the electrode through the sulfur. Quantitative measurements of the adsorption confirmed all of this. of trans-CrTEDTA The adsorption increased abruptly over a very narrow range of bulk concentrations and was strongest at positively charged electrodes. The potential dependence was corroborated by the fact that the charge on the electrode in the presence of the complex was more positive. The concentration and potential dependences are attributed to back bonding from Cr(III) to the thioether sulfur which a-donates electron density to the electrode. vi The abrupt change in the surface coverage over a narrow charge of bulk concentration indicates that the adsorption of trans-CrTEDTA is cooperative. Chronocoulometric measurements of the adsorption of the cis isomer at pH 4 and 7 demonstrated that the neutral complex was more strongly adsorbed than the anions especially at more negative potentials. The adsorption appeared to increase as the electrode potential was made more negative. However, the electrocapillary curves were discrepant. Electrode charge measurements supported the electrocapillary data. The adsorption of the c i s complex depressed the double layer capacitance at all potentials. This behavior is similar to that of organic adsorbates. Although the non-adsorption of the free ligand was attributed to steric difficulty 1n exposing the sulfur atom to the electrode, the fact that the La(III) and Zn(II) complexes of TEDTA were also not adsorbed shows that the role of Cr(III) in inducing the adsorption of TEDTA in the ci s isomers is complicated. vii TABLE OF CONTENTS Page Chapter I . 1 Chapter II .96 Chap!er I II 142 Proposition 1 157 Proposition 2 164 Proposition 3 169 Proposition 4 177 Proposition 5 183 CHAPTER I COORDINATION CHEMISTRY OF CHROMIUM(III) WITH THIOBIS(ETHYLENENITRILO)TETRAACETIC ACID (TEDTA) 2 INTRODUCTION As part of a continuing investigation of the chemical factors which induce certain classes of complex ions to attach themselves to the surfaces of electrodes by spontaneous adsorption (1-5), we have sought ligands which can induce the adsorption of the widest possible variety of metal cations. Halide and pseudohalide anions are effective in inducing the adsorption of many d 10 cations (6-9), but large excesses of free ligand are usually necessary to stabilize these rather weak, labile complexes. As a result, analysis of the surface coordination chemistry involved is complicated by competition between these anions and their metal complexes for adsorption sites on the electrode surface . • Thiocyanate anion induces the adsorption of certain of the transition metal cations (5), but the extent of adsorption is strongly dependent on the d electronic configuration of the transition metal-isothiocyanate complex and competitive ligand adsorption remains a problem with all substitutionally labile complexes. An attractive solution to this problem is to choose a ligand of high denticity and affinity for metal cations so that large ligand excesses are not required 3 to achieve essentially complete complexation. EDTA(ethylene- dinitrilotetraacetic acid) is a good example of such a ligand, but with the exception of cobalt (10), EDTA complexes of most metal cations show little or no tendency to adsorb on the surface of mercury electrodes. In seeking a molecule with chelating capabilities similar to those of EDTA but with a much greater tendency towards adsorption on mercury surfaces, we investigated the sulfur-containing analog of EDTA, thiobis(ethylenenitrilo) tetraacetic acid (TEDTA), (l,7-diaza-4-thiaheptane-1,1,7,7,-tetraacetic acid), S[CH 2 CH 2 N(CH 2 COOH) 2 ] . Many metal complexes of this ligand exhibit a propensity to adsorb on mercury surfaces. This chapter concentrates on the unusual coordination chemistry displayed by chromium(III) complexes of TEDTA. 4 EXPERIMENTAL APPARATUS pH Titrations were performed using a Beckman Model 76 Digital pH Meter. Single wavelength absorbance measurements were done on a Beckman DU Spectrophotometer. UV-Visible spectra were recorded on a Cary Model 17 or 118 Spectrophotometer. Elemental analyses were performed by the Caltech microanalytical laboratory unless otherwise specified. Cyclic voltammograms were obtained using either a conventional multipurpose electrochemical instrument based on operational amplifiers which was constructed in this laboratory or a Princeton Applied Research Model 173 Potentiostat/Galvanostat, Model 1 75 Universal Programmer and Model 179 Digital Coulometer. Voltammograms were recorded on a Tektronix Model 564 CRO. All measurements were made at 25±2° C. MATERIALS Reagent grade chemicals were used without further purification. Methyl mercuric hydroxide was obtained as .1 1.27!:!aqueous solution from Alfa. All solutions were prepared with triply distilled water. Stock solutions of sodium perchlorate were prepared by neutralizing 60 % HClOi+ with sod i urn carbonate (11) . Stock so 1 u t ions _of chrorni um (I I I) perchlorate were prepared by reducing Cr0 3 with excess 5 H2O2 in perchloric acid and boiling to remove unreacted H2O2 . These stock solutions were made 0.5 to 1 Min HClO4 to minimize the formation of chromium(III) polymers. The extent of a polymer formation was monitored spectrophotometrically following the method of Altman and King (12). Standardization of Cr(III) solutions involved oxidation to chromate and comparison of the absorbance at 375 nm with a standard potassium chromate solution (13). SYNTHESIS OF THIOBIS(ETHYLENENITRILO)TETRAACETIC ACID, (TEDTA) The synthetic procedure of Smolin and coworkers (14) was followed with the following modifications and precautions: Preparation of Thiobis(ethylamine). The reaction between H2S and ethyleneimine is extremely exothermic and considerable caution was exercised during the introduction of H2S. The rate of addition of H2S was such that the temperature of the reaction mixture could be easily maintained at 17-20° C with an ice-water bath. If the pot temperature driftea above 25°, the reaction became autocatalytic. Control of the reaction rate was regained by discontinuing the addition of H2S and introducing Nz. 6 To facilitate monitoring of the rate of consumption of H2S, a U-tube filled with mineral oil was connected to the top of the chilled water (5°) condensor fitted to the reaction vessel. As the addition of H2S was judged complete when the weight of the reaction vessel I had increased by the theoretical amount, provision was made to isolate and close off the pot (via the conventional ·use of stopcocks) before removing it to the scale. The yield observed was always 60-70 % in excellent agreement with that reported (14). Preparation of the Tetraacid. The yield of the tetraacid was strongly dependent on the rate of addition of base and chloroacetic acid. The yield was 40% greater when the addition rate was such that maintenance of the temperature at 40-50° did not require external heating than when gentle heating was necessary. Although the reaction is somewhat exothermic, an external source of heat was required to maintain the pot temperature at the recommended 40° overnight. Water was added when necessary to dissolve the reagents. Sulfuric acid was used to acidify the solution, instead of the recommended HCl, to take advantage of the lower solubility of NaHSO4. Acidification lead to a somewhat viscous solution of high ionic strength 7 from which the desired product did not readily crystallize. The following procedure ~roved necessary to remove the excess salt present. The acidified reaction mixture was allowed to stand overnight at room temperature in a large crystallizing dish. The crystals of NaHSO 4 which formed were removed and discarded (losses of the tetraacid arising from coprecipitation with NaHSO 4 were minimal). The filtrat e was diluted four-fold with distilled water, the pH adjusted to ca . 2 by addition of H2SO 4 if necessary, and the overnight evaporation repeated. The solid film which formed on the bottom of the crystallizing dish was broken up, the solution agitated and the yellow, and now rather viscous, supernatant was diluted again threeto four-fold with water. After refrigeration for ca. 2 hours, crystalline product amounting to 60% of the total yield was obtained. The dilution-evaporation procedure was repeated twice. The product was filtered, washed several times with cold water and dried at 110° for 24 hours. Total overall yield: analysis(%): C (40.65); H (5.48); N (7.75). for C1 2H20OeN2S: 33 %. Elemental Calculated C (40.90); ti (5.72); N (7.95). The purity of the unrecrystallized material was verified by acidimetric titration of a sample with standard sodium hydroxide. The observed equivalent 8 weight was 177.1 compared with the theoretical value of 176.2. PREPARATION OF PENTAMETHYLENEDINITRILOTETRAACETIC ACID, (PMDTA) This ligand was synthesized from 1,5-diaminopentane (Aldrich) and chloroacetic acid using essentially the same procedure described above for TEDTA. Purification of the crude product by decolorization and multiple recrystallizations from both water and dilute HCl yielded colorless, well-formed crystals. Unfortunately, this improvement in the appearance of the product was not accompanied by a corresponding improvement in the analytical data (Table 1). Based on an average of the carbon and nitrogen analyses of a sample from treatment 4, the calculated purity of the product was 74.8%. The observed equivalent weight of a sample from treatment 4 was 117.5 which compared to the theoretical value of 167.2, indicated that this sample was 70.3 % pure. Titration of a sample of PMDTA from treatment 4 with standard Zn+ 2 solution in the presence of ammonia buffer and EriochrDme Black T yielded results consistent with the elemental analyses and equivalent weight determination. The impurity was some hydrogen- containing non-titrateable, non-complexin& i~rganic PMDTA ~fter 1st treatment above 3. for C1 3 lh 2N2O8 : %C ( 4 6 . 70 ) ; %N ( 8 . 38 ) ; %H 13.2:2.0:25.7 12.9:2.1:25.7 5.70 5.70 b 6.01 6.65b ( 6 . 6 3) . Recrystallized from 0.05 M 34.90 HCl, washed 2X with 0.05 ~ b HCl, recrystallized again 34.34 from0.05 M HCl, washed several times with ice water . and then ov en dried at 105° for 5 days. 13.1:2.0:25.9 5.76 6.19 Decolorized, recrystallized 34.76 from 0.1 M HCl, washed crystals ZX with I-1 2O, dried overniQht in vacuum dessicator, then dried overnight at 78° over CaC1 2 . 1 3 . 1 : 2 . 0 : 25 . 8 C:N:H 13.2:2.0:26.1 5.71 %H 5. 79 6.30 %N 6.02 34.44 %C 34.96 Decolorized, 2X recrystallized from H2O, washed crystals 3X with Il 2O and then dried under vacuum for 3 days. Washed 6-7X with H2O, dried in dessicator at <10 mm Hg for 2 days and then in oven at 110° for 2 days. Treatment bElemental analysis by Spang Microanal y tical Laborator y , Eagle Harbor, MI. a Ca 1 cu 1 at e d Decolorized, recrystallized, PMDTA from treat/ . ments 2 and 3 a bove PMDTA after 1st treatment above 2. 4. Crude Product 1. Sample Elemental Analysis Data for PMDTAa Table 1 <.O 10 material which did not appear to interfere with the preparation of Cr(III) complexes from the impure ligand. PREPARATION OF CHROMIUM(III)-CHELATE COMPLEXES CHROMIUM(III) COMPLEXES OF TEDTA Two isomeric complexes, I and II, were prepared: Isomer I. Dilute (~25 m~) stock solutions of this isomer were prepared by adding a stoichiometric amount of TEDTA to a solution of Cr(ClO4) 3 at pH ~1.80 and heating to ca. 90° for 1 1/2 hours. The completeness of the complexation reaction was verified by raising the pH of an aliquot of th e reaction mixture above 5. A blue precipitate (chromic hydroxide) resulted until complexation was complete. Solutions of Isomer I were stored at pH 1.5 without detectable decomposition for 30 days. Prolonged storage at pH values below 1 resulted in decomposition of the complex, which was reflected by a reduction in the intensity of the violet color of these solutions. Isomer II. Stock solutivns (10 m~) of this isomer were prepared from solutions of Isomer I by raising the pH to 4-4.5 and maintaining it within this range while the solution was boiled for an additional 3 hours. The 11 resulting red-violet solution was cooled to room temperature and the pH increased to 6.8-7.0. The visible spectrum of this solutio~ which was nearly identical to that of a pure solution of Isomer II, is shown in Figure 1. Complete conversion to Isomer II occurred after a few weeks at room temperature and pH 6.87.0. Alternatively, a solution of pure Isomer II can be obtained immediately by the anion exchange procedure described below. Solutions of Isomer II were stored at pH 6.8-7.0 and appeared to be stable indefinitely. Complex III. Stock solutions (10 m~) of this complex were prepared by heating an aqueous mixture containing equimolar quantities of Cr(C1O 4 ) 3 and TEDTA at an initial pH of 3.2 for an hour at 90°. Reaction of TEDTA with Cr(NO 3 ) 3 -xH 2 O resulted in the formation of Complex III even if the pH were initially set to 1.5. The reddish-violet solution could be stored at pH 2 for ca. a week. CHROMIUM(III) COMPLEXES OF OXYBIS(ETHYLENENITRILOTETRAACETIC ACID (EEDTA) Cr(ClO4) 3 and EEDTA (Ciba-Geigy: from a ten year old stock in this laboratory; the material is apparently no longer commercially available) were mixed in stoichio- 12 FIGURE 1 Visible Spectra of Solutions of Isomer II. Curve A is the spectrum of the Isomer II reaction mixture after it had cooled to room temperature and the pH had been raised to 6.89. Curve Bis the spectrum of a pure solution of Isomer II at pH 6.9 (11.3 m~), which had been equilibrating for more than 6 weeks. The spectra were superimposed at the low energy maximum. 13 ~ 0 0 0) 0 0 14 metric quantities in water (pH 1.8). heated for 1 1/2 hours at 90° C. The mixture was The resulting blue- green solution turned violet on cooling and was stored at pH 4 where it appeared to be stable indefinitely. CHROMIUM(III) COMPLEXES OF PMDTA Two isomers were prepared, one violet and one peach 1n color. Violet Isomer. A stock solution of the violet isomer ~as obtained by heating a small excess of the ligand with Cr(ClO4)3 at pH 1.8 for ca. one hour. (The ligand purity was taken into account.) Peach Isomer. A solution of the violet isomer was adjusted to and kept at pH 5-6 and boiled for 2 hours. The solution became red in color. This red solution was divided -- one half was maintained at room temperature and pH 6.4 and the other half was heated at pH 6.4-6.6 for an additional 20 hours. The color of the solution gradually changed to peach. This solution appeared indefinitely stable at room temperature and pH 6.5-6.6. The red solution at pH 6.4 di~ not change visibly in over 4 months at room temperature. 15 ION EXCHANGE RESINS Selectacel DEAE anion exchange resin was purified according to reference 15 and converted to the perchlorate form by stirring with 1 M HClO4. A column (2.2 x 10 cm) of the resin was rinsed thorou ghly with distilled water before use. AG 1 X-8 and AG SOW X-8 resins were purified by overnight oxidation with a strongly alkaline solution of H2O 2 , followed by several washes with 1 M HCl and distilled water. The AG 1 X-8 (2.2 x 13 cm) and AG SOW X-8 (2.2 x 16 cm) columns were thoroughly rinsed with distilled water before use. PROCEDURES FOR pH TITRATIONS The substitutional inertness of chromium(III) complexes made it necessary to heat mixtures of each ligand and Cr(OH 2 ) 6 3 + until complexation equilibrium was complete before commencing each titration. For the titrations of Isomer I and Cr(III)-EEDTA the procedure was as follows: A stoichiometric quantity of the solid ligand in the tetraacid form was added to a 10 mM solution of Cr(ClO 4 ) 3 which also contained a slight excess of perchloric acid. The mixture was heated for ca. 60-90 minutes at 90° while maintaining its volume by addition of distilled water. The 16 resulting homogeneous solution (pH 1.5-1.7) was cooled to room temperature. For Isomer I, completeness of conversion was verified by spectral examin a tion. The solution was then titrated with standard 0.010 M NaOH. The concentration of perchloric acid present in the stock solution of Cr(ClO4)3 was determined by a similar titration following the conversion to the CrEDTA complex,which is known to produce four moles of protons per mole . of Cr(III) for titration to the first end point (16). For Isomer II, a slightly different procedure was followed: After preparing a 10 mM solution of Isomer I as described above, the pH of the solution was raised to 4-4.5 by addition of a measured volume of standard 0.01 M NaOH solution. The mixture was heated to 95° C and the pH was maintained at 4-4.5 by continual quantitative additions of NaOH. After three hours, the solution was cooled to room temperature and titrated with standard NaOH. The visible spectrum of this solution at pH 7 was nearly identical to that of pure Isomer II ( cf . prep. of Isomer II). The da t a from pH 3.8 to 2.8 were obtained by back titration with standard 0.01 N HCl. ANALYSIS OF TH E SOLID ADDUCT OF MERCURIC ION AND ISOMER II A 34% excess (over 1:1) of standard 50 m~ Hg (ClO4)2 (pH 1.4) was added to a 10 mM solution of Isomer II; the 17 resulting mixture was centrifuged and samples of the colorless supernatant were carefully withdrawn for Hg+ 2 analysis. The mercury content of these samples was determined by addition of excess EDTA and back titration of the unreacted EDTA with a standard solution of Zn +2 . The precipitate was collected and dried under vacuum for the chromium analysis. A weighed sample of the pink solid was dissolved in water by addition of excess EDTA. This solution was made alkaline and the chromium content assessed by oxidation to chromate as previously described. 18 RESULTS PREPARATION, PURIFICATION AND ATTEMPTED ISOLATION The product of the reaction between Cr(ClO 4 ) 3 and TEDTA depended on the pH of the reaction mixture as summarized in Scheme 1. When the pH was maintained below 1.8 during heating, a violet complex (Isomer I) was obtained which turned blue-green when the pH was adjusted to pH 7 at room temperature. If the initial pH of the reaction mixture was in the vicinity of 3, heating produced a solution which was redder than is characteristic of Isomer I. These reddish-violet solutions (Complex III) turned gray rather than blue-green at pH 7-8 and appeared to contain dimeric Cr(III) complexes which slowly converted to pure Isomer I at pH 1 - 1.5. If Cr(NO 3 ) 3 was used, Complex III was obtained even at pH 1.5. Isomer I was converted to Isomer II by adjusting the pH to 4-4.5 and heating to 95° for several hours. Figure 2 shows the spectral changes that occurred during this transformation. Isomer II spontaneously reverted to Isomer I overnight at pH <l. Raising the pH of a hot solution of Isomer I above 5 and heating at pH 5-6 led again to the dimers,which were not readily converted to Isomer II. in Figure 3. The observed spectral changes are shown / go° C ~ » TEDTA pH)3 + ~ "\, Form pH <s l1 Gray pH>7 Comnlex III Reddish-violet Form Cr(Cl04)3 <-,·, 0 C, e., i}- "v<-, goo 25° pH>7 25° pH<4.5 Isomer I (violet) go 0 25° ~1 day pH 1, pH 4-4.5; Blue-green Complex pH 1.5-1.8 SCHEME 1 Isomer II (burnt pink) I-' \.0 20 FIGURE 2 Spectral Changes during the Preparation of Isomer II. Curves 1-8 are the spectra of aliquots removed from a reaction mixture Cr(III)-TEDTA ea. 10 mM in at the following intervals: Curve 1, prior to raising the pH to 4-4.5 (pH 1.29); Curve 2, after 1/2 hour (pH 5.03); Curve 3, after 1 hour (pH 4.14); Curve 4, after 1 3/4 hours (pH 4.38); Curve 5, after 3 hour~ (pH 4.08); Curve 6, after 2 1/2 hours at room temperature, before heating was resumed; Curve 7, after 4 hours total heating (pH 4.24); Curve 8, after raising pH of aliquot 7 to 6.99. ,...., N J> • 0 (J); CP 01 (.J) • 0 ..p;. ;o z CP J> rrJ 0 400 500 WAVELENGTH/NM 600 22 FIGURE 3 Sn e ctral Changes during the Preparation of Complex III. Curves 1-4 are the spectra of aliquots removed from a reaction mixture c a . 10 m~ in Cr(III)-TEDTA at the following intervals: Curve 1, prior to raising the pH to 5-6 (pH 1.58); Curve 2, immediately after raising the pH of the hot reaction mixtur e (pH 5.30); Curve 3, after 1 hour (pH 5.12); Curve 4, after 5 hours total heating (pH 5.40); Curve 5, aliquot 4 after 3 days at room temperature and pH 6.94. 23 :E t fTl r fTl z G) -{ I '-(Jl zo so 24 Isomers I and II and Complex III were readily separated on a Selectacel DEAE anion exchange column in the ClO 4 form. Isomer I exhibited no affinity for this resin and was rapidly eluted with either 0.01 N NaC1O 4 or 0.01 N HClO4. By contrast, Isomer II and the gray form of Complex III were tightly bound in a narrow band which was eluted with 1 M NaClO 4 ; Isomer II moved somewhat faster than gray Complex III and, thus, they could be se~arated. Attempts to isolate Isomer I as the acid or lithium salt and Isomer II as the potassium or sodium salt by preparation of concentrated solutions (0.1-0.5 ~) from which their precipitation could be eff e cted by the addition of ethanol were unsuccessful. Both powders obtained for Isomer I ,gave poor elemental analyses. The violet powder obtained as the acid form of Isomer I was insoluble in water and dissolved only slowly in hot, 0.2 M HC1O 4 . Both preps for Isomer II yielded burgundy colored solutions which gradually blackened at pH >9. One solution of the potassium salt (0.1 ~), which was allowed to reflux 30 hours at pH 7, became extremely viscous and gelled on cooling to room temperature. All of these materials were probably polym e ric. Ricard (17) has reported that the formation of Cr(III) polymers of long chain (n - 3,4,6) diamino- 25 alkyltetracarboxylic acids was favored by prolonged heating of concentrated solutions (~0.1 ~) at pH >5. Hydrolytic cleavage of these polymers was effected by heating at 50° in 0.05 ~ HC1O 4 • Hamm (16) reported that all efforts to isolate the potassium salt of either the violet or blue forms of CrEDTA failed and that extremely viscous, oily materials always resulted. As we lusted for the crystal structure of Iso mer II, we persisted in trying to coax its crystallization. Dilute (10 m~) stock solutions of Isomer II were allowed to concentrate by evaporation at pH 7 at room temperature, but clear gels were always obtained. Pink powders could be obtained from these transparent gels by washing them successively with ethanol and ether; however, these were always contaminated with NaC1O 4 and possibly some excess TEDTA. Attempts to purify Isomer II by recrystallization as either the sodium, ammonium, tetramethylammonium or tetraethylammonium salts from either water or a mixed water-ethanol system by evaporation at room temperature or by slow, controlled cooling were unsuccessful. Typically, opaque, semi-fluid gels were obtained in which were suspended small, dark partic}es. In one experiment with the tetraethylammonium salt, a nearly transparent, viscous fluid containing solid particles was obtained after standing overnight at 4° C. 27 N(CH3)4Br can be removed by sublimation at 0.1 mmHg at 150° (19). We have succeeded in separating NH 4Br from a solid mixture of HCrEDTA and NH 4Br at 0.1 mmHg and 60°. The lower temperature is preferred in order to minimize complications from the thermal matrix reaction of the chromium complex and the halide salt (19,20). Ogino, Chung and Tanaka reported the preparation of solid, monomeric complexes of Cr(III)-diaminoalkyltetraacetate complexes by using chromous ion to drive the complexation reaction without heating (21). This procedure was followed for Cr(III)-TEDTA. After the addition of Cr(II), the reaction mixture gradually turned deep blue and a blue solid precipitated from solution. No change was observed when the vessel was opened to the air. On standing overnight, the solution turned violet (probably aquation of the blue, chloro complex). The precipitated solid, which was filtered off and washed with ethanol and ether, was pale bluishviolet in color when dry and insoluble in water. vaping led Roto- to a deep blue, glassy solid which was also insoluble in water. CHARACTERIZATION OF Cr(III)~TEDTA COMPLEXES Characterization of the Cr(III) complexes of TEDTA included an examination of their pH behavior, visible 28 and UV absorption spectra, substitutional lability with excess N3 - , behavior on Selectacel anion exchange resin, reactivity with heavy metal ions and methyl mercury cation, limited IR evidence and distinctive electrochemistry. For purposes of comparison, the Cr(III) complexes of two other ligands related to TEDTA were also studied: oxybis(ethylenenitrilo)tetraacetic acid (EEDTA), (more properly, l,7-diaza-4-oxaheptane-1,1,7,7,-tetraacetic acid) and pentamethylenedinitrilotetraacetic acid (PMDTA), (l,5-diazapentane-1,1,5,5-tetraacetic acid), in which the thioether group in TEDTA is replaced by an ether linkage and a methylene group, respectively. pH BEHAVIOR pH Titrations. The two isomers of CrTEDTA were subjected to acid-base titrations (as described in the Experimental Section) in order to examine the metal coordination and ligand denticity. Because the reaction of the ligand with substitutionally inert Cr(III) was driven largely to completion by heating before the pH titrations were conducted, the titration curves do not reflect the equilibrium pH dependence of the Cr(III)ligand reaction. With Isomer II, the single inflection in the titration curve (Figure 4a, Curve 3) near pH 7 corresponded 29 FIGURE 4 pH Titration Curves. except as otherwise noted. All solutions were 10 mM Part a: Curve 1, free TEDTA ( ca . 3 m~); Curve 2, Isomer I; Curve 3, Isomer II (5 m~). Cr(III)-EEDTA. Part b: Cr(III)-EDTA. Part c: 30 a 10 pH 6 2 4 MOLES BASE / MOLE CR 6 b 10 pH 6 4 6 MOLES BASE/ MOLE CR C 10 pH 6 2k------2 MOLES BASE/MOLE CR 4 31 to the consumption of Cr(III), 3.9 the reaction. of moles of Therefore, groups are dinated carboxyl 4.1 coordinated group moles which either of base per were added durin g all to Cr(III) has a four or mole carboxylate any uncoor- pK too low to be a distinguishable from the titration of a strong acid. During the titration, the originally pink solution of Isomer II changed color only slightly, becoming somewhat more orange at pH 10. However, on standing overnight at a pH ~11, the solution took on the green color characteristic of Isomer I at this pH. Solutions of Isomer I produced titration curves with two less well defined inflections (Figure 4a, Curve 2). The positions of these inflections corresponded reasonably well to the points where 3 and 4 moles ~f base per mole of Cr(III) had been consumed. values were 3.1 and 4.1, respectively.) (The precise The first three protons titrated as strong acids, but the fourth proton titrated over a somewhat wider pH range and appeared to correspond to a weak acid with a pK a near 5.5. The color of the solution of Isomer I varied extensively during the titration: The solution retained its original violet color up to ca. pH 4.3 but became progressively bluer at higher pH values and was green 32 above pH 11. The blue-green solution present at pH 7 returned to its original violet color only slowly when the pH of the solution was adjusted quickly to 3,whereas the green solution at pH 11 immediately turn e ' violet. In the Cr(III)-EDTA complex only three of the four carboxylate groups are coordinated to Cr(III); the sixth coordination position is occupied by a water molecule (16, 22-25). This complex yielded a titration curve (Figure 4b) containing two sharp inflections which corresponded to 4 and 5 moles of base added per mole of Cr(III). The solution remained violet, turning blue only after sufficient base had been added to substantially convert the aquo complex to the hydroxy complex (16,23-25); the observedpK of the coordinated water molecule was a 7.50 in excellent agreement with the literature value of 7.52 (16). At pH ~12, the solution was green. This green color has been ascribed to the formation of the dihydroxy complex (23). Titration of a solution of the Cr(III)-EEDTA complex (Figure 4c) resulted in a simple titration curve with a single inflection corresponding to the titration of 4 moles of protons per mole of chromium(III). The solution was violet at pH values below 3.5 and blue-green above pH 4.5. The blue-green form of the complex regained its violet color only slowly at room temperature when the pH was decreased to 2.5. 33 Although a potentiometric pH titration of Complex III was not done, its color as a function of pH was noted. It was reddish-violet up to pH 5.6, gray at pH 7-11 and turned green on standing at pH >12. As with CrTEDTA, two geometrical isomers of CrPMDTA appear to exist. A violet isomer formed at low pH, underwent color changes identical to those exhibited by Isomer I as the pH was increased. A peach-colored isomer formed at higher pH values behaved similarly to Isomer II, exhibiting only minor variations in color between pH 2 and 11. pH Hysteresis Experiments. In order to further investigate the hysteresis phenomena noted above, a series of experiments was performed to determine whether there was any difference in the amount of base required to increase the pH from pH 2, 3 or 4 to 8 compared to the amount of acid required to restore the pH to its original value. The results are summarized in Table 2. These data demonstrate that 30 to 35% of the acid groups titrated were not immediately available for reprotonation when the pH was returned to 3· or 4, respectively. However, when the solution was strongly acidified (pH ~2), these groups were released and could be reprotonated . Similar experiments with Isomer II and Complex III revealed no hysteresis in their pH behavior . ' 34 Table 2 pH Change mmoles of Titrant Consumed Quantit_l of Base Used Quantity of Acid Used 2.00-+ 8.10 8.10-+ 2.00 0.879 0.902 1. 03 3.06-+ 8.10 8 .10 -+ 3.07 0.556 0.390 1. 43 4.16-+ 8.10 8.10-+ 4 .12 0.472 0.307 1. 54 1. 98 -+ 8.17 8.'27-+ 1. 99 0.116 0.124 0.94 3.04-+ 8.17 8.27-+ 3.03 0.0267 0.0260 1. 03 3. 56 -+ 9.17 8.27-+ 3.53 0.0156 0.0167 0.93 Complex III 3 2. 02 -+ 8. 0 7 8.07-+ 2.02 0.450 0.492 0.91 3.12 -+ 8.07 8.07-+ 3.10 0.254 0.269 0.94 Isomer 11 Isomer 11 2 1 Sample: 25.00 ml of a solution 10.02 mM 1n complex. - 2 Sample: 5.00 ml of a solution 10.44 mM 1n complex 3.00 ml triply distilled water. 3 Sample: 20.00 ml of a solution 9.87 mM in complex. + 35 Equilibrium Distribution of Cr(III)-TEDTA Species as a Function of pH. In order to gain a greater under- standing of the relationship between the various Cr(III)TEDTA complexes, a series of experiments was performed to determine their equilibrium distribution as a function of pH. Three sets of five jars were each filled with 15 ml of a 10 mM solution of either Isomer I, predominantly Isomer II or predominantly Complex III and the pH was set to 1.5, 3.0, 5.0, 7.0 or 9.0. These solutions were allowed to equilibrate at room temperature over a period of four months. The pH was checked periodically and readjusted when necessary. It was assumed that equilibrium was attained when no changes were observed in the pH, color and visible spectrum of the solution for at least a week. A summary of the data is presented in Table 3. Spectra for each series of solutions are shown in Figures Sa, band c for Isomer I, Isomer II and Complex III, respectively. The percentage of Isomers I and II present at / equilibrium at pH 3 and 5 was calculated from the data in both Figures Sa and Sb using the observed absorbances at 558 and 512 nm. The resul~s of these calculations are presented in Table 4. The equilibrium distribution of Isomers I and II as a function of pH from pH 1,5-7 is shown in Figure 6. 1.5 3.0 5. 0 7. 0 9.0 1.5 3.0 5. 0 7. 0 9.0 Isomer II Complex III 2. 3. Original Color Violet Violet Purple Blue Grayish-Green Reddish-Violet Reddish-Violet Reddish-Purple Grayish-Red - Violet Gray Red-Violet Reddish-Violet Reddish-Violet Grayish-Purple Grayish-Purple Gray-Purple Observed tipHa None Decreased Decreased Decreased Decreased None C Increased Increased ~C onstant Decreased None C Increased Increased Decreased Decreased Violet Red-Violet Pink-Violet Burnt Pink Gray-Pink Violet Red-Violet Pink-Violet Burnt Pink Gray-Pink Violet Red-Violet Pink-Violet Burnt Pink Gray-Purple Final Color 2 weeks 8 weeks 7 weeks 15 weeks 13 weeks 1 week 4 weeks 4 weeks 4 weeks 11 weeks 0 4 weeks 13 weeks 16 weeks 6 weeks Equilibration Timeb 372 nm (£37.0) d 372 nm ( £3 5 . l)e 513 nm (£38.5)e 366 nm (£29 . 9)d Isosbestic Points e Intersection of spectra for solutions set at pH 1 . 5, 3.0 and 5.0. d Intersection of spectra for solutions set at pH 1. 5, 3. 0, 5. 0 and 7. 0. c At such high aciditi es ,the pH electrode error was comparable to any variation observed in the solution pH. b Time required to reach equilibrium. a Observed change in the pH as the solution equilibrated. 1.5 3.0 5.0 7. 0 9.0 Isomer I 12H 1. Predominant Starting srecies Summary of Data for the Cr(III)-T EDTA pH Eq uilibrium Experiments Table 3 vi 0-, 37 FIGURE 5 Sp ectra of solutions originally ca . 10 mM in: Isom e r I, Part a; Isomer II, Part b; Complex III Part c, which had been allowed to come to equilibrium at room temperature over a period of over four months with th e pH maintain e d at: pH 1.5, Curve l; pH 3.0, Curve 2 ; pH 5.0, Curve 3; pH 7 .0, Curve 4; pH 9.0, Curve 5. 38 WAVELENGTH; NM WAVELENGTH / NM 4 WAVELENGTH/ NM 39 Table 4 Equilibrium Composition of Cr(III)-TEDTA Solutionsa pH 3 5 % Isomer I %Isomer II 46.5 53.5 50.6 49.4 5.2 94.8 8.2 I 91. 8 aFirst set of numbers at a given pH was obtained from an analysis of the spectra in Figure Sa; second set, from Figure Sb. The extinction coefficients used were: at pH 3, E:g53 = 63.8, £!12 = 37.4' cfi3 = 30.3 and cl12 = -1 -1 I I -1 40.0 M cm ; at pH 5, same except c = 66.0 Mcm 558 40 FIGURE 6 Equilibrium Distribution of Isomers I (e) and II ( ■) as a function of pH from pH 1.5 to 7.0. Points a t pH 3 and 5 were calculated from the spectra in Figures Sa and Sb as described in the text and in Table 4. Point~ at pH 1.5 and 7 WLr e known from previous experiments. 41 0 0 • <O % COMPOSITION 0 (\J 42 VISIBLE AND UV ABSORPTION SPECTRA Visible spectra of solutions of Isomers I, II and Complex III as a function of pH are shown in Figures 7, 8 and 9. UV spectra are shown in Figure 10 and A max values are summarized in Table S along with those for the Cr(III)-EEDTA and Cr(III)-PMDTAcomplexes. Note that only for Isomer II does a UV band appear to be present at 240 nm. For Complex III, this band may be obscured by the intense UV absorption. REACTIONS WITH AZIDE ION The water molecule which occupies the sixth coordination position in the pentadentate EDTA complex of Cr(III) is relatively labile and can readily be replaced by another ligand such as acetate (26) or azide (27,28). Tanaka and coworkers (27) found that such anation by azide intensified and red-shifted the two prominent absorption bands of CrEDTA. The two isomers of CrTEDTA were therefore reacted with azide to compare their behavior with that of the EDTA complex. A 9.4 mM unbuffered solution of Isomer II at pH 4.3 was treated with a fifty-fold-excess of NaN 3 which produced an immediate increase in the pH to 6.7. Heating this solution to boiling for one hour caused the pH to increase to 8.4 and the originally pink solution took on 43 FIGURE 7 Visible Spectra of 10 m~ Solutions of Isomer I: Curve A, pH 4.28 (violet solution); Curve B, pH 7.23 (blue-green solution). 44 ABSORBANCE 0 (JJ 0 0 ~ 0 0 ::, (J1 3 0 0 (j) 0 0 .0 N p p (j) (X) 45 FIGURE 8 Visible Spectra of 10 rnM solutions of Isomer II: Curve A, pH 3.01; Curve B, pH 7.15; Curve C, pH 10.78. 46 0 0 <..O 8 C:E LO 0 0 ~ <1 0.6 0.3 ABSORBANCE 47 FIGURE 9 Visible Spectra of 10 m~ Solutions of Complex III. Curve A, pH 1.98 (reddish-violet); Curve B, pH 8.53 (gray). 48 8 c.o 0 0 <;j"" 0.6 0-4 ABSORBANCE 0.2 49 FIGURE 10 Ultraviolet Spectra of ca . 0.2 mM Cr(III)TEDTA Solutions. Curve A, Isomer I at pH 4.49; Curve B, Isomer I at pH 7.13; Curve C; Isomer II, at pH 7.15; Curve D, Complex III at pH 8.86. 50 0.400 ~ - - , - - - r - - - - - r - - - - - r - - - - - r - - - - - , D 0.300 0.020 w u z <I >-max= 240 E = 1050 0.010 OJ a:: 0.200 0 Cf) OJ <I 0.000 200 250 300 nm 250 350 300 400 450 51 Table 5 AbsorEtion SEectral Data a \ pH Color of the Solution max (nm) b (M- cm- 1 ) 4.28 Violet 558 393 65 51 195 8,650 584 414 54 58 195 17,800 1 E CrTEDTA Isomer I 4,49 7.23 Blue-green 7.13 Isomer II 3.01 Reddish-pink 4. 7 2 Comp 1 ex I II 513 36 5 ~415(weak max) 240 195 40 30 1,050 19,100 7.15 Burnt pink 512 365 ~415(weak max) 41 31 10.78 Orangish-pink 509 36 5 ~415(weak max) 37 29 1. 9 8 Reddish-violet 554 390 309(sh) 58 45 ~5 8.53 Gray 575 407 303(sh) 37 45 ~9 ' 52 Table 5 Cont'd CrEEDTA 1. 7 5 Violet 561 396 4.65 Blue - green 585 415 203(sh) Vi olet Isomer 1. 46 Vio let 557 393 Peach Isomer 6.40 Peach 514 410~} 370 53 47 78 78 ~4,800 CrPMDTA a 64 51 ~60% of those for Violet Isomer Visible spectra were obtained with 10 mM solutions of the complexes in 1 cm quartz cells; UV spectra, with 0.2 mM solutions in 0.1 cm quartz cells. b Extinction coefficients were calculated per mole of Cr(III). c Split band--maxima of approximately equal intensity. .S3 an orange tint. However, the visible spectrum of the solution matched that of the control. Isomer II was not anated by N3 Therefore, The maxima in the visible absorption spectrum of Isomer I undergo a significant and not completely reversible shift to the red in the pH range from 4. 3 to 7 in the absence of N3 Therefore, for the reaction of Isomer I with N3 - , the solution was carefully maintained at pH 4.25-4.35. Addition of a fifty-fold excess of NaN 3 to a 10 mM solution of Isomer I caused an immediate color change from violet to blue-violet. After two hours standing at room temperature, the visible absorption spectrum exhibited two bands: (68 ~ - i cm- 1 ) \ max (E), 570 nm (102); 414 nm which were more intense and red-shifted from those of Isomer I (Table 5). After standing over- night, the pH of the solution increased to 4.6 and the absorption spectrum changed: 424 nm (91 ~ - 1 cm - 1 ). \max (E), 582 nm (136); This final spectrum was immune to further increases in pH, but reducing the pH to 1.8 and gentle heating restored the spectrum of pure Isomer I. By heating the reaction solution at 60°, the final spectrum was obtained in one hour. This behavior closely matched that of CrEDTA (27) and strongly indicates that Isomer I also possessed a labile coordination position. 54 However, anation of Isomer I did not occur even after one hour at 60° C when the solution was adjusted to pH 7 (producing a blue-green solution) before the azide was added. The substitutional lability of both the reddishviolet and gray forms of Complex III were examined by adding a SO-fold excess of N3 each form. to 10 mM solutions of For the reddish-violet form, the initial pH of the blank and the N3 - -containing solutions was 4.60 and 4.58, respectively; for the gray form, 8.54 and 8.39, respectively. Although no hysteresis was observed in the interconversion of the two forms of Complex III, care was taken in these experiments to control the pH during the addition of N3 • All the solutions were boiled for 40 minutes. The color of the N3 - -containing solution of the reddish-violet form turned purple after a few minutes at room temperature. After 5 minutes on the hot plate, the color had changed to an intense, grayish-blue; additional heating had no effect. The pH rose to 6.50 and was lowered to 4.62 without any perceptible change in the color of the solution prior to recording the visible spectrum. and 433 nm (E44 ~ Maxima were observed at 585 (E60) - 1 cm - 1 ). After standing overnight at pH 4.6, these bands red-shifted to 592 (E67) and 441 nm 55 1 (£49 ~ - i cm- ). No further changes in the visible spectrum occurred. Interestingly enough, the color of the blank was not constant during the heating period, becoming somewhat redder and less intense. The pH dropped to 2.77 and was raised to 4.30 before running the visible spectrum. All bands blue-shifted: (£44), 390 ➔ 378 (E33 ~ - l cm - l ) and 309(sh) ➔ 554 ➔ 537 305(sh) nm (£constant); the asymmetry of the principal high energy band increased. The pH decrease and these spectral changes are consistent with partial conversion to Isomer II which apparently did not occur in the N3 - containing solution. Thus, a better blank for comparison with the azide induced spectral shifts was the blank solution before heating (pH 4.24): and 390 (E41 ~ - l cm - l max (E), 554 (E52) A ). The color of the N3 --containing solution of the gray form became slightly greener within minutes at room temperature; heating had no effect. to 8. ·55. (£50 ~ - i The pH rose slightly Maxima were observed at 572 (£38) and 406 nm cm- 1) at pH 8.40. The spectrum was unchanged after standing for 2 days at room temperature and pH 8.3. Although the positions and extinction coefficients of the maxima were essentially unaltered from those of the gray complex before heating (Table 5), the band shapes had been modified. The high energy azide band became 56 more symmetrical and the width of the low energy band decreased. The greenish tint to the color was due to the increased transmission of light in the 500 nm region by the N3 - -containing solution. Heating had a much larger effect on the color and spectrum of the blank solution. Its color changed to grayish-purple, the visible bands shifting to: Amax (E) at pH 8.34, 558 (31), 404 (39) - l and 308(sh) nm (~9 M cm rose slightly to 8.88.) - l ) (cf. Table 5). (The pH These changes in the blank are again ascribed to formation of Isomer II. azide result is interesting. However, the Excess azide seems to inhibit the conversion of the gray form of Complex III to Isomer II. The blue-green form of CrEEDTA was not anated by azide. Unlike Isomer II, the peach isomer of CrPMDTA did react with azide, albeit slowly. The usual conditions were used and the solution heated for 1 1/2 hours. The pH rose from 6.46 to 8.59 as the color changed from peach to brownish-orange. A blank solution set at pH 8.6 and boiled for 1 1/2 hours did not change color. The color of the azide-containing solution darkened slowly on standing for over a week at room temperature and pH 6.6, whereas the blank was unchan~ed. summarized in Table 6. The spectral data are 57 Table 6 CrPMDTA + SO-Fold Excess N3 Reaction Time Sequence A max (nm) Color Ha E._!_ Before addition of N3 :- Peach 6.40 3 70] spit l. peak 410 514 After boiling 1.5 hrs. BrownishOrange 6.39 415 (asym. on blue side) 513J sp 1 1t • pea k 553 After 21 hrs. at room temperature BlackishOrange 6.44 415 (asym. on blue side) 560 (asym. on blue side) After 8 days at b room temperature Gray 6.01 421 (n~arly symmetrical) 567 (symmetry increased) a pH of solution during recording of visible spectrum. b Color had been constant for several days. 58 ION-EXCHANGE BEHAVIOR Attempts to determine the charges of Isomers I and II from their behavior on ion-exchange resins were only moderately successful. Isomer I behaved as a cation when loaded onto AG SOW X-8 cation exchange resin from a 0.1 M HClO4 solution, but it could be subsequently eluted by washing with dilute (0.1 ~) NaC1O 4 . It seems likely that the cationic form existing at pH 1 represents the tetradentate form in which there are two uncoordinated carboxylate groups because CrEDTA exhibits such behavior at low pH (23). Isomer II showed no affinity for either an AG SOW X-8 cation exchange resin or an AG 1 X-8 anion exchange column at pH 4 in 0.1 ~ NaC1O 4 . CrEDTA complex behaved similarly. The monoanionic The gray form of Complex III behaved as a highly charged anion on an AG 1 X-8 column in the Cl form. The band did not move discernably even when washed with 5 M NaC1O 4 . It was finally removed by washing with 1 M HCl. A solution of Isomer II at neutral pH was loaded onto a Selectacel DEAE anion column in the ClO 4 form. No movement was discernable after washing for an hour with 0.1 N HClO4 The behavior of the various Cr(III) complexes on Selectacel was investigated. It was found that CrEDTA at pH 5.4 (violet, monoanionic), CrEEDTA at pH 6.95 59 (blue-green, monoanionic) and Isomer I at pH 7 (bluegreen, anionic) banded on loading, were not removed with distilled water and eluted readily with 0.01 N NaClO4 (although the bands broadened considerably). ~ The elution rate of these complexes with 0.01 NaC1O 4 matched that of Isomer I at pH 2 (violet, neutral) with 0.01 N HC1O 4 . Isomer II (pH 7, monoanionic), the gray form of Complex III (pH 8.5, polyanionic) and the peach form of CrPMDTA (pH 6, monoanionic) exhibited considerable affinity for the Selectacel resin. complexes could be moved with 0.01 None of these ~ NaC1O 4 . Rinsing with 0.1 ~ NaC1O 4 broadened these bands somewhat, but did not remove them from the column. Rapid elution and considerable sharpening of these bands resulted on washing with 1 M NaC1O 4 . Isomer II and peach CrPMDTA moved at approximately the same rate with 1 M NaC1O 4 ; gray Complex III, somewhat slower. The disparity in the behavior of these two groups of Cr(III) complexes on Selectacel is especially notable because it is independent of complex charge. REACTIONS WITH HEAVY METAL CATIONS AND METHYL MERCURY CATION Isomer II reacted with Hg+ precipitates. 2 , Ag+ and Pb+ 2 to produce When pre~ipitation was incomplete, the 60 color of the supernatant diminished in intensity, but no change in its hue was observed. Complete precipitation (as judged from the absence of color in the supernatant) occurred when an equimolar quantity of Hg+ added. 2 had been Precipitation was incomplete when less Hg was used. Hg+ 2 The Pb+ and Ag+ 2 solid was soluble in acid. +2 The precipitates were insoluble in water or acid but dissolved readily in 1 M HCl. The Hg +2 precipitate was also soluble in excess EDTA and the spectrum of the resulting pink solution matched that of Isomer II. Analysis of the pink Hg+ 2 solid (Experi- mental Section) yielded a mercury to chromium ratio one to one. of The precipitate formed with Ag+ was quite photosensitive and turned dark red from its initial pink color unless protected from the light. Both the gray and violet forms of Complex III behaved similarly to Isomer II. Complete precipitation was observed when an equimolar quantity of Hg+ added. 2 was The lavender precipitate was insoluble in acid. Isomer I and the free TEDTA also formed precipitates with Hg +2 + and Ag , but complete precipitation required much higher concentrations of· the heavy metal cation. The Ag+-Isomer I solid was also light sensitive. The Cr(III)-EEDTA complex in which an oxygen atom replaces 61 the sulfur atom of CrTEDTA exhibited no apparent reaction 2 with Hg + • The behavior of Isomers I and II with CH 3Hg+ paralleled their behavior with Hg +2 + and Ag . A pink precipitate was obtained at pH <4 on addition of excess CH 3HgOH to a solution of Isomer II, but no change in the hue of the supernatant occurred. was decreased As the pH of the solution + below 4, smaller quantities of CH 3Hg were needed to precipitate Isomer II. Complete removal of Isomer II from solution was observed at pH 1.6. Dissolution of the pink solid was effected by raising the pH (pKb of CH 3HgOH = 9.4 (29)) as the comnlex dissociated. A solution of Isomer I became only slightly turbid after the addition of a 60-fold excess of CH3Hg + at pH 1.6. When the solid had settled, the color of the supernatant was unchanged from that of a pure solution of Isomer I. IR EVIDENCE The IR spectra of TEDTA and the Hg +2 precipitates of Isomers I, II and Complex III were recorded. Little information was obtained from the carbonyl and N-H stretching regions as the bands were broad and poorly resolved. The most distinctive region was from 850 to 300 cm- 1 and this portion of each spectrum is reproduced in Figure 11. Notable are the relative intensities of 62 FIGURE 11 The 850 to 300 cm (KBr): - 1 Region of the IR Spectrum Curve A, TEDTA; Curve B, Hg+ of Isomer I; Curve C, Hg+ Curve D, Hg+ 2 2 2 precipitate precipitate of Isomer II; precipitate of Complex III. Circled part of each spectrum indicates a discontinuity caused by a grating change. 63 700 400 64 the absorptions at 750 and 700 cm - 1 (743 and 690 cm - 1 in Complex III) for Isomer II and Complex III compared with Isomer I, in particular, the intensity of the 750 band is enhanced in the spectra of the former complexes; the presence of a moderately strong band at 620 cm - 1 in the spectra of Isomer II and Complex III and the absence of this band in the spectrum of Isomer I; and the presence of a weak absorption at 830 cm - 1 only in the spectrum of Complex III. ELECTROCHEMISTRY Isomers I and II were both adsorbed on mercury electrodes but they exhibited marked differences in their electrochemical reductions. voltammograms for both isomers. Figure 12 shows cyclic Isomer II was adso r bed so intensely that the peak potentials for the reduction of the initially adsorbed and the unadsorbed molecules were separated by 110 mV. The much weaker adsorption of Isomer I produced only a single reduction peak with a peak current that increasingly exceeded the diffusion controlled value as the scan rate was increased (30). The cyclic voltammograms of Complex III were very similar to those for Isomer II at the same pH. The cyclic voltammetry of the CrEEDTA complex under the same conditions produced a single reversible wave with no evidence of significant adsorption. 65 FIGURE 12 Cyclic voltammetry of 1 m~ solutions of Isomers I and II at pH 4, scan rate: 10 v/s. 66 <1 ~ 1-- zw 40 0 0:: 0::. =:) (_) ----,q() -.I -.9 V VS SCE 40 <t ~ 1--- 0 zw f2-4o =:) (_) I • -, 2 V VS SCE- ' 2 67 Table 7 lists representative data on the adsorption of both isomers of CrTEDTA and of CrEEDTA as determined chronocoulometrically (31). As inferred from the cyclic voltammetry, CrEEDTA is not adsorbed. Isomer II is strongly adsorbed and the adsorption of Isomer I 1s moderate. The adsorption of Isomer II increased as the electrode potential was made more positive. The sign of the potential dependence of the adsorption of Isomer I could not be established unambiguously but the magnitude of the potential dependence was certainly much less than for I so mer I I. 68 Table 7 Adsorption of Cr(III)-EEDTA and Cr(III)-TEDTA Complexes on Mercurya Complex r X 10 l 0 moles cm Cr (I I I) -EEDTA b 0 Isomer IC 0.30 Isomer I Id 2.0 - 2 a Measured at an electrode potential of -0.4 V vs. SCE. b 0.94 mM CrEEDTA in 0.5 M NaCl04, pH 4 . 6 . C 0.024 rnM Isomer I in 1 M NaCl04, pH 4.30. d 0.027 mM Isomer II in 1 M NaCl04, pH 4.77. 69 DISCUSSION The structures we propose for the two isomeric Cr(III)-TEDTA complex are shown 1n Figure 13. In Isomer I, we believe TEDTA acts as a pentadentate ligand resembling the behavior of EDTA in the CrEDTA complex (16, 22-25). Only three of the four carboxylate groups are coordinated and a water molecule occupies the sixth coordination position on Cr(III). By contrast, in Isomer II, TEDTA is believed to act as a sexad entate ligand with the thioether sulfur atom coordinated to the metal. The evidence and reasoning upon which these proposed structures are based will be elaborated in what follows. DENTICITY OF THE Cr(III)-TEDTA COMPLEXES pH BEHAVIOR It will be useful to discuss the behavior of the two isomers of Cr(III)-TEDTA by comparison with the extensively studied Cr(III)-EDTA complex. The two inflections in the pH titration curve for the latter complex correspond to the titration of 4 and 5 moles of acid per mole of chromium and they have been identified with the titration of the free protons present along 70 FIGURE 13 Proposed structur es of the two monomeric isomers of Cr(III)-TEDTA. 71 ISOMER I ISOMER II 72 with the single, uncoordinated carboxyl group (pK 1 = followed by the coordinated water molecule (pK 2 7.5) (16). = 3.1) The originally violet solution becomes blue as the pH traverses the range between the first and second inflection and the monohydroxy complex is formed (reaction 2): CrHEDTA(OH 2 ) + OH (1) violet CrEDTA (OH 2 ) violet + OH pK2=7.5 CrEDTA(OH) violet - 2 + H2 0 (2) blue This interpretation of the pH behavior of CrEDTA received recent additional support from kinetic measurements by Thorneley, et.al. (23). The titration curve for Isomer II resembles that for CrEEDTA in that only a single inflection is observed which corresponds to the titration of four moles of protons per mole of chromium. This implies the absence of a titrateable, coordinated water molecule and, therefore, that all six of the coordination positions around the metal are occupied by the- ligand after four moles of acid have been neutralized. The inertness of the coordination environment to changes in pH is demonstrated for Isomer II by the absence of significant color changes for pH values from 73 2 to 11 and for CrEEDTA by the slowness of the color change when the pH 1s reduced from 8 to 3. The reasons for proposing that the sulfur atom of TEDTA occupies the sixth coordination position rather than the fourth carboxylate group, are set out below. The titration curve for Isomer I (Figure 4a) exhibits two poorly defined inflections marking the consumption of 3 and 4 moles of base per mole of chromium. The drawn out nature of the titration curve and the unusually high pH observed after 3.5 moles of base (ca. were added 5.5) (32) suggest a more complicated picture than simply conversion of the violet pentadentate complex to a blue-green hexadentate complex as is observed for CrEEDTA. (Note: The pH after addition of 3.5 moles of base for CrEEDTA was 3.85.) The ability of the free acetate group in CrEDTA(OH 2 ) to labilize the coordinated water molecule has been well established (26-28). Azide substitution experiments (described above and discussed below) have demonstrated the lability of the coordinated water in the violet form of Isomer I. Therefore, it is proposed that f ol lowing neutralizati on of the dangling carboxyl group, the labilized water is displaced by the free acetate arm of another molecule of Isomer I . Then the two possible sources of acid neutralized between 3 and 4 moles of base added are the proton released from the bridging acetate 74 group and the proto n on the non-bridging, uncoordinated acetate group. The combined action of these two different acids is believed to be responsible for the absence of a sharp inflection at 4 moles of base added. The addition of a fifth mole of base results in dissociation of the dimer and formation of a green hydroxy complex. The model is summarized in Scheme 2. Ample support for the proposed model can be found in the literature. Tanaka and co-workers have shown that acetate readily substitutes on CrEDTA(OH 2 ) ( 2 6) . More recently, Ogino, et.al. synthesized several binuclear complexes in which the free acetate arm of NiEDTA(OH2)was bound to [Co(NH 3 ) 5 ] +3 2 , (33). Carboxylatebridged dimers have frequently been proposed to explain the coordination in many metal ion complexes of amino acids (17, 34-36). Moreover, several additional lines of experimental evidence strongly favor the formulation of the blue-green form of Isomer I as an acetate bridged dimer. hysteresis experiment is corroborative. The pH If none of the dimer had dissociated when the pH was reduced from 8 to 4, 50% less acid would have been required to effect this pH change than base was used in originally raising the pH from 4 to 8, As 35% less acid was used and the color turned bluish-violet, some dissociation occurred. However, these results are also consistent with hexadentate I coordination in the blue-green form. But the cyclic 75 Scheme 2 (1) violet violet relativelyl! slow slow I (2a) fast n (2b) slow - 2 + H blue-green blue-green - 2 blue-green green green + (3) 76 voltammetric behavior cannot be satisfactorily expl a ined in terms of this latter proposal. At pH 7-9 where the solution is blue-green, three cathodic waves were observed for Isomer I - - the first representing reduction of the violet form followed by two, one-electron waves of equal hei ght at mor e negative potentials. By contra s t, the blue~green form of Cr(III)-EEDTA exhibited only a single reversible, one-electron wave from pH 5.5 to 9, which gradually disappeared at higher pH. The two more cathodic peaks for Isomer I are assigned to the successive reduction of the two nonequivalent metal centers in the dimer. Assignment of these two waves to reduction of the hexadentate and hydroxy complexes of Isomer I is incompatible with the fact that their peak currents are always equal and with the ob servation that CrEDTA(OH) - 2 and the green form of Cr(III)-EEDTA (presumably a mono- or dihydroxy complex) are reducible. not The green form of Isomer II was also irreducible. Potential step experiments as a function of concentration substantiated the existence of a dimer (KD = 2410 ~ - l ) (Table 1, Chapter II). A complete discussion of the electrochemistry follows in Chapter II. The observed conversion of Complex III from reddishviolet to gray implies a change in the coordination of the metal as a function of pH. The ease of this inter- 77 conversion indicates that it occurs without making or breaking bonds to Cr(III). REACTIONS WITH AZIDE ION Isomer I reacts readily with azide to produce spectral changes resembling those obtained with the pentadentate CrEDTA(OH 2 ) complex. This is consistent with the presence of a labile water molecule in the coordination sphere of Isomer I as shown in Figure 13 and Scheme 2. The lack of apparent reaction between the proposed acetate bridged dimer of Isomer I and azide is attributed to the inefficiency of the anchimeric c ~s effect (37) under these conditions. The unbound carboxylate group can labilize the coordinated water cis to it by nucleophilic attack. There are two likely geometrical isomers of the dimer; one of these is shown in Scheme 2, the other is shown below: 78 In the former isomer, cis attack is not possible. In the isomer depicted above, although cis attack is possible, freedom of rotation of the two halves of the molecule about the bridged group significantly decreases the probability that the two halves of the dimer will be properly oriented for cis attack. Considering this orientation problem and the existence of another isomer in which cis attack is impossible (Scheme 2), one would predict a much slower and incomulete reaction of azide with the dimer relative to the monomer. Indeed, it seems likely that when the dimer was properly oriented for c i s attack, formation of a second acetate bridge could also occur. The failure of Isomer II to react with N3 conditions where Isomer I and CrEDTA(OH2) under are readily anated is strong evidence that Isomer II has no open coordination position (Figure 13). Hexadentate coordination is also indicated for the blue-green form of CrEEDTA which also failed to react with N3 - . The reaction of the peach isomer of CrPMDTA with azide is superficially inconsistent with the pH behavior of this complex which showed it to be hexadentate. The apparent lability of the coordinated carboxylates may result from the bulkiness of the 8-membered chelate ring in the trans isomer which the peach complex is 79 believed to be. As can be seen from the diagram below, the pentamethylene group bridging the two N's crowds acetate groups a and b. Therefore, these groups might be expected to be somewhat labile. Steric interactions in transition metal complexes have been shown to weaken metal-ligand bonds (38-40). Hamilton and Alexander attributed the considerably higher rate of aquation of trans-(RR,SS)-dichloro(l,4,8,11tetraazaundecane)cobalt(III) relative to the trans-(RS) isomer to the greater crowding of the axial chlorides in the (RR,SS)-racemate (40). The greater lability of the peach CrPMDTA complex compared with Isomer II supports sulfur coordination in Isomer II. PMDTA and TEDTA are virtually the same size (the nitrogen-nitrogen djstance in TEDTA is 0.56 i longer). Therefore, if the two nitrogens and four acetates were coordinated to Cr(III) in Isomer II, it would be expected to undergo anation by azide as does 80 trans-CrPMDTA. The failure of Isomer II to react with azide indicates the absence of labilizing steric interactions with the coordinated acetate groups and points to the participation of the sulfur atom rather than a carboxylate group in achieving hexacoordination. The reddish-violet form of Complex III is readily anated and behaves similarly to Isomer I. Consequently, it is concluded that there is a free acetate group cis to a coordinated water in this form of Complex III. However, the results obtained with the gray form are atypical. The absence of a shift in the visible bands indicates that azide substitution did not occur. But, the increased absorbance at the high energy maximum, the increased transmission in the 500 nm region, and the resistance of the gray form to isomerization in the presence of azide imply that azide is interacting with the gray complex in some manner. The nature of this interaction is not understood. CIS AND TRANS ISOMERS OF Cr (III) -TEDTA In the preceding discussion the existence of cis and trans (with respect to N'5) isomers of Cr(III)-TEDTA has been proposed on the basis of the solution chemistry of the two isomers. Considerable spectral evidence can also be marshalled to support this proposal. 81 VISIBLE ABSORPTION SPECTRA Wilkins and Williams have summarized the utility of visible absorption spectroscopy for distinguishing between cis and trans isomers of transition metal complexes (41). Theory predicts greater splitting of thee and t 2 levels in the trans isomer compared to the cis isomer. In practice, the cis bands are only broadened, whereas, the trans bands are actually split and usually, considerably diminished in intensity. cis-Cr(en) 2 (OH 2 ) 2 + 485 nm (E67. 6 ~ - 1 3 For example, the spectrum of displays maxima at 366 (E43.7) and • cm - 1 ) , while for the trans isomer, spl 1• tt1ng of the low energy band is observed: - 1 (E39.3) and 508 nm (E22.5 M cm - 1 361 (E39.2), 442 ) (20). The CrEDTA(OH 2 ) complex, in which the two nitrogens are known to have the cis configuration from crystal structure measurements (22), exhibits a spectrum with two, relatively symmetrical absorptions at 395 and 545 nm. Similar spectra are observed for all forms of Isomer I, both the violet and blue-green forms of CrEEDTA and the violet form of CrPMDTA (Table 5). It therefore seems likely that the nitrogen atoms are cis in each of these complexes. The bands in the ·spectrum of Isomer I I and the peach form of CrPMDTA are diminished in intensity, the low energy band is broadened and the high energy band is split. These spectral features are consistent 82 with the nitrogen atoms in these two complexes being trans to each other. Busch and coworkers (42) and Bosnich and coworkers (43) have shown that the stabilities of geometrical isomers are affected greatly by the chelate ring size. In particular, they found that the stability of the trans isomer increased as the chelate ring size increased. J¢rgensen observed that the lower the ring stability, the lower Dq (44). The trends in the spectral data in Table 8 for various hexadentate Cr(III) chelate complexes are consistent with these expectations. As the size of the nitrogen chelate ring increases in the cis isomers, the visible bands red-shift, reflecting the decreasing stability of the se isomers. The greater distance between the nitrogen atoms in CrPMDTA and CrTEDTA leads to the formation of as blue-shift evidenced by the the t r ans of isomer the visible band s . BEHAVIOR ON SELECTACEL ANION EXCHANGE RESIN The remarkable difference between the mobility of Isomer I and a known cis complex, CrEDTA(OH 2 ) compared with Isomer II and a probable trans complex, peach CrPMDTA , on Selectacel anion exchange resin supports the assignment of the cis configuration to Isomer I and the trans configurauion to Isomer II. 556,397 585,415 517,420,370 512,415,365 cis-CrTDTAc cis-CrEEDTA trana-CrPMDTA trans-CrTEDTA 8 8 8 Peach Burnt Pink 7 6 Blue-Green Violet Red Color this work this work 9.12 9.68 this work 21 45 8.90 7.58 6.04 Ref. c TDTA is tetramethylenedinitrilotetraacetic acid. b TRDTA is trimethylenedinitrilotetraacetic acid. a Bond distances obtained from the 51st ed. of the Handbook of Chemistry and Physics. 509,385 (nm) cis-CrTRDTAb Complex >- max Size of the N-N Chelate Ring Distance Between the N'sa No. of Members (A) Trends in the Stabilities of Hexadentate Cr(III) Chelate Complexes as a Function of the Nitrogen Chelate Ring Size Table 8 00 vi 84 SULFUR COORDINATION TO Cr(III) In his review, J¢rgensen states that thioethers, which are part of multidentate ligands containing amino groups, are much more attractive to central metal ions which normally display no affinity for R2 S (46). Models show that when the nitrogens are in the trans configuration in CrTEDTA, the sulfur is forced into proximity to the Cr(III), thus facilitating its binding. A few examples of thioether sulfur coordination to Cr(III) have been reported (47,48). Additional evidence for sulfur coordination in Isomer II is discussed below. UV ABSORPTION SPECTRA The UV absorption spectrum of Isomer II contains a shoulder at 240 nm (E1050) which is not present in the spectrum of Isomer I or the Cr(III) complexes of EDTA and EEDTA. The intensity of the band suggests that it arises from a charge transfer transition. Sulfur-to- chromium charge transfer would be consistent with the structure proposed for Isomer II (Figure 13). The higher energy of this band compared to those for Cr(III) thiolate complexes, which are .generally observed at 265275 nm (49), is consistent with the lesser reducing power of thioethers compared with thiols. The absence of such a band in the UV spectrum of Isomer I supports the absence of a sulfur-chromium bond in this isomer (Figure 13). 85 REACTIONS WITH HEAVY METAL CATIONS AND METHYL MERCURY CATION There is little doubt that the reaction between . I somers I an d II an d Hg +2 , Ag + or CH3Hg + involves the binding of the cation to the sulfur moiety. binding of Hg +2 , Ag + . and CH 3Hg + Monodentate to the thioether sulfur atoms in methionine (50,36,51) and S-methylcysteine (51,34) has been established at low pH. The fact that hig he r concentrations of these cations are required for reaction with Isomer I than with Isomer II is added evidence of the different coordination chemistry exhibited by the thioether sulfur in the two isomers. This difference in behavior may reflect an enhancement of the nucleo~hilicity of the coordinated sulfur due to the thioether n-accepting electron density (52,53) from Cr(III), thus increasing the attractiveness of the sulfur atom to Hg +2 , Ag + and CH 3 Hg + (54). IR EVIDENCE The similarity of the IR spectra of Isomer II and Complex III in the fingerprint region suggests that they are structurally similar and unlike Isomer I. at 620 cm- 1 , The band which is present.only in the spectra of Isomer II and Complex III, may originate from a Hg-S-Cr + stretch as it is absent in the spectrum of the Ag -Isomer II adduct. The weak band at 830 cm- 1 , which is present only 86 in the from spec tr urn a Cr-0 of Complex I I I, stretch of the Cr-0- Cr suggests that Complex McAuliffe and Perry have the IR spectrum at cm 830 - 1 in III is probably an results sys tern (5-:S) and oxo-bridged reported a weak of dimer. absorption µ-oxodi[bis(rnethio- nato)aquochromium(III)], which they assigned to a Cr-0 stretch within the Cr-0-Cr unit (56). ELECTROCHEMISTRY The marked difference in the extent of the reactions of Isomers by the difference to the surface The intensity adsorption many I of organic and Isomer I are specific surface. The very strong + and Hg of comparable a tom because similar adsorption is (Table 7). to the those of more signals a stronger the to mercury certainly a 1 mos t of -the believed of the non- adsorption CrEEDTA complex. is matched However, Isomer II s it e is attachment dependence interaction with binding +2 adsorptive adsorbates (57). and more the their and potential adsorption sulfur in CH 3 Hg of mercury. electrodes intense the I I with The be the of unusually result 87 of the back donation involved in the formation of the chromium-sulfur bond in Isomer II which "softens" the sulfur atom thus enhancing its adsorbability (58). To summarize the chemical data on the Cr(III)TEDTA complexes: i) The pH behavior and indicate that Isomer I azide experiments is pentadentate with a coordinated water molecule and that Isomer I I is hexadentate. ii) The splitting of the visible bands, their decreased intensity for/ Isomer I and the the mobilities of complex, nitrogens Isomer II Isomer to I and a known cis complex, Isomer peach CrPMDTA , are cis relative discrimination on Selectacel between compared to CrEDTA(OH2) trans Isomer in Isomer II and a probable suggest I and that the trans in II. iii) The S + Cr(III) CT band at 240 nm, the inertness of Isomer II relative to hexadentate, transCr!MDTA to reaction with azide, its apparently strong binding to Hg +2 and CH 3 Hg + and strong adsorption on positively charged mercury electrodes are all consistent with the presence of a ch r omium-sulfur bond in Isomer II. 88 On the basis of these data, the structures for Isomers I and II in Figure 13 are proposed. That is, Isomer I is cis-[thiobis(ethylenenitrilo)tetraacetato]aquochromium(III), cis-CrTEDTA(OH 2 ) and Isomer II is trans-[thiobis(ethylenenitrilo)tetraacetato-S]chromate(III), trans-CrTEDTA-. With regard to the dimers, iv) The drawn out titration curve for cis- CrTEDTA(OH 2 ), the sluggishness of conversion of the blue-green form to cis-CrTEDTA(OH 2 ) , the lack of reaction of the blue-green form with azide, the electrochemical nonequivalence of the two metal centers, the results of the potential step experiment which were consistent with a dimerization constant of 2410 ± 140 ~ - 1 (cf. p. 76) are all consistent with the formulation of the blue-green form of cis-CrTEDTA(OH 2 ) as the acetate-bridged dimer, cis-[thiobis(ethylenenitrilo)tetraacetato]aquochromium(III)-µ-N-acetato-cis-[thiobis(ethylenenitrilo)tetraacetato]chromate(III), TEDTA(OH2)Cr(NCH 2 COO)CrTEDTA (Scheme 2). v) For Complex III, the high negative charge of the gray form and the IR band at 830 cm - l suggest that the complex is an oxo-bridged dimer; the asymmetrical visible bands and high affinity for Selectacel suggest that the nitrogens are trans; the apparently strong 89 binding with Hg+ 2 and strong adsorption on mercury electrodes as shown by cyclic voltammetry are indicative of chromium-sulfur coordination. The azide experiment demonstrates the presence of a coordinated, labile water molecule in the reddish-violet form. The ready inter- conversion of the two forms requires that no bonds to Cr(III) are made or broken. Structures for the two forms of Complex III are shown below: n reddish-violet n-2 gray µ-oxodi(trans-[thiobis(ethylenenitrilo)tetraacetato-S]aquochromate(III)), µ-oxodi(trans- [thiobis(ethylenenitrilo)tetraacetato-S]hydroxochromate(III)), [tr ans -CrTEDTA(OH2)]20 [ trans -CrTEDTA(OH)]20 maximum charge: maximum charge: -4 -6 90 DRIVING FORCE FOR THE ISOMERIZATION OF Cr(III)-TEDTA COMPLEXES Figure shows 6 the the equilibrium between the conversion of was found that pH dependence Isomers I Isomer I one mole of per mole of believed that this results decrease 1n the pKa when the configuration changes from Although other factors may the of enhancement It is a significant the atoms the fourth carboxylate of the two to trans. cis be involved, of acidity uncoordinated acetate group appears be the source of free energy controls cis - trans the the nitrogen of the the atoms. nitrogens which of consequence a being the to isomerization As it was from nitrogen During acid chromium. of I. Isomer II, to released group and of trans , the sulfur atom is brought into close proximity with the Cr(III) center thus facilitating its sulfur the Coordination coordination. mo i ety replacement doubt no is of the of the favored by large, eight-membered 91 ring in the c~s isomer with membered rings in Isomer I I. two, five- 92 REFERENCES (1) H. S. Lim and F. C. Anson, J. Electroanal. Chem., ~ ' 297 (1971). (2) F. C. Anson and R. S. Rodgers, J. Electroanal. Chem., !Z._, 287 (1973). (3) S. N. Frank and F. C. Anson, J. Electroanal. Chem., ~ ' 55 (1974). (4) J. Gulens, D. Konrad and F. C. Anson, J. Electrochem. Soc., 121, 1421 (1974). (5) F. C. Anson, Accts. Chem. Res.,~' 400 (1975). (6) F. C. Anson, J. H. Christie and R. A. Osteryoung, J . Electroanal. Chem., .!_l, 343 (1967). (7) G. W. O'Dom and R. W. Murray, J. Electroanal. Chem., _!i, 327 (1968). (8) F. C. Anson and D. J. Barclay, Anal. Chem., 40, 1791 (1968). (9) D. J. Barclay and F. C. Anson, J. Electroanal. Chem., ~ ' 71 (1970). (10) a) F. C. Anson, Anal. Chem., 36, 520 (1964); b) A. T. Hubbard and F. C. Anson, Anal-.-Chem., ~ ' 1601 (1966). (11) D. J. Barclay, E. Passeron and F. C. Anson, Inorg. Chem., ~' 1024 (1970). (12) C. Altman and E. L. King, J. Am. Chem. Soc.,~' 2825 (1961). (13) G. W. Haupt, J. Res. Nat'l. Bur. Standards, ~ ' 414 (1952). (14) D. D. Smolin, L. M. Kazbitnaya and Yu. M. Viktovov, Gen. Chem. USSR, l!, 376~ (1964). (15) S. Kaufman and L.S. Keyes, Anal. Chem.,~' 1777 (1964). (16) R. E. Hamm, J. Am. Chem. Soc., (17) A. Ricard, Ann. Chim.,~' 213 (1974). .zi, 5670 (1953). 93 (18) G. H. Stout and L. H. Jensen, X-Ray Structure Determination. A Practical Guide, The Macmillan Co., N.Y. (1968) pp. 62-66. (19) W. W. Wendlandt and L. K. Sveum, J. Inorg. Nucl. Ch em. , ~, 3 9 3 ( 19 6 6) . (20) C. S. Garner and D. A. House in Transition Metal Chemistry, V. 6, 3d., R.L.Carlin, Marcel Dekker, Inc., N.Y. (1970) pp. 59-295. (21) H. Ogino, J -J. Chung and N. Tanaka, Inorg. Nucl. Chem. Letts., l, 125 (1971). (22) J. L. Hoard, C.H. L. Kennard and G. S. Smith, Inorg. Chem.,~' 1316 (1963). (23) R. N. F. Thorneley, A. G. Sykes and P. Gans, Chem. Soc. (A), 1494 (1971). (24) G. Schwarzenbach and H. Biedermann, Helv. Chim. Acta, ~' 459 (1948). (25) H. Brintzinger, H. Thiele and V. Muller, Z. Anorg. Allgem. Chem., 251, 285 (1943). (26) H. Ogino, T. Watanabe and N. Tanaka, Chem. Letts., 91 (1974). (27) l· H. Ogino, T. Watanabe and N. Tanaka, Inorg. Chem., !.!, 2093 (1975). (28) Y. Sulfab, R. S. Taylor and A. G. Sykes, Inorg. Chem., 12., 2388 (1976). (29) G. Schwarzenbach and M. Schellenberg, Helv. Chim. Act a, ~, 28 ( 19 6 5) . (30) R.H. Wopschall and I. Shain, Anal. Chem.,~' 1514 (1967). (31) J. H. Christie, R. A. Osteryoung and F. C. Anson, J. Electroanal. Chem., !_l, 236 (1967). (32) Similar titration curves for the tetra- and hexamethylenedinitrilotetraacetate complexes of Cr(III) have been reported: A. Ricard, Ann. Chim.,~' 203 (1974). 94 ( 3 3) H. Ogino, K. Tsukahara and N. Tanaka, Inorg. Chem., .!.§_, 1215 (1977). (34) C. A. McAuliffe, Inorg. Chem., ( 3 5) g, 1699 (1973). S. E. Livingstone and J. D. Nolan, Inorg. Chem., '}_, 144 7 (1968). ( 36) C. A. McAuliffe, J. V. Quagliano and L. M. Vallarino, Inorg. Chem.,~' 1996 (1966). (37) M. V. Olson, Inorg. Chem., (38) S. F. Pavkovic and D. W. Meek, Inorg. Chem.,!, 20 g, 1416 (1973). (1965). (39) a) J. I. Legg and D. W. Cooke, Inorg. Chem.,!, 1576 ( 19 6 5) ; b) Ibid. , i, 5 9 4 ( 19 6 6) . (40) H. G. Hamilton, Jr., and M. D. Alexander, J. Am. Chem. Soc.,~' 5065 (1967). (41) R. G. Wilkins and M. J. P. Williams in Modern Coordination Chemistr . Princi les and Methods, e . J. Lewis an R. G. Wil ins, Interscience Publ., Inc. , N. Y. ( 19 6 0) pp. 18 7 -191 . (42) J. H. Worrell, T. E. MacDermott and D. H. Busch, J. Chem. Soc. Chem. Communs., 661 (1969). ( 43) B. Bosnich, R. D. Gillard, E. D. McKenzie and G. A. Webb, J. Chem. Soc. (A), 1331 (1966). (44) C. K. J¢rgensen, Acta Chem. Scand., ~' 1362 (1955). ( 4 5) J. A. Weyh and R. E. Hamm, Inorg. Chem.,'}_, 2431 (1968). ( 4 6) C. K. J¢rgensen, Inorg. Chim. Acta Rev.,~. 65 (1968). (47) R. J. H. Clark and G. Natile, Inorg. Chim. Acta,!, 533 (1970). ( 4 8) a) J. Podlaha and J. Podlahova, Inorg. Chim. Acta,!, 521,549 (1970); b) Ihid., i, 413,420 (1971). { 4 9) R. !-I. Lane, F. A. Sedor, n. J. Gilroy and L. E. Bennett, Inorg. Chem., ~, 102 (1977). (SO) D. F. S. Natusch and L. J. Porter, J. Chem. Soc. Chem. Communs, 596 (1970). 95 (51) M. T. Fairhurst and D. L. Rabenstein, Inorg. Chem., .!_!, 1413 (1975). (52) R. J. P. Williams, Ann. Reports Progr. Chem., 56, 87 (1959). (53) S. E. Livingstone, Quart. Rev. (London),~' 386 (1965). (54) Hg+ forms much stronger complexes with thiols than with thioethers: a) A. E. Martell and R. M. Smith, Critical Stability Constants, V. 1, Plenum Press, N.Y. (1974); b) A. J. Carty and N. J. Taylor, J. Chem. Soc. Chem. Communs, 214 (1976). (55) 2 C. G. Barraclough, J. Lewis and R. S. Nyholm, J. Chem. Soc., 3552 (1959). (56) C. A. McAuliffe and W. D. Perry, Inorg. Nucl. Chem. Letts., lQ_, 367 (1974). (57) B. B. Damaskin, 0. A. Petrii and V. V. Batrakov, Adsor tion of Organic Com ounds on Electrodes, Plenum Press, N.Y. 1971). (58) D. J. Barclay and J. Caja, Croat. Chem. Acta,~' 221 (1971). 96 CHAPTER II ELECTROCHEMICAL STUDY OF THE ADSORPTION OF THE CIS AND TRANS ISOMERS OF CHROMIUM(III)-TEDTA ON MERCURY ELECTRODES 97 INTRODUCTION In Chapter I, the coordination chemistry of Cr(III) with TEDTA was described and evidence for the formation of both cis and trans isomers (with respect to the nitrogens) was presented. In the complex having the trans configuration, the sulfur atom of the thioether group is forced to lie close enough to the Cr(III) center for a chromium-sulfur bond to be formed. Because of our general interest in ligands which may be used to induce the adsorption of metal cations on electrode surfaces (1-4), the electrochemistry of these two, new complexes of chromium(III) was examined. Striking differences in the cyclic voltammetry and adsorption of the two isomers at mercury electrodes were observed and are described in this chapter. 98 EXPERIMENTAL APPARATUS A small volume (10 ml), two compartment cell was employed which isolated the sodium chloride saturated calomel reference electrode (Sargent-Welch) from the hanging (Brinkmann Instruments, Inc.) or dropping mercury working electrode (dme) and the platinum wire auxiliary electrode. The area of the hanging mercury drop electrode was 0.032 cm 2 • dme was 0.65 mg sec - 1 . The mercury flow rate for the Electrocapillary curves were measured using a digital drop timer previously described (5) in conjunction with a Princeton Applied Research (PAR) Model 174 Polarograph. Cyclic voltammograms were obtained using either a conventional multipurpose electrochemical instrument based on operational amplifiers which was constructed in this laboratory, or a PAR Model 173 Potentiostat/Galvanostat, Model 174 Universal Programmer and Model 179 Digital Coulometer. Voltammograms were recorded on a Tektronix Model 564 CRO and photographed when permanent records were desired. Chronocoulometric determinations of the quantity of adsorbed reactants were performed by means of the computerized data acquisition and analysis system that has been previously described (6). 99 pH measurements were performed using a Beckman Model 76 Digital pH Meter. All measurements were made at 25±2° C. MATERIALS Stock solutions of cis-CrTEDTA(OH 2 ) sodium [transCrTEDTA] and CrEEDTA were prepared, purified and standardized as described in Chapter I. Spectral measurements were used to verify that the stock solutions were pure. Stock solutions of the chromium(III) complex with EDTA (ethylenedinitrilotetraacetic acid) were prepared by mixing stoichiometric quantities of Cr(Cl0 4 ) 3 and the tetraacid in water and heating for 1 hour. Supporting electrolyte solutions were prepared from triply distilled water and consisted of unbuffered 1 M NaC10 4 unless otherwise specified. The pH of each solution was adjusted after oxygen was removed by bubbling with prepurified nitrogen. 100 RESULTS CYCLIC VOLTAMMETRY Figure 1 shows a series of cyclic voltammograms for cis-CrTEDTA(OH2 ) at pH 4.0, 5.3 and 7.2. Depending on the pH, as many as three cathodic peaks appeared. As the pH increased, the peak currents of the two more cathodic waves increased at the expense of the first wave. On repetitive scanning at pH 7.2, the waves at -1.26 V and 1.40 V gradually disappeared as the wave at -1.04 V grew. If the electrode was maintained for 30 seconds at -1.60 V before initiating a (positive going) potential scan, the waves at -1.26 V and -1.40 V did not appear in the cathodic half of the cycle. Both CrEDTA(OH2) and CrEEDTA exhibited a single set of uncomplicated cathodic and anodic voltammetric waves under the conditions where cis-CrTEDTA(OH2) showed multiple cathodic waves. In order to verify the existence of the proposed acetate bridged dimer (cf. Chapter I, pp. 73 - 76), th e follo wi n g potential step experiment was performed on solutions 0.5, 1.0 and 2.5 mM in cis-CrTEDTA(OH2) at pH 7.3: The potential was stepped from some initial potential to -1.15 V, where only cis-CrTEDTA(OH 2) was reduced and to -1.50 V, where both the monomer and dimer were reduced. 101 FIGURE 1 Cyclic Voltarnrnetry of 1 rnM Solutions of Isomer I. Curve A, pH 4.0; Curve B, pH 5.3; Curve C, pH 7.2. Scan Rate: 50 v/s. 102 100 <[ ~ 0 rz w o::-100 0:: :=) 0 -. . V VS SCE -1.7 103 The slope of a plot of charge vs. k (time) 2 for the first step was proportional to the concentration of the monomer whereas, the difference in the slopes for the first and second steps was proportional to the dimer concentration. As the diffusion coefficient of the dimer was expected to be smaller than that of the monomer, both it and the dimerization constant, KD, were fit to the data. The results of this experiment which are presented in Table 1 strongly support the existence of the dimer. The peak current for the first reduction wave of cis-CrTEDTA(OH 2 ) is plotted against the square root of the scan rate in Figure 2 along with the calculated behavior for an uncomplicated diffusion limited electron transfer reaction. The increasing deviation between the observed and calculated behavior as the scan rate increases is in the direction expected for initially adsorbed reactants. Cyclic voltammograms for trans-CrTEDTA in Figure 3. are shown In contrast with the cis isomer, these voltammograms were unaltered as the pH was increased from pH 4 to 10. Two reduction waves were present on the first scan at -1.19 V and -1.30 V. A new reduction wave appeared at -1.04 Von the second cathodic scan and the height of the two original waves diminished. Upon repetitive scanning at a high scan rate ( e . g . 100 V 104 Table 1 Evaluation of the Dimerization of Cis-CrTEDTA(OH 2 ) at pH 7.3a Cone. Cm~!) Slope at -1. 15 V -!-,; - 2 s 2) (µC cm 0. 5 72.4 1.0 2.5 Slope at -1. 50 V -2 -!.:2 ( µC cm s ) % Monomer KD (~-1) 156 42.9 2470 100 273 33.9 2560 171 586 26.2 2200 - 1 Average: 2410±140 M 6 1 a The monomer diffusion coefficient was 6 x 10- cm 2 sThe dimer diffusion coefficient obtained from fittin g the data was 4.5 x 10- 6 cm 2 s- 1 , which is quite reason a bl e. 105 FIGURE 2 Scan rate dependence of the peak current of the first reduction wave of cis-CrTEDTA(OH 2 ): pH 4. 1 m~, The dashed line represents the behavior of an uncomplicated, diffusion limited electron transfer reaction. 106 • 280 • / / <i / ~ / / I- / z / w12 er:: 0::: / • :=) / / 0 / ~ •/ / <! w Q_ / / 40 / / /' 2 4 6 8 12 (SCAN RATE) / 10 107 FIGURE 3 Cyclic Voltammetry o f 1 rru-2 Solutions of Isomer II at pH 7. is the first scan. Successive scans at 100 v/s. Curve 1 108 2 <I: ~ ~ w 0:::: 0::: ~-200 0 -0.4 -1.0 V VS SCE -1.6 109 sec - l ) a steady-state voltammogram resulted in which the original pair of reduction waves was absent and only the wave at -1.04 V remained. If the electrode was maintained at -1.60 V for 30 seconds before initiating the (positivegoing) scan,only the wave at -1.10 V appeared during the cathodic half of the voltammogram. A series of cathodic voltammograms at increasingly higher sweep rates is shown in Figure 4. Note that the reduction wave at -1.19 V grows more rapidly than the wave at -1.30 Vas the scan rate is increased. This behavior is consistent with the first wave being assigned to the reduction of reactant which is adsorbed strongly enough to shift its reduction potential from that of the diffusing reactant (7). This interpretation is also supported by the concentration dependence of the voltammograms: Decreasing the concentration of the complex from 1 mM to 80 µ~ produced a proportional decrease in the peak current for the wave at -1.30 v; however, the peak current at -1.19 V decreased by only 20% (adequate time was allowed to assure adsorption equilibrium before the scan was initiated). ELECTROCAPILLARY CURVES The prominence of adsorption in the voltammetric behavior of both isomers of Cr(III)-TEDTA led us to 110 FIGURE 4 Cathodic voltammograms of a 1 m~ solution of Isomer II at pH 4 at increasingly faster scan rates: Curve 1, 2 v/s, 5 µA/cm; Curve 2, 10 v/s / 20µA/cm; Curve 3, SOv/s, SOµA/cm; Curve 4, lOOv/s, lOOµA/cm. 111 1- z w a: 0:: :::, 0 -0.4 -1,0 V VS SGE 112 examine the ·electrocapillary curves for solution s of each complex by measuring the natural drop time of a dme as a function of potential. The resulting electro- capillary curves are shown in Figure 5. The larger change 1n interfacial tension produced by the trans isomer is indicative of its stronger adsorption on mercury as is the relative shift in the position of the electrocapillary maximum (ecm): -0.55 V and -0.65 V for the cis and trans isomers, respectively. The electrocapillary curve for a 1 m~ solution of free TEDTA in 0.5 ~ NaC1O 4 at pH 4.2 was indistinguishable from the curve for 0.5 M NaC1O 4 alone. Thus, the free ligand is adsorbed much less strongly than either of the Cr(III) complexes. The lower adsorbability of the free ligand may reflect greater steric hindrance in exposing the sulfur atom to the electrode surface when the ligand is not wrapped back around the metal to which it is coordinated. DETAILED STUDY OF THE ADSORPTION OF CIS -CrT EDTA(OH 2) AND TRANS-CrTEDTA- ON MERCURY ELECTRODES Trans-CrTEDTA-. Double potential step chronocoulome t r y (8) served well as a method for the determination of the quantity of adsorbed trans-CrTEDTA~ The ratio of the slopes of the chronocoulometric plots (charge vs. the 113 FIGURE 5 Electrocapillary Curves for Isomers I and II. 1 M Na C1 0 '+ , (0) (I) ; 0 . 9 3 mM I s om er I , pH 3 . 9 0 , 0. 9 3 mM Isomer I , pH 7 . 2 0 , (4) ; 0. 2 8 mM Isomer II, pH 3.97, (0). 114 4.9 ••••••• • • • i• •••• ~~~.• i. • ii 0 w 4. (/) 0 0000 i 0 O o o o o<Xf 00 i .i w 2 □ i--- 0 0:: • .i □ .......... n. ~ A 4-.1 0 -0.3 -0.7 V VS SCE i • 115 square root of a time function) for the reductive and oxidative steps were in good accord with the valu e expected given the quantity of adsorbed complex,which had been evaluated from the intercepts of these -same plots ( 8) . This agreement demonstrated that the chromium(II) complex produced by the reduction of trans-CrTEDTA was not adsorbed on the electrode at the potential ( - 1.6 V) at which it was formed. Figures 6 and 7 display the concentration and potential dependences, respectively, of the adsorption of the trans isomer at pH 5. Especially noteworthy are the extremely low concentrations at which the complex is significantly adsorbed and the abrupt changes in coverage that result over a very narrow range of bulk concentration or electrode potential. of trans-CrTEDTA This feature of the adsorption is better revealed in Figure 8 where the data are plotted using Langmuir coordinates. In each case, the dashed line of unit slope represents a hypothetical system which obeys the Langmuir isotherm. The horizontal distance between the experimental points and the dashed line is a measure of the magnitude of the interactions between the adsorbed molecules. Positive deviations from Langmuir's isotherm signify attractive interactions and negative deviations, repulsive interactions between the adsorbed species. At potentials 116 FIGURE 6 Concentration Dependence of the Adsorption of Trans-CrTEDTA , pH 5. -.3 V Initial potential: ce) ; -. 5 V (II) ; - . 8 V (A) . 11 7 AMT. ADSORBED lµCCM-2 N ('\) • 0 0 0 z0 rn z -I .::0 ~ 0 z -.... 1:: IS: 0 0 0 118 FIGURE 7 Potential Dependence of the Adsorption. 0.027 rnM tr>ans-CrTEDTA , pH cis-CrTEDTA(OH 2 ) , pH 4 (I). 5 (II). 0.24 rnM 119 N1 z 0 0 ~ 15 0 UJ en er 0 (./) 0 <r ~ 2 <! 5 -.2 V VS SCE 120 FIGURE 8 Langmuir Plots for Trans-CrTEDTA Initial potential: Part c, - . 8 V. at pH 5. Part a, -.3 V; Part b, -.5 V; Dashed line represents a hypothetical system which obeys the Langmuir isotherm. 1 21 0 / / / LNJL / 0+ 1 / / / / -15 -II LN C / / b / / / / / / / • / LN..fr.. 01-1 / / • / / / / -3 - 15 -11 LN C / / C / / / / / / / / / / -11 LN C • 122 positive of the point of zero charge (pzc in 0.27 m~ was -0. 641 V.) where the electrode bears a net trans-CrTEDTA positive charge, the repulsiveness of the intermolecular interactions abruptly decreases over the very narrow range of bulk concentration (3 to 5 µ~) where full coverage is achieved (Figure 8a). At potentials near the pzc, this discontinuity is not evident in the Langmuir plot (Figure 8b). At potentials negative of the pzc, the interactions are so extremely repulsive at low bulk concentrations that the apparent reduction in these forces which accompanies the sudden increase in surface coverage(between 3 and 5 µ!i:) is reflected in a slight discontinuity in Figure 8c. These adsorption data could not be fit to a single, continuous coverage independent Frumkin isotherm (accounts for interactions between the adsorbed molecules) (9). Enhancement of the ligand field at the nitrogen atom in CrNCS +2 on binding Hg +2 . has been demonstrated (10) and attributed to increased chromium to ligand bonding (11). Similar arguments have TT- been advanced to explain the stronger adsorption of cisCr(OH 2 )~(NCS) 2 + than of free SCN at comparable concentrations on positively charged mercury surfaces (12). We believe that back donation is responsible for the marked preference of trans-CrTEDTA for positive electrodes 123 shown in Figwre 7. Greater adsorption at constant composition at more positive electrode potentials requir es that th e e lectrod e char ge increase at constant potential ( 13). Consequently th e electronic charge densities in the presence and absence of trans -CrT EDTA were evaluated. Th e total charge flowin g into a dm e during the gTowth of each drop was measured and the faradaic and nonfaradaic components separated accordfng to previously described pr ocedures (14). are plotted in Figure 9. The results The pzc shifted from -0.492 V to -0.641 Vin accordance with the greater positive charge on the electrode in the presence of trans-CrTEDTA Since the adsorption independent at pzc esse ntiall y potentials Figure (cf. is th e 7), th ese electrode at slightly affected positive th e of charge the on th e is onl y presence of potentials by potential the complex. The chronocoulometric plots of char ge vs. for trans -CrTEDTA at concentrations than 3 µM were curved. (time) ½ greater The curvature was most pronounced ' at the most positive potentials and decreased as the bulk concentration increased. The·initial slope was larger than the diffusion controlled slope and approached it after ca. 5 msec. Such behavior has been previously 124 FIGURE 9 Electrode Charge-Potential Curves. Curve A, 1 M NaC1O 4 , pH 4; Curve B, lrn~ cis-CrTEDTA(OH 2 ) , pH 4.15; Curve C, 0.27 rnM t~an s-CrTEDTA , pH 3.89. 125 8 ~0 0 -...::t w 0 (9 0::: <t :r: Q -s -0.2 -0.6 V VS SCE 126 obs e rved in th es e laboratories and was cau se d by slow reduction of the adsorbed comple x (15). To determine if t r ans -Cr(III)T EDTA rapidly desorbed is reduced, at its desorption was examined. Using in at complex -1.6 V a from -0.1 V reactant was initially where significant faradaic reaction -1. 5 V, where a·ttached and The solution 27 potential -0.5 V, where de s orption, occurred to data the of -1.1 V, but no then both to the was fast. obtained were anal y zed treatment of were The with instantaneous desorption data Osteryoung at consi s tent -1.1 V by the diffusion-controlled a new adsorption approximately 60 % of the initially of the reactant Christie (16). ment was to and and followed l-!M the adsorbed, diffusing it -1.1 V reduction charge-time according or before at pH 6.3, stepped is establish- equilibrium. Since adsorbed reactant was desorbed at - 1.1 V, it highly probable that all of complex is rapidly desorbed at -1.6 V . the is 127 Cis-CrTEDTA(OH 2). Do uble potential step chrono- coulometry was not useful for the evaluation of the quantity of adsorbed cis-Cr(III)TEDTA(OH 2 ) at pH 4 because the resulting chromium(II) complex appeared to remain adsorbed throughout the duration of the reductive step. This adsorption of the electrode reaction product was evident from the intercept and slope of the chronocoulometric plots for the oxidation step which were too high and too low, respectively. It was ascertained that the persistence of the chromium(II) complex on the surface was not the result of its spontaneous adsorption at equilibrium by performing a chronocoulometric experiment with the initial electrode potential maintained at a value (-1.4 V) where the chromic complex was reduced and then stepping the potential to a more positive value (-0.5 V) where the chromous complex was oxidized at a diffusion controlled rate. k (charge vs. (time) 2 ) Chronocoulometric plots for such experiments had intercepts which were much smaller than those for the oxidative half of a double potential step experiment and were, in fact, within ten percent of the double layer charging blank for the pure supporting·electrolyte. Therefore, although the chromium(II) complex produced by reduction of cis-CrTEDTA(OH 2 ) at pH 4 is not adsorbed at equilibrium at -1.4 ½ its rate of desorption is somewhat slow. 128 The slow desorption of the chromous complex forced us to use the difference in the intercepts of the chronocoulometric plots in the presence and absence of cisCrTEDTA(OH 2 ) to estimate its adsorption. Values for the apparent adsorption obtained in this way are plotted in Figure 10. The cis isomer seems'to be much more weakly adsorbed than the trans isomer and the maximum obtainable coverage appears to be a rather sensitive function of the electrode potential, at least at the highest bulk concentration employed (2 ~). However, there is a discrepancy between the apparent potential dependence of the adsorption shown in Figure 7 and the slope of the electrocapillary curve for the c~s isomer shown in Figure 5. If the adsorption really increased as the electrode potential was made more cathodic, it is required by thermodynamics that the electrode charge become more negative (13). However, the slope of the electrocapillary curve for the cis isomer at any potential, which is proportional to the electrode charge, is less negative than that of the blank. To corroborate the electrocapillary behavior, the electrode charge was measured as a function of potential in a solution 1 mM in cis-CrTEDTA(OH 2 ) at pH 4.1 as described above. The data are plotted in Figure 9 and confirm the conclusions reached by inspection of the electrocapillary curves. 129 FIGURE 10 Concentration Dependence of the Apparent Adsorption of Cis -CrTEDTA(OH 2 ), pH 4. potential: Initial -.3 V (e); -.5 V (II); -.8 V (4). 130 f-! 2 <r 4 10 1000 CONCENTRATION I µM 131 The apparent adsorption of cis-CrTEDTA(OH 2 ) was also measured at pH 7. in Table 2. Representative data are listed All of the adsorption observed at pH 7 can be explained in terms of the monomer alone at total Cr concentrations below 0.05 mM. However, at higher concentrations, a signifi c ant fraction of the cis isomer is dimerized. As the scan rate dependence of the dimer waves indicates that the dimer is adsorbed, both the monomer and dimer contribute to the total quantity of adsorbed complex at pH 7. At pH 7, the uncoordinated carboxyl group is deprotonated and the originally neutral ci s complex is now anionic. As the data in Table 2 show, these complexes are less strongly adsorbed than the uncharged monomer at all potentials, but the greatest difference is observed at the most negative potentials as would be expected for anionic adsorbates. = - 1 2410 M 0.18 0.87 1. 4 --- 2.4 2.4 b Quantities of adsorbed compl e x r e port e d ar e for bulk con ce ntr a t i ons o f the cis monom e r which a r e as close as possible to those calculated to be present, e . g . for 0.53 mM, the values reported are those measured at 0.47 mM. - Calculated using KD 0.21 0.12 0. 2 7 0.21 0.17 0.0090 0.00019 0.0094 0.65 0.45 0.32 0.32 0.24 0.022 0.0011 --- - -- 0.92 0.79 1. 6 1. 7 1. 2 1. 3 0 . 75 0.024 ) 1. 2 1. 3 1. 4 2 Monomer (pH 4)b -3 V -5 V -8 V (V v s . SCE) xmol cm- 0.54 0.47 0.44 0.040 0.0038 0.047 a 1 0 0.59 0.57 0.43 0.070 0.012 0.094 0.61 0.55 0.14 0.049 0.24 0.90 0.82 0. 2 2 0.12 0.47 1.1 0.96 0. 35 0.30 0.94 on Mercury (10 Acetate Bridged Dimer (pH 7) -.3 V -.5 V -.8 V (V vs . SCE) ) 1. 2 a 2 1.1 0.53 0.67 1. 88 Monomer Cone. (mM) Dimer Cone.a (mM) [Cr] (m~) T ~pparent Adsorption of c is-CrTEDTA(OH Table 2 f---1 N l,-1 133 DISCUSSION CYCLIC VOLTAMMETRY The cyclic voltammetric behavior of the two isomers shown in Figures 1 and 3 can be explained by means of the combination of electrode and chemical reactions summarized 1n Scheme 1. In Figure 1 1 the reversible one-electron wave at -1.04 V results from the reduction of cis-CrTEDTA(OH 2 ) and is the only wave observed at pH 4, where the complex exists in its neutral, monomeric form. At pH 7, the two other one-electron waves at -1.26 V and -1.40 V, which are of equal size, represent the reduction of the acetate-bridged dimer. The electrochemical nonequivalence of the two Cr(IIi) centers is a reflection of the difference in their coordination environments. By contrast, the symmetrical dimer, (EDT A- Fe ( II I) ) 2 0 - l+ , e xh i b its on 1 y a singl~ two-electron wave (17). The observation of a wave for each Cr(III) also dictates the order of their reduction. The metal donating the bridging acetate group must be reduced fir3t. If the other Cr(III) center were reduced first, the bridging O-Cr bond would be labilized, the dimer would dissociate and only two cathodic waves would have been observed -- one for the monomer and a single, one-electron wave for the dimer. Moreover, as the coordination environ- ,,,,,,o .<4 _,, -sv -~v-<a rD;\ .,Cr ~ S p:#: OH 1/ I ~ S "---NV [cr 11 ~ o/"-o s~N ~o 0 + E p = E ~ E p = = v,,, \a ~IL- 0 E 'o - .-( p D 1ofu ~Cr 4lrr-:::::=t01lz Hif "'O 'N~OH2 R aw I +H+ ~ fast 0/ ~ ~o /IN~~ '1 ~~ / 26 V - 0. 98 V -1.04 ➔ = -1.30 V _ P -1. e + e- +e- E - 2 Scheme 1 ~ o7\o S~N ....- 1 fl.~r I / _ "I ~Cr "- s07 +>- 0s o "a N~N-0~7 > (i;cr "-/ f---1 vJ ;°t~<" (adsorbed) Ep = -1.40 V t { > O H , = -1.19 V 135 ment of the metal donating the bridging group is less affected by the formation of the dimer, it is reasonable that its reduction potential would be shifted less from the monomer potential. Dissociation of the dimer follows reduction of the second metal center as shown by the gradual growth of the monomer wave and disappearance of the dimer waves on successive scans. Reduction of either adsorbed or diffusing transCr(III)-TEDTA produces trans-Cr(II)TEDTA - 2 which rapidly relaxes to cis-Cr(II)TEDTA(OH 2 )-, as evidenced by the anodic wave at -0.98 V and the appearance of a new cathodic wave at -1.04 Von the second scan (Figure 3). greater stability of cis-Cr(II)TEDTA(OH2) The is demonstrated by the fact that after the electrode was held at -1.6 V for 30 s., only cis-Cr(III)TEDTA(OH 2) was reduced during the cathodic half of the sweep. The concentration and scan rate dependence of the peaks at -1.19 V and -1.30 Vindicate that the first wave originates from the reduction of the strongly adsorbed complex and the second, from reduction of the diffusing complex. The reduction of attached complex at a less negative potential than the diffusing complex is attributed to the strength of the Hg-S bond and consequent weakening of the S-Cr bon~ facilitating the stretching of the Cr-S bond to reach the transition state. 136 ADSORPTION BEHAVIOR The electrocapillary curve for ci s-CrTEDTA(OH 2) at pH 4 (Figure 5) indicates that the depression of the interfacial tension in the presence of the adsorbed complex was the greatest near the pzc (-0 . 486 Vin 1 m~ ci s-CrTEDTA(OH 2 )). Moreover, the electrode charge measurements show that the double layer capacitance is depressed at all potentials by the adsorption of the c~s isomer. These results suggest that ci s-CrTEDTA(OH 2) is behaving as an organic adsorbate ( 1 8) and that th e ap p ar ent in crease in th e a ds orption of cis-CrTEDTA(OH 2) at more cathodic potentials is misl ea d i n g . The lack of adsorption by CrEEDTA demonstrates the complex is attached to the electrode through the sulfur. The accessibility of the sulfur atom to the electrode is enhanced by the coordination of the ligand to Cr(III). Therefore, the reason that free TEDTA is not adsorbed is probably because of the steric difficulty of exposing the sulfur moiety to the mercury electrode. However, the role of Cr(III) in inducing the adsorption of TEDTA on mercury is more complicated since the La(III) and Zn(II) complexes of TEDTA were not adsorbed. At pH 7, where part of the cis co mplex exists as the acetate bridged dimer, both the monomer and dimer contribute to the total apparent adsorption (Table 2). The scan rate dependence of the dimer waves is consistent with its adsorption. The much weaker adsorption of these complexes 137 at the more cathodic potentials at pH 7 results from the electrostatic repulsion of the negative ions from the negatively charged electrode. The adsorption of trans-CrTEDTA on mercury electrodes is extraordinarily strong and comparable to that of Cr(NCS) 6 - 3 , which it is believed forms three bonds to the electrode (19). This isomer exhibits a marked preference for positive electro des (Figure 7) which probably arises from the capacity of the thioether sulfur to TT-accept d electrons (20,21) from Cr(III) and a-donate electron density to the electrode. The resulting increase in the ligand field stabilization energy which accompanies this shift in electron density is the driving f orce for the adsorption of the trans isomer (lb, 12). Clearly, this mechanism would operate most effectively at positively charged electrodes. The over abrupt change in very narrow range the (Figure concentration the onset adsorption of indicated lS strongly by the 6) - coverage observed . of bulk indicates that cooperative. attractive discontinuity The sudden interactions in the lS Langmuir plot 138 (Figure 8a). At -.3 V, the cooperative mechanism apparently operates very effectively and saturation is almost achieved at 5 µ~ (Figure 6). At -.5 V where the electrode is less positively charged, back bonding is less efficient, the electrostatic repulsion between the negatively charged adsorbed molecules is more important and higher bulk concentrations are required to achieve the same surface concentration. When the electrode is negatively charged (at -.8 V), back bonding is least effective and the electrostatic repulsion of trans-CrTEDTA from the electrode strongly interferes with its adsorption. To summarize, the lack of adsorption of CrEEDTA-, in which the sulfur atom of TEDTA is replaced by oxygen, demonstrates that the sulfur is the site of attachment of the Cr(III)-TEDTA complexes to the electrode. Steric hindrance in exposing the sulfur moiety to the electrode probably prevents free TEDTA from adsorbing on mercury. Although these steric problems are greatly reduced by the coordination of ligand around a metal ion, the nonadsorption of the La(III) and Zn(II) complexes of TEDTA shows that additional factors are involved in the adsorption of cis-CrTEDTA(OH 2 ) on mercury. Back bonding by Cr( I II) enhances its ligand field stabilization energy and is responsible for the marked preference of trans-CrTEDTA 139 for positive electrodes. A cooperative adsorption mechanism accounts for the abrupt charge in coverage of the electrode by trans-CrTEDTA concentrations. over a very narrow range of bulk 140 REFERENCES (1) a) F. C. Anson, J. H. Christie and R. A. Osteryoung, J. Electroanal. Chern., 13, 343 (1967); b) S. N. Frank and F. C.Anson, J. Electroanal. Chern., ~, 55 (1974). (2) D. J. Barclay and F. C. Anson, J. Electroanal Chern., ~' 71 (1970). (3) A. P.Brown, C. Koval and F. C. Anson, J. Electroanal. Chern., ?_J_, 379 (1976). (4) a) A. P. Brown and F. C. Anson, J. Electroanal. Chern., 83, 203 (1977); b) C. A. Koval and F. C. Anson, Anal. Chern., ~' 223 (1978). (5) P. J. Peerce and F. C. Anson, Anal. Chern., 49, 1270 (1977). (6) G. Lauer, R. Abel and F. C. Anson, Anal. Chern.,~' 765 (1967). (7) R.H. Wopschall and I. Shain, Anal. Chern.,~, 1514 (1967). (8) J. H. Christie, R. A. Osteryoung and F. C. Anson, J. Electroanal. Chern., 1:._l, 236 (1967). (9) A. N. Frumkin, Z. Phys. Chern., 116, 466 (1925). (10) C. E. Schiffer, Spec. Publ. No. 13, The Chemical Society, London (1959), p. 153, (11) J. R. Wasson and C. Trapp, J. Inorg. Nucl. Chem., lQ_, 2437 (1968). (12) D. J. Barclay, E. Passeron and F. C. Anson, Inorg. Chem., 9, 1024 (1970). (13) P. Delah~y, Double Layer and Electrode Kinetics, Interscience , Publ., Inc., N.Y. (1965). (14) G. Lauer and R. A. Osteryoung, Anal. Chem., ~, 1866 (1967). ( 15) F. C. Anson and R. S. Rodgers, J. Electroanal. Chem., .!Z_, 287 (1973). (16) R. A. Osteryoung and J. H. Christie, J. Phys. Chem., 71, 1348 (1967). 141 (17) H. J. Schugar, A. T. Hubbard, F. C. Anson and H. B. Gray, J. Am. Chern. Soc., ~' 71 (1969). (18) B. B. Damaskin, 0. A. Petrii and V. V. Batrokov, Adsor tion of Or anic Com ounds on Electrodes, Plenum Press, N.Y. 1971 . {19) M. J. Weaver and F. C. Anson, J. Electroanal. Chem. , §__Q_' 19 (1975). (20) R. J. P. Williams, Ann. Re:eorts Progr. Chern. , 87 (1959). (21) s. E. Livingstone, Quart. Rev. ~. (London), !2_, 386 (1965). 142 CHAPTER III DIGITAL DEVICE FOR PRECISE DETERMINATION OF DROP TIMES AT DROPPING MERCURY ELECTRODES* Pamela J. Peerce and Fred C. Anson From Arthur A. Noyes Laboratory, t California Institut e of Technology, Pasadena, CA 91125 * Anal. Chem., ±2._, 1270 (1977). t Contribution No. 5526 143 INTRODUCTION Thermodynamic information on the adsorption of ions and molecules at mercury electrodes is often obtained from electrocapillary curves (1). The necessary values of the interfacial tension can be determined directly by means of a capillary electrome te r or evaluated from measurements of th e natural drop time of a dropping mercury electrode (2). The tedious experimental aspects of the former approach are well known. The latter procedure is more attractive because it is experimentally simpler and readily automated (2-6) Several methods for detecting drop fall h ave been proposed previously (5). The two most common employ photoelectric or impedance measurements to detect the end of drop life. Corbusier and Gierst (2) were among the first to suggest the photoelectric method, which requires that the capillary be positioned very precisely in the light beam. However, their device and several of those based on impedance measurements (3,4) do not allow the lifetimes of consecutive drops to be measured. In this note, a simple, digital timer is described which provides a rapid, convenient means for determining the lifetimes of up to 1024 consecutive drops. The potential of the electrode can be controlled by a standard, commercial pola~ograph. (A Princeton Applied Research Model 174 Polarographic Analyzer was utilized, but any polarograph having a current measuring circuit with an output which can be adjusted for d.c. level and polarity would be acceptable.) 144 The e xperimental arrangement for measuring drop times is the same as that for recording a polarogram. The current output of the polarograph is connected directly to the input terminal of the drop timer. Since the device can be set to time up to 1024 consecutive drops at each potential, drop lifetimes can be determined very accurately. Once started, the timer runs continuously without operator intervention and displays the number of clock pulses elapsed during the last run while performing the next one. This permits the operator to record the result of each run while the next run is in progress. The digital output of the timer is well suited for interfacing to a computer if even greater automation were desired. 145 EXPERIMENTAL DESCRIPTION OF THE INSTRUMENT The drop timer is shown schematically in Figure 1. The abrupt change in current which accompanies the detachment of each drop from the capillary is used to sense the end of drop life. The input signal to the timer is sharpened by a Schmitt trigger, ICl, which is set to fire when the negative edge of the signal at its input crosses the threshold level of +2.6 V. (Since the timer accepts only positive input voltages, it is necessary to reverse the polarity of the current output when the current changes sign). The clock frequency is 120 hz. The unrectified, 60 hz signals from the transformer secondary are differentiated and fed to separate inputs of the Schmitt trigger, IC2. Normally, as one input is high while the other is low, the exclusive OR output is high. However, just before and after the voltage reverses direction (twice per cycle), both inputs are low causing the exclusive OR output to change state, thus generating the clock pulses. The inputs of both ICl and IC2 are protected by Zener diodes (5 volt). An experiment is initiated by depressing the START button sometime during the growth of a drop. This clears 146 FIGURE 1 Schematic Diagram of Drop Timer. All integrated circuits are Motorola CMOS devices. ICl and IC2 are dual Schmitt triggers, MC14583; IC3 is a programmable timer, MC14536; IC4 is a quad 2-input NAND gate, MC14011; res is a dual monostable multivibrator, MC14528; IC6 and IC7 are three-digit BCD counters, MC14553; IC8 is a BCD counter, MC14510. Not shown are the latch/decoder/drivers, MC14511, and the 7-segment LED's MAN-74, which are conventionally wired. 147 F' I LTER ON/O FF IOK 15 0 1( To Powe r To Polorooroph _ _ _ _.. I 2j,£, Supply IC I e,.>----- ~ IC2 0 0 120 hz IC6 I C3 Scanner h.-.J~----~=:---=--,----7 o,, Multiple Overf lo w Reul C l ock Di sable } To ( 3 ) PNP Trans i stors } To o L otch / Oecodtr / Or i ver and {3) 7 • StQtntnl L ED 's } To 0 LO!Ch /O tc od er /O r i vt r and (3) 7 • StQmtnt LED ' s Lotcri CLOCK PULSE COUNTERS IC7 Rese t C l ock Oi soblt Latch ICB IC5 ' - - - - - - - < Rent 0 EXP T a, 0, '---1-~ii, a, c, COU NTER A, t - - t - - - - - - - - - - - - - - - ~ C1oc "- To o Lotch / Decoder / Oriver } and ( I ) 7-Segment LEO 148 IC8 and when th e drop falls, IC3, IC6 and IC7 are reset, IC8 is incremented by one and the counting begins. When the preset number of drops has been counted, the output of IC3 undergoes a brief negative transition which triggers the firing of ICS, thus disabling IC6 and IC7 and updating the two digital readouts which display the number of clock pulses counted and the number of times the measurement will have been repeated at the conclusion of the current run. The relative timing of these events is shown in Figure 2. The execution time of the device is ca. 10-15 µsec which represents a negligible correction for electrodes with natural drop times of several seconds. The low pass filter shown in Figure 2 was employed to reduce 60 hz noise (RC= 0.3 sec). The presence of the filter caused the apparent lifetimes of each drop to increase by a constant amount (ca. 0.02 sec). 149 FIGURE 2 Timing Diagram. The purpose 0£ this figure is to show the relative timing of the indicated events. However, the time scales for the various events are not necessarily the same. For example, the clock pulses (IC2) are separated by 8.33 msec, whereas the pulses from the exclusive OR output of ICl occur only every 3-7 sec. 150 Clock IC 2 ©Output PAR 174 Current Output IC 1 (±)Output START Exp't Counter IC 8: Reset Count I I ~-----------------L---- 11 ,-,- ~ - - - - - 1 - - - - - - - - - 2 1,0, Drop Counter IC 3: Set Output : : I I I I I I I I I LJ LJ I 1 I I Count Clock Pulse Counters IC6 and IC 7: Disable Latch Moster Reset I I ,______ 8 drops----------8-- 151 RESULTS AND DISCUSSION CHARACTERISTICS AND PROPERTIES OF THE TIMER As the clock frequency employed in this device is 120 hz, the uncertainty in the lifetime of a single 5 sec. drop is ca. 0.2%. Such precision in the measurement of individual drop times is comparable to that reported by Corbusier and Gierst (2) but not as high as has been achieved with some of the previously described devices (3,5,6). However, when the lifetimes of many drops are averaged, the accuracy and precision of the present timer are very high indeed. A precision time mark generator (Tektronix Model 180 A) was used as the input source to test the inherent accuracy and precision of the drop timer. For marks separated by 5 sec. and counting 8 marks per run, the interval could be measured to within 0.2%. In twenty replicate experiments a precision of ±0.008% was observed. The time mark generator was also used to test the ability of the timer to measure short time intervals. For marks exactly 0.01 sec. apart, the timer yielded an interval of 0.010006 sec. after counting 512 marks. ELECTROCAPILLARY DATA The timer was used to measure the electrocapillary curve for 0.5 M NaC1O 4 (pH 4.2) at 25° C. At least three 152 runs consisting of the measurement of eight drop times were performed at each potential. Even near the pzc the drop time could be measured with a precision better than ±0.1% by using the most sensitive current range on the PAR 1 7 4 ( 7) . The data are plotted in Figure 3 along with the surface tension data of Parsons and Payne for aqueous 0.5426 M HC1O4 at 25° C (8). The agreement between the two sets of data is excellent. Thus, the greater convenience of the drop timer described herein has been achieved without sacrificing the quality of the electrocapillary data obtained. based on drop time Of course, all methods measurements share the limitation that adsorption equilibrium must be reached rapidly for the drop time to be a reliable measure of the equilibrium interfacial tension. ACKNOWLEDGEMENTS Charles Klopfenstein provided much valuable advice in the design of the digital circuit. Helpful discussions with Irving Moskovitz are also acknowledged. CREDIT This work was supported by the National Science Foundation. 153 FIGURE 3 Comparison of electrocapillary data obtained from differential capacitance and drop time measurements. (I) - This work; (O) - Taken from Parsons and Payne (8). The two data sets were superimposed at the electrocapillary maximum. 154 5.50 0 0 Q) 0 ~ • 400 0 • I 0 Q) .E 5.00 E • to. 0 Q) C 0 >- • ~ 0 ""O ~ • • 350 • 4.50 • 4.oo L L L _ _ L _ J L . _ L~ !:-;~__L___L_-1.,~o~oo~___.__---'--___.____, 50 0 0 -E /mV vs. SCE 310 155 REFERENCES (1) D. C. Grahame, Chem. Rev., 41, 441 (1947); D. M. Mohilner in ElectroanalyticaT Chemsitry, A Series of Advances, V. 1, ed. A. J. Bard, Marcel Dekker, Inc., New York (1966). (2) P. Corbusier and L. Gierst, Anal. Chim. Acta, _!i, 254 (1956). (3) L. Meites and J.M. Sturtevant, Anal. Chem.,~' 1183 (1951). (4) A. J. Bard and H.B. Herman, Anal. Chem.,~' 317 (1965). (5) B. Nyg~rd, E. Johanss on and J. Olofsson, J. Electroanal. Chem., g, 564 (1966). (6) G. Papeschi, M. Costa and S. Bordi, Electrochim. Act a , ~, 2 0 15 (19 7 0) . (7) Triggering of the drop counter requires that the negative edge of the input signal cross +2.6 V. By careful adjustment of the d.c. offset in the current output of the polarograph,this requirement can usually be satisfied despite the smaller magnitude of currents flowing near the pzc. However, problems can occur under conditions where the current-time curves contain maxima which cause the input signal to cross the triggering level prematurely. (8) R. Parsons and R. Payne, Z. Phys. Chem. N.F., ~' 9 (1975).