Metal complexes of cyclic tetra-azatetra-acetic acids

July 9, 2017 | Autor: Rita Delgado | Categoría: Analytical Chemistry, Metal Complexes, Acetic Acid
Share Embed


Descripción

Ta/alita, Vol. 29, pp. 815 to 822, 1982 Printed in Great Britain. Ali rights reserved

0039-9140;82,1 100815-08$03.00;0 Copyright © 1982 Perga mon Press Ltd

METAL COMPLEXES OF CYCLIC TETRA-AZATETRA-ACETIC ACIDS RITA DELGADO and J. J. R. FRAÛSTO DA SILVA Centro de Quimica Estrutural, Instituto Superior Técnico, Lisbon, Portugal (Receil'ed 5 February, 1982. Accepted Il May 1982)

Summary--The cyclic tetra-aza complexones cOOT A ([12]ane N 4 · 4ac), cTRIT A ([13]ane N 4' 4ac) and cTET A ([ 14]ane N4 ' 4ac) have been synthesized and characterized by elemental analysis, titration, melting-point determination and NMR (and infrared) spectroscopy. The ionization constants and the stability constants of the MH 2 L, MHL and ML complexes formed with al kali, alkaline-earth and sorne transition metals were determined at 25.0 ± 0.1 0 and ionic strength O.IOM [KN0 3 and (CH3)4NN03]. It was confirmed that cOOTA forms the most stable Ca 2 + and Sr 2 + complexes but the reported inversion of the order of stability of the complexes of these two ions with cTRIT A was not confirmed. Also, the values of the stability constants determined in this work differ substantially from those previously reported for ML species. cOOT A is an interesting alternative to classical non-cyclic complexones for the complexometric determination of Ca 2 + and Mg 2 + but neither this ligand nor the other two offer advantages over EOT A or OCT A for the complexometric titration of transition metals. Cyclic polyoxa and polyaza ligands which form stable complexes with a va ri et y of metal ions are of current interest either because of their usefulness in new processes of organic synthesis or as models for the study of certain biological reactions or systems, The effects of cavity size and of conformation on the spectroscopic and magnetic properties of the metal complexes, on the kinetics of complex formation and on the thermodynamic stability of the speci es formed have also been studied by man y inorganic chemists but, so far, few analytical chemists have been concerned with such ligands as possible reagents for complexometric titrations. Indeed, the macro cycles that have been studied are not suitable for this purpose, but derivatives of them might be of interest. For example, the polyaminocarboxylic acids derived from cyclic polyaza ligands should be relatively soluble in water, and should form quite stable complexes with metal ions. Very few such derivatives have been reported in the literature; references have been found to a dipropionic acid derivative of a diazatetraoxa macrocycle,l to the tetra-acetic acid derivatives of three tetra-aza macrocycles 2 and to the tetra-acetic acid derivative of a dimethyltetra-aza macrocycle. 3 In the tirst two cases, the authors were indeed concerned with the determination of stability constants of alkaline-earth and transition metal complexes; the third case, however, is a preparative and spectroscopic study of the cobalt, nickel, copper and zinc species formed with the particular ligand chosen. The dipropionic acid derivative mentioned above 1 forms only very weak complexes and is not of special analytical interest; in contrast, the tetra-acetic acid derivatives of tetra-aza macro cycles synthesized by Stetter and Frank 2 form very stable complexes and one of them (a) was reported to form the most stable Ca 2 + complex in aqueous solution known to date.

For a second ligand of this group (b) the stability constant of the strontium complex was reported 2 to be higher than that of the calcium complex by a factor of almost 104, which would be a quite exceptional result for classical non-cyclic complexones, but possible, at least in principle, for derivatives of macrocycles with the correct cavity size. Stetter and Frank reported later 2a that they were unable to contirm their unusual initial results. On the other hand, the stability constants of the complexes formed with the transition-metal ions are high but not exceptional. HOOC-CH

2

( "'N] ,

HOOC-CH ...... 2

am

(CH)

N,

CH COOH

...... N

(CH) m

2

'CH COOH

2

= 2, n = 2:

1,4,7,10- tetra -azacyclododecane- N, N', N", N'''- tetraacetic acid ([12]ane N 4 · 4ac) or cOOTA bm

=

2, Il = 3: 1,4,7,1 O-tetra-azacyclotridecane- N,N',N",N'" -tetra-acetic acid ([13]ane N 4'4ac) or cTRITA

cm

= 3, n = 3: 1,4,8,11- tetra -azacyclotetradecane- N,N', N", N'" - tetraacetic acid ([14]ane N 4'4ac) or eTETA

These reports led us to undertake the synthesis of the ligands to allow more detailed s.tudy of the species formed in aqueous solution, the thermodynamic data for the reactions with metal ions and their suitability as reagents for complexometric titration, In the present paper we report the synthesis and analysis of the ligands, and give values for the stability constants of the MH 2L, MHL and ML species formed with various metal ions. Our values ditfer, sometimes quite considerably, from those reported by Stetter and Frank: 2 in particular, no inversion of the normal 815

RITA DELGADO and J. 1. R. FRAUSTo DA SILVA

816

a

order of stability of the calcium and strontium complexes of b was found. The trend for the three cyclic complexones is the

b

same as for the non-cyclic classical ones. EXPERIMENTAL

Synthesis and characterization of the ligands The cyclic tetra-azatetra-acetic acids were prepared by reaction of the corresponding cyclic amines with chloroacetic acid in aqueous alkaline solution. The cyclic tetramines [12]ane N 4 and [13]ane N 4 were synthesized as described in the literature;4,5 [14]ane N 4 was obtained from a commercial source (Strem Chemicals), The method of preparation of the cyclic tetramines involved the condensation of a linear tosylated amine with a tosylated diol at 100-120°, in dry dimethylformamide as solvent. Tosylation of the amines and of the diols was do ne as described in the literature,6,7 but the yield from the diols, particularly from 1,3-propanediol, was rather low (40-60%). Hydrolysis of the tosyl groups in the cyclic tetramines was better achieved by refluxing with a mixture of glacial acetic acid with 48% hydrobromic acid (9: 16 v/v) for 48-60 hr. The condensation of the amines with chloroacetic acid offered no difficulties but the pH had to be kept below 10 and the temperature between 40 and 60°. The reaction mixtures were acidified to pH 2 with hydrochloric acid, and the products (cDOT A and cTET A) crystallized as sm ail bright colourless needles on standing overnight in the refrigerator; the crystallization of cTRITA required the previous separation of the potassium chloride present. Ail products were recrystallized from water. Both cDOT A and cTRIT A crystallized with two moles of potassium chloride per mole of product and recrystallization did not change their composition. cTET A was obtained free from potassium chloride. Ail intermediate and final products were characterized by elemental analysis, melting points, infrared and/or NMR spectra, and by the titration curves in the case of the tetra-acetic acids. cDOTA (with 2 KCl): m.p. 298 (dec,); m,w. (titration) 553.5. Found: C 34.7%, H 5.4%, N 10.3%; required: C 34.70%, H 5.06%, N 10.12%. Proton NMR spectrum (solvent D 2 0, reference DTSS, pD = 4.50), Fig. I(a): /j 3.683 (8 H, acetate groups), /j 3.281 (16 IXCH 2 ring protons). cTRITA (with 2KCI): m.p. 242 (dec,); m,w. (titration) 567.6. Found: C 36.1%, H 5.6%, N 9.9%; required: C 35.97%, H 5.33%, N 9.87%. Proton NMR spectrum (solvent D 2 0, reference DTSS, pD = 3.05), Fig. I(b): /j 3.71, 3.67 (2 singlets, 8 H, acetate groups), /j 3.33 (multiplet, 16 IXCH 2 ring protons), /j 2.05 (quintuplet, 2 fJCH 2 ring protons). cTETA: m.p. 313 0 (dec.); m.w. (titration) 432.5. Found C 49.7%, H 7.3%, N 12.9%; required C 50.00%, H 7.40%, N 13.00%. Proton NMR spectrum (solvent D 2 0, reference DTSS, pD = 3.40), Fig. I(c): /j 3,475 (8 H, acetate groups a), /j 3.210 (singlet, 8 IXCH 2 ring protons b), /j 3,142 (triplet, 8 IXCH 2 ring protons cl, /j 1.924 (quintuplet, 4 fJCH 2 ring protons dl.

c

(c)

(b)

(a)

2

4

8

ppm vs. DTSS

Fig. 1. Proton NMR spectra of (a) cDOT A at pD 4.5; (b) cTRITA at pD 3.05; (c) cTETA at pD 3.40.

0

0

Reagents Metal salts, Metal nitrates of analytical reagent grade were used and solutions were prepared in demineralized water and standardized by EDTA titration or gravimetry (Be). The ionic strength was adjusted with solutions of potassium or tetramethylammonium nitrate (prepared from tetramethylammonium hydroxide and ni tric acid and recrystallized twice from 80% ethanol). Carbonate-free potassium and tetramethylammonium hydroxides. Carbonate-free solutions of these titrants were

prepared according to Schwarzenbach and Biederman,8 un der purified nitrogen. The solutions (ca. 0.050M) were standardized by titration with O.OIM hydrochloric acid. Carbonate was tested for regularly 9 and the solutions were discarded when the concentration reached 0.5% of the hydroxide concentration.

Potentiometric titrations The experimental set-up has been described previously;10 a Radiometer pHM 4 measuring instrument was used together with a Radiometer G 202 B glass electrode and a K 401 saturated calomel reference. The temperature was controlled at 25,0 ± 0.1 0 by circulating water through the jacketed titration cell, The ionic strength was kept to O.lOM by use of potassium nitrate or tetramethylammonium nitrate as background salts; the ionic product of water in these media was taken as 1.68 x 10 - 14.1 The glass electrode was calibrated in terms of [H +] by titrating solutions of hydrochloric acid and potassium hydroxide of known concentrations and correlating the m V readings with ca\culated values of [H +]. The [H +]-dependent junction potentials were found to be negligible (from Gran plots 9 ) and the correlation between measured e.m.f. and calculated [H +] was strictly represented by E = EO' + Q log [H +] for both the acid and alkaline zones, with slightly different values of EO,. In each zone the relevant value of EO' was used; in intermediate pH ranges (4.5-8,5) an average EO' value was adopted. With this procedure experimental and calculated titration curves were completely superimposable.

°

Cyclic tetra-azatetra-acetic acids To obtain the titration curves of the complexones, 50.0 ml ofapproximately 1O- 3M solutions of the ligands, 5.0 ml of demineralized water and 5.0 ml of 1.200M potassium nitrate or tetramethylammonium nitrate were titrated with 0.050M carbonate-free potassium hydroxicte or tetramethylammonium hydroxide. To obtain the titration curves of the complexones in the presence of metal ions, 5.0 ml of 1.00 x 10- 2 or 2.00 x 1O- 2M solutions of the metal salts were added instead of the demineralized water. Other measurements

NMR spectra were recorded with a lOO-MHz Jeol JNM 100 PTF spectrometer coupled to a Jeol 980 A computer. When deuterated chloroform or acetone was used as solvent, tetramethylsilane (TMS) was the reference compound; when deuterium oxide was the solvent, the reference compound was the sodium salt of 2,2,3,3-tetradeutero3-(trimethylsilyl)propionic acid (DTSS). Melting 'points were determined with a Reichert-Thermovar instrument provided with a microscope, and are uncorrected. Elemental analyses were done with a Perkin-Elmer 240 Elemental Analyzer.

817

11

10

9

8

6

5

4

Ca/cu/ation of the stability constants

The stability constants of the various species formed were obtained from the experimental data with the aid of the program MINIQUAD 11 .12 and an IBM 360 computer. The program uses a non-linear least-squares method to optimize data and requires previous knowledge of approximate values for the constants. Suitable approximate values for the constants were obtained by use of simpler programs based on methods of calculation previously described in other publications from our laboratory. 10 The values were improved by comparison of the experimental titration curves of the complexones in the presence of the various metal ions with calculated titration curves for values of stability constants close to the estimated values, until a satisfactory superimposition of the curves was achieved. This was done on a 2200WANG computer coupled to a 2212 piotter. The f3 values for the best superimposition were then refined by MINIQUAD. In the present work, the values selected were those for which the calculated relative standard deviation was less than 10%, the "fitting index" R was less than 0.003 and the confidence level was higher th an 95% for 6 degrees of freedom, following the procedure of other authors. 13 .14 The standard deviations q uoted refer to calculation from data obtained in one experiment; however, the logarithmic values obtained from a series of titrations performed on different occasions were differed by not more than ± 0.05 from those presented, even in the most unfavourable cases. RESULTS AND DISCUSSION

Titration curves for the three complexones studied, alone and in the presence of several metal ions, are shown in Figs 2, 3 and 4. As can be seen, ail the ligands are tetrabasic, with two pKs of the order of 3-4 and two of the order of 10--11. The values refined by MINIQUAD are presented in Table 1 together with those determined by Stetter and Frank 2 at 20.0° and f1 = 0.1 (potassium chloride) and by Desreux et al. 15 at 25.0° and f1 = 1.0 (sodium chIo ride). Our values were determined at 25.0 ± 0.1 ° and J1 = 0.10 (tetramethylammonium nitrate or potassium nitrate); both sets of results are presented to allow comparisons. Agreement between the several sets of values is fair, taking into account the different experimental con-

3

2

o

1.0

2.0

3.0

5.0

4.0

a

Fig. 2. Titration curves for cDûT A alone and in the presence of met al ions in' 1: 1 ratio. T = 25.0 ± 0.1 oc. Il = O.IM [(CH3)4NNû3]. I~DûTA alone, 1O- 3M, and with 2,3-Na + or Li +; 4-Mg2+; 5-Ba2+; 6---Be2+; 7-Sr2+; 8-Ca2+;9, 100Co2+ or Ni2+; II-Zn2+; 12-Cu2+. ditions in which they were determined. The major differences are found in the values of pK 4 for cDOT A and cTET A and of pK 3 for cTET A. The discrepancy in the case of cDOT Amay be explained by the fact that we used non-complexing tetramethylammonium nitrate medium and Desreux et al. used lM sodium chIo ride. Although they made a correction for the formation of a sodium complex this may have been insufficient. In our determinations we obtained a value for the stability constant of this sodium complex that was higher than their guessed value, and if our value is adopted, the constants of Desreux et al. come close to ours. The values are already close for O.IM potassium nitrate medium (but potassium also forms a complex with cDOT A). The discrepancy in the case of cTET Amay also be explained in the same way, but now the correction made by Desreux et al. may be too high. We found for the sodium complex of this ligand log K = 0.4 ± 0.1 (with a 100: 1 molar ratio of sodium to ligand) and if this value is used instead of log K = 1.64, again closer agreement between their value and ours is obtained. Tt seems, from the available data, that the first two protons ionize from two trans acetic acid moieties and the last two from protonated trans nitrogen atoms, but hydrogen-bonding to pairs of nitrogen atoms is not excluded. 15 ,16

RITA DELGADO and J. J. R. FRAUSTO DA SILVA

818

o

1.0

2.0

3.0

o

5.0

4.0

1.0

2.0

4.0

3.Q

5.0

a

a

Fig. 3. Titration curves for cTRIT A alone and in the presence of metal ions in 1: 1 ratio. T = 25.0 ± 0.1 oC, J1 = O.IM (KNû 3). I~TRITA alone. 1O- 3 M. and with 2__ ~g2+; 3--8a 2+; 4--Sr 2+; 5--Ca 2+; 6--8e 2+; 7--Zn2+; 8--Co2+; 9--Ni2+; 1000Cu2+.

Fig. 4. Titration curves for eTETA alone and in the presence of metal ions in 1: 1 ratio. T = 25.0 + 0.1 OC. J1 = O.IM (KNû 3). I~TETA alone. 1O- 3M. a;d with 2--8a 2 +; 3--Sr2+; 4--Ca 2+; 5--8e2+; 6,7--Co2+ or Zn2+; 8--Ni 2 +; 9--Cu2+.

Regarding the formation of the metal complexes, two observations immediately arise from Figs 2-4. First, the stabilities of the complexes of the alkalineearth metals decrease as the size of the cavity of the

ligands increases. Secondly, there is no inversion of the order of stability of the calcium and strontium complexes of cTRIT A, contrary to the original report by Stetter and Frank,2 and log K CaL > log KS'L for

Table 1. Ionization constants of cyclic tetra-azatetra-acetic acids

Constant

[12]ane N 4 ' 4ac (cOûTA)

[13]ane N 4 '4ac (cTRITA)

4.130 ± 0.003 4.36 ± 0.01 4.41 4.18 ± 0.03

(a) (h) (c)

4.548 ± 0.003 4.37 ± 0.01 4.54 4.24 ± 0.02

(a) (h) (c)

9.680 ± 0.001 9.750 ± 0.002 9.73 9.23 ± 0.02

(a) (h)

12.09 ± 0.04 11.22 ± 0.03 11.36 11.08 ± 0.07

3.323 ± 0.004 3.28

[14]ane N 4 '4ac (eTETA)

(h) (c)

(d) 4.157 ± 0.006 4.59

(h) (e)

(d)

(c)

9.734 ± 0.002 9.18

(h) (c)

(d) (a) (h)

(e) (d)

11.35 ± 0.05 11.22

(h) (c)

(a) Present work: T = 25.0 ± O.loC; J1 = O.IOM [(CH3)4NNû3]. (h) Present work: T = 25.0 ± O.loC; J1 = O.IOM (KNû3)' (c) Stetter and Frank: 2 T = 20°e, J1 = 0.1 M (KCI). (d) Desreux et al.: 15 T = 25.0°C, J1 = 1.0M (NaCl), corrected.

3.347 ± 0.004 3.46 3.38 ± 0.04

(h)

± 0.004 4.31 4.05 ± 0.02

(h) (c)

10.136 ± 0.002 9.75 10.18 ± 0.02

(h) (c)

10.682 ± 0.005 11.07 11.56 ± 0.02

(h) (c)

4.091

(d)

(d)

(d)

(d)

...

r

'"

h.>

r-

>

ML

K+

MH 2 L MH ML

MH 2 L MH ML

MH 2L MH ML

MH 2L MHL ML

MHL ML

MH 2L MHL ML

MH 2L MHL ML

MHL ML

7.01 13.145 21.049

6.49 11.45 20.03

8.316 14.416 22.21

6.05 12.08 20.17

6.415 12.873

2.28 7.80 15.22

3.11 8.68 17.226

3.917 11.915

2.26 7.68 13.64

1.64

4.38

4.32

17.25

18.90

± 0.03 ± 0.09 ± 0.03

± 0.01 ± 0.006 ± 0.009

19.06

18.42

± 0.03 ± 0.02 ± 0.02

± 0.008 ± 0.007 ± 0.01

1l.3(e)

12.80

15.85

± 0.005 ± 0.002

± 0.03 ± 0.05 ± 0.02

± 0.06 ± 0.04 ± 0.005

5.56 12.138 19.42

7.175 13.639 20.821

(e) Stetter and Frank. 2a

(a) Present work: T = 25.0

7.21 14.03 21.53

6.17 12.73 20.10

3.641 8.342

3.688 9.995

5.451 12.085

2.781 7.620

± 0.005 ± 0.005 11.03

± 0.07(d)

2.41 7.58 13.36

2.52

(c)

± 0.03 ± 0.04 ± 0.01

± 0.05

± 0.02

± 0.03

(a)

cOOTA

± O.I°C; Il = 0.10M [(CH3)4NN03]. (h) Present work: T = 25.0 ± O.l°C; Il = O.IOM (KN0 3). (c) Stetter and Frank: 2 T = 20°C, Il = O.IM (KCI). (d) Oesreux et al.: 15 T = 25.0°C; Il = 1.0M (NaCl).

Zn2+

Ni 2 +

Cu 2 +

Co 2 +

Ba 2 +

Sr 2 +

Ca 2 +

Mg 2 +

MH 2L MHL ML

ML

Na+

Be 2+

ML

Species

Li+

Metal ions

± 0.01 ± 0.007 ± 0.02

± 0.006 ± 0.006 ± 0.006

± 0.04 ± 0.04 ± 0.04

± 0.02 ± 0.01 ± 0.01

± 0.005 ± 0.002

± 0.005 ± 0.002

± 0.005 ± 0.003

± 0.005 ± 0.004

± 0.05 ± 0.03 ± 0.01

(h)

cTRITA

14.42

15.75

17.29

14.98

7.24

8.5(e)

11.70

1O.4(e)

8.06

6.36

(c)

± 0.1

9.84 16.27

6.52 13.35 19.91

7.36 14.60 21.60

2.63 9.949 16.557

2.519 3.854

3.987 5.728

5.09 8.322

1.743 1.967

2.47 7.82 13.38

± 0.03 ± 0.01

± 0.03 ± 0.02 ± 0.03

± 0.03 ± 0.02 ± 0.03

± 0.07 ± 0.005 ± 0.004

± 0.005 ± 0.004

± 0.005 ± 0.005

± 0.004

+ - 0.05

± 0.005 ± 0.001

± 0.05 ± 0.05 ± 0.03

negligible

0.4

(h)

Table 2. Stability constants (log K) of met al complexes of cOOTA, cTRITA and cTETA cTETA

1.64

15.81

15.26

18.60

15.00

4.32

6.15

9.48

3.02

± 0.02(d)

(c)

0.-

'-D

00

~

()

O'

n'

()

Po $P.

O'

$P. ...,

~

N

Po

O'

$P. ...,

n'

(] '<

RITA DELGADO and J. J. R. FRAUSTO DA SILVA

820

Table 3. Comparison of stability constants (log K) for sorne alkali, alkaline-earth and transition-metal complexes of cDûTA, ethylenediamine-N,N' -diacetic acid (EDDA), ethylenediaminetetra-acetic acid (EDTA) and trans-I,2-diaminecyclohexanetetra-acetic acid (DCT A) T = 25.0°C, J.I. = 0.1 (potassium nitrate or tetramethylammonium nitrate) Metal ion

cDûTA

Li+ Na+ Be2+ Mg 2 + Ca2+ Sr2+ Ba 2 + Co2+ Ni2+ Cu2+ Zn2+

4.32 4.38 13.64 11.92 17.23 15.22 12.87 20.17 20.03 22.21 21.05

EDDA

3.95

11.25 13.65 16.2 11.22

4.13 4.55 9.68 12.09

6.53 9.59

EDTA

DCTA

2.85* 1.79*

4.13 2.70

8.83 10.61 8.68 7.80 16.26 18.52 18.70 16.44

11.07 13.15 10.58 8.6 19.58 20.2 21.92 19.35

2.0 2.61 6.11 10.17

2.4 3.5 6.12 12.3

*J.I. = 0.3M CsC\.

these cyclic complexones, as with the classical noncyclic ones, and as later found by Stetter and Frank. 2a For the transition metals the situation is not so cléarly defined; from the titration curves the only conclusion which can be reached is that the cobalt and zinc complexes of cTET A are markedly less stable than those of nickel and copper, in contrast with what happens with the other two ligands. Values for the stability constants of the various species identified in 1: 1 mixtures of the met al ions and of the ligands* are given in Table 2, along with the corresponding values for sorne ML species. 2 ,l5 These values confirm the qualitative observations made above but differ quite considerably from those reported by Stetter and Frank, 2 which are generally lower than ours by a factor of 10 2 _10 3 . Since these authors did not consider MH 2 L and MHL species, their values might be expected to be higher rather than lower, so we see no obvious reason for the difference. It should, however, be stressed that our values were found to be reproducible from titration to titration, and to vary little with differences in the nature of the ionic medium. To obtain such reproducibility required great care, particularly with the transition metals which gave rather slow reaction. Usually, the mixtures of metal and ligand were left to equilibrate for 2 hr before the titration, and in the case of Ni 2+ and cDOT A it took 24 hr to reach stable pH values. After each addition of titrant, the reading was not taken until the pH was stable, and this took 1-3 min, sometimes more, particularly in the case of

*2: 1 ratios were also used but the partial formation constants of the 2: 1 complexes were too low to be ca1culated with precision in these conditions; however, in the case of Ni 2+ -cTET A we obtained log fJM2L = 23.01 ± 0.05.

nickel and cobalt complexes. With cTRIT A, the reactions were generally faster. For the alkali-metal complexes of cDOTA and for the alkaline-earth met al complexes of ail ligands the stability follows the order of the ionic potential, except for magnesium and beryllium, which behave "abnormally", as usual, because of their small radii. Beryllium forms quite stable complexes with the three complexones and, at first sight, this seems of interest for the development of complexometric methods for this metal, but the side-reaction of beryllium with hydroxide prevents this (see below). For the transition metals, the Irving-Williams order of stability is observed; it is interesting to note, however, that the difference in stability between the complexes of the sa me metals with cDOT A, cTRIT A and even cTET A (in the cases of nickel and copper) is rather small, unlike the behaviour with the alkalineearth metals and also with the complexes of the transition metals with the parent cyclic amines (judging from the few values available for copper l7 ) and with the corresponding non-cyclic amines. This seems to imply that for the alkaline-earth metals the size of the internai cavity of the ligand and its conformation are critical and that ail nitrogen atoms may be involved in bonding to the metal, as postulated for the complexes of rare-earths with cDOTA. lB In contrast, the transition metals do not seem to coordinate to ail the nitrogen atoms of the ligands. This conclusion was also reached by Hiiflinger and Kaden on the basis of a spectral study. of the complexes formed by cobalt, nickel, copper and zinc with meso-5, 12-dimethyl-l ,4, 8,11-tetra-azacyclotetradecaneN,N',N",N"'-tetra-acetic acid, which is similar to cTET A. A comparison with the values of stability constants for the complexes of a few classical non-

Cyclic tetra-azatetra-acetic acids

16

~

12

~

"

.- ~-

"

j; .

"

J~--

~.-.-._'-'-'-

10

/

.,,/

/1//

/1"

//l

8

/" 1

/ /// . 1//

1/

6

/.1;"

/

I!

,1"

YI

4

/

? /

/

2

6

4

10

8

12 pH

Fig. 5. Variation of the conditional stability constants (log K~ad of the calcium complexes of cDOT A (--), transDCTA (---), EGTA (~~) and EDTA (_._) with pH.

cyclic complexones (Table 3) pro vides sorne support for this hypothesis. Although these complexones are not directly comparable with those studied in the present work a few conclusions can be extracted from the values presented in this table. 1. Co-ordination by one strongly basic nitrogen atom (pK - 10), one nitrogen atom of intermediate basicity (pK - 6) and two carboxylate groups, as in ethylenediaminediacetic acid (EDDA), gives com12

/.-.-."":.toc:: ....

.

8

,

',~,

",/ ~

/ / / / ~

6

/

"

/

.1

4

./

/

//

.1

/

/

---.....

'"

/

/ 2

/"

/

;"

1 /

1

/

821

pl ex es with stability constants lower than those determined for the corresponding cDOT A complexes by a factor of about 106~1O1O, but that are not too different from those determined for the cDOT A MHL complexes. 2. Co-ordination by a ligand such as EDT A, i.e., EDDA plus two carboxylate groups, gives transitionmetal complexes with stability constants lower by a factor of up to 104 than those determined for the cyclic complexones. 3. When the basicity of one of the nitrogen atoms is increased as in DCTA (pK 4 - 12.3), the complexes of the transition metals compare in stability with those of the cyclic complexones but the complexes of the alkaline-earth metals have stability constants that are 4~6 orders of magnitude below the values found for those of the strongest cyclic complexone, cDOT A. Hence, it seems that for the transition metals the MHL complexes of the cyclic complexones may be like those of EDDA i.e., co-ordinated to two nitrogen atoms and to two carboxylate groups. Their ML complexes require more ligand atoms to be co-ordinated and it is tempting to suggest that the two remaining carboxylate groups may be involved. In any case it is unlikely that the four nitrogen atoms can participate since then much higher stability constants would result (the log K values for the copper complexes of the cyclic amines are 17 about 25~29). The fact that the less stable complexes of the cyclic complexones form in preference to the more stable ones involving the four nitrogen atoms, as in the parent amines, may be the result of the combination of various factors (conformational, thermodynamic and kinetic) but at the present stage of our work it is not possible to ad vance a concrete soundly based explanation. Our final comments refer to the possible use of the cyclic complexones, particularly cDOT A, as reagents for complexometric titrations. In Figs. 5 and 6 the logarithms of the conditional constants of the calcium and magnesium complexes of cDOT A, EDT A, EGT A and DCT A are plotted vs. pH. These show that cDOT A is indeed the most powerful ligand for both metal ions and that the selectivity ratio log K~aJlog K~gL is as good for cDOT A as for EGT A, at pH > 10 (and is better still for cTET A). These ligands could then be interesting alternatives for titrations of these IOns. For beryllium, the conditional constants of the complexes are too low (log K~eL < 3) even at the most favourable pH values, so none of the new ligands is suitable for beryllium determination. For the transition metals, the cyclic complexones have no advantages over the classical non-cyclic ones such as EDT A or DCT A.

1

4

6

8

10

REFERENCES 12 pH

Fig. 6. Variation of the conditional stability constants (log K~gd of the magnesium complexes of cDOT A (--), trans-DCTA (---), EGTA (~~) and EDTA (-.-) with pH.

1. M. Takagi, M. Tazaki and K. Ueno, Chem. Letf., 1978, 1179. 2. H. Stetter and W. Frank, Angew. Chem. Imern. Ed. Engl., 1976,15,686

822

RITA DELGADO and 1. 1. R. FRAUSTO DA SILVA

2a. H. Stetter, W. Frank and R. Mertens, Tetrahedron, 1981,37,767. 3. H. Hiiflinger and A. Kaden, H elv. Chim. Acta, 1979, 62, 683. 4. 1. E. Richman and T. J. Atkins, J. Am. Chem. Soc., 1974, %, 2268. . 5. W. L. Smith, 1. D. Ekstrand and K. N. Raymond, ibid., 1978, 100, 3539. 6. L. Y. Martin, C. R. Sperati and D. H. Busch, ibid., 1977, 99, 2968. 7. E. R. Nelson, M. Maiethal, L. A. Lane and A. A. Benderly, ibid., 1957, 79, 3467. 8. G. Schwarzenbach and W. Biederman, Helv. Chim. Acta, 1948, 31, 331. 9. F. J. C. Rossotti and H. Rossotti, J. Chem. Educ., 1965, 42, 375.

10. 1. J. R. Frausto da Silva, Rev. Port. Quim., 1965, 7, 230. Il. P. Gans, A. Vacca and A. Sabatini, Talanta, 1974,21, 53. 12. Idem, Inorg. Chim. Acta, 1976, 18,237. 13. A. Sabatini and A. Vacca, Coord. Chem. Rev., 1975, 16', 161. 14. R. N. Sylva and M. R. Davidson, J. Chem. Soc. (Dalton), 1979, 465. 15. J. F. Desreux, E. Merciny and M. F. Loncin, Inorg. Chem., 1981,20,987. 16. R. Delgado, 1. Ascenso and 1. J. R. Frausto da Silva, to be published. 17. A. Anichini, L. Fabbrizzi, P. Paoletti and R. M. Clay, J. Chem. Soc. (Dalton), 1978,577, 18. J. F. Desreux, Inorg. Chem., 1980,19,1319.

Lihat lebih banyak...

Comentarios

Copyright © 2017 DATOSPDF Inc.