Uranyl complexes of α-carboxypolymethylenediaminetetra-acetic acids

Tal~fa, Vol. 34, No. 5, pp. 519-524, 1987 Printed in Great Britain. Al1 rights reserved

~39-91#~87 53.00 + 0.00 Copyright Q 1987 Pergmon Journals Ltd

ANALYTICAL DATA

URANYL

COMPLEXES OF a-CARBOXYPOLYMETHYLENEDIAMINETETRA-ACETIC ACIDS

A. MAT~LLA HRRNANDEZ, S. GONZALEZ GARCIA and J. M. TRRCERO MORENO ~~~amento de Q&mica Inorganica, Facultad de Farmacia, Unive~idad de Granada, 18071-Granada, Spain M. CANDIDA T. A. VAZ and L. V~LAS BOAS Centro de Quimica Estrutural, Instituto Superior Tecnico, 1096 Lisboa Codex, Portugal (Received 18 Judy 1986. Aecepred 20 December

1986)

Summary-The uranyl complexes of N,N,N’,N’-tetrakis(carboxymethyl)-2,3-diaminopropionic acid, N,N,N’,N’-tetrakis(carboxymethyl)diaminobutyric acid, N,N,N’,N’-tetrakis(carboxymethyl)omithine and N,N,N’,N’-tetrakiscarboxymethyl)lysine have been studied by potentiometry, with computer evaluation of the titration data by the MINIQUAD program. Stability constants of the 1: 1 and 2: 1 metahligand chelates have been determined as well as the hydrolysis and polymerization constants at 25” in O.lM potassium nitrate. Results are compared with those obtained for the many1 complexes of the corresponding members of the series of the pol~ethylen~i~netetra-attic acids.

We have recently prepared a new series of’complexones, the a-carboxypolymethylenediaminetetra-acetic acids (cpdta) and studied their complexation reactions with some metal ions.‘” These acids have the general formula: (HOOCCH,),NCH(COOH)

EXPERIMENTAL Reagents

All ligands were prepared by condensation of the cormsponding diamino acid with mon~hlo~~tic acid in alkaline medium.‘” Uranyl nitrate was analytical grade (Fluka), used without further purification. The stock I@+ solutions was standardized gravimetrically. The potassium hydroxide solutions used as titrants were prepared under nitrogen from Merck “Titrisol” vials, with C&-free demineralized water. The concentrations of these solutions and the absence of carbonate were checked regularly by ~~tiomet~c titration with standard hydrochloric acid.

(CH,),N(CH,COOH)r

where n = 1,2,3 and 4, and are respectively named cxcarboxyethylenediaminetetra-acetic acid (CEDTA), ~,~,~‘,~ - tetra~s(~r~x~ethyl} - 1,3 - diaminobutyric acid (DBT), D,L-N,N,N‘,N’-tetrakis(carboxymethyl)omithine (OTC) and N,N,N’,N’-tetrakis(carboxymethyl)lysine (LTC). The asymmetry of these compounds may lead to some properties of possible interest as complexforming agents, since the two complexing groups will behave differently, one being more powerful than the other. In this work, we present results obtained for the formation in solution of complexes of the uranyl ion and the newly synthethized ligands. The complexes formed by the uranyl ion and the parent ~l~ethylen~i~ine~tra-attic acids (@a) have been thoroughly investigated and the influence of the length of hydrocarbon chain on the stability of the complex species formed has been discussed.m The present study aimed to check the effects of the asymmetry of the ligands on the number and type of species formed and their stability. The potentiometric data were evaluated by computer with the MINIQUAD program;9V’o we selected probable models for the series of complexes formed and calculated the corresponding stability constants.

Potentiometric measurements

A Crison digital potentiometer and Ingold electrodes were used. The measured potentials were converted into [H+]values according to the expression E = K + a log&I+]. The experimental value of a, determined at 25*, was 59.2 f 0.1 mV, in good agreement with the theoretical value of 59.15 (the drift of the liquid-junction potential during the measurements in the pH range of the titrations was found to be negligible and was not considered in the expression used). The cell constant K was calculated from a previous titration of hydrochloric acid with potassium hydroxide with end-point location by C&an’s method.” The cahbration was repeated before and after each titmtion. The value of the ionic product of water (in 0.1 M KN4) used in the calculations was 1.68 x lo-“. Method

Potentiometric titrations were Performed on mixtures with Uq+/cpdta ratios of 1: 1 and 2: 1 at a range of ligand concentrations from 5 x lo-* to 1.5 x 10e3M. At coneentrations below 5 x 10A4Mno polymeric species are formed, and at concentrations above 1.5 x IO-‘&f the polymeric species precipitate as soon as the metal and ligand solutions are mixed. The titrations were performed in a double-walled titration cell with the temperature controlled at 25.0 f 0.1” by water 519

520 Table 1. Stability constants of the proton complexes of ffie ligands (25”, 0. I M KNQ,) L&and CEDTA DBT OTC LTC

B011 1.182 x IO’O 1.680 x lOi0 3.022 x IOr 2.455 x lo’s

B012 3.580 x lOI 5.079 x 1O’8 3.254 x IOr 5.248 x tot9

B013 7.707 x IOt9 3.807 x 10” 2.415 x IO= 7.413 x 1021

B 014

4.227 x 6.425 x 4.677 x 1.778 x

B 0,s

1p2 l@’ I@’ lo2f

6.796 x 5.900 x 3.196 x 1.288 x

IOU 10zs tOa6 to=

Table 2. Hydrolysis constants of uranyl ion” B10-1 1.16 x 1O-6

BE-2 1.26 x lo-&

BXI.3 4.90 x to-‘3

circulated from a thermostat. The titrant was added from a Metrohrn automatic burette. After each addition of base, the equilibrium potential was reached within 3&!80 sec.

The MINIQUAD program, through a leastsquares refinement, gives the formation constants of the complex species defined by:

where r = b - u, M is UO:+ and L the cpdta iigand, corresponding to the reaction: pM $ qL + aH20~M~LqH6(0H)o

B 1.74 zo-23

B 3.47 :;,-I,

The behaviour of the titration curves of the different UO$+/cpdta systems is similar, and clearly suggests the formation of MLH and M,L species (Fig. 1). However, it is not possible to rule out the presence of other protonated species or polynuclear hydrotysed species, and the value of the R parameter does indeed decrease when these arc introduced into the model. The number of possible complex species in the U@+/cpdta system is high; the following equilibria are compatible with the titration curves. 1: 1 species LH,+M r MLH3 1 MLH, 1 MLH 1 MLH(OH)

t- (a - b)H+.

The program also calculates an agreement factor, R, and other statistical parameters which aflow comparison of the models proposed. First, the protonation constants of the ligands were calculated (Table 1); the values obtained are in good agreement with those reported previously.‘-’ These constants, together with the hydrolysis constants of the uranyl ion (Table 2), were introduced as fixed data in the calculations for the different uranyl systems. The potentiometric titrations of the cpdta show the increasing difference in nitrogen basicity as the chain lengths decrease,‘-5 as expected from electrostatic and hydrogen-bonding effects.

“I

$ MLH(O~)~

~~L~H~(OH)~

2 : 1 species 2M + LH, r M,LH r M,L

/ M .+,L,+~(OH)t+<

W.JdOW,n

Mz L($H) \ r M,L(OH), f /

Mz,Lz (OH),

MJ-,K'WS+-+M~WW~

The stability constants (log/?) of the complexes corresponding to the models for which the smallest Fig. 1. Titration curves of LTC (a) and LTC-uranyl ion in molar ratios 1: 1 (b) and 2: 1 (c), at 25” and ionic strength 0. IM KNO,; C,, = 7.12 x 10e4M; a = degree of neutralization; 1 indicates precipitation.

values of R (R < 0.04) were obtained are given in Table 3. Species containing more than four UO:+

ions are rejected by the program. M,LOH and MrL(OH), species, which may be intermediates for the formation of some other complexes, are also

ANALYTICAL

521

DATA

Table 3. Stability constant of UOz+-cpdta complexes for the best model obtained with the MINIQUAD standard deviations In % (25”, O.lM KNO,) Ligand

log 8113 23.04 + 0.08 25.00 f 0.05 25.17 + 0.14 25.32 + 0.05

log BllZ 20.11 f 0.03 22.25 + 0.04 22.96 f 0.06 23.17 + 0.01

log Bill 17.06 k 0.02 19.39 f 0.02 19.91 f 0.03 20.05 f 0.01

log BllO

log Bll.,

CEDTA DBT OTC LTC

13.64 * 0.03 14.14 f 0.06 14.27 + 0.02

7.33 f 0.06 7.73 *0.13 7.73 f 0.08

CEDTA DBT OTC LTC

19.99 f 22.16 + 22.99 f 23.08 f

16.53 + 18.61 f 19.22 + 19.42 +

10.63 + 0.29 13.15 * 0.13 -

7.62 f 0.20 -

r

100

0

2

0.02 0.01 0.01 0.01

0.01 0.02 0.02 0.01

26.32 + 30.12 f 3 1.86 f 32.30 +

0.07 0.13 0.03 0.02

program, with

log Bl20 26.27 + 29.62 f 31.34 f 31.03 +

0.05 0.40 0.22 0.14

15.31 f 19.21 f 20.86 f 21.15 +

0.39 0.18 0.04 0.03

0.0026 0.0035 0.0032 0.0035

100

UO:/CEDTA:

UO;/DBTA=l/l

III

6

5

4

3

M% 100

UO;/OTC=I/I

0

2

3

4

5

0

6

2

3

4

5

PH

Fig. 2 Fig. 2(a). Fraction of the metal present in each complex as a function of pH, for the systems UO:+/cpdta in l/l ratio.

6

IO0 UOE+/CEDTA B

=2/l

Fig. 2(b). Fraction of the metal presr?nt in each complex as B function crfpH, for the systems UO$+/cpdta in 2/l ratio.

for somesysta~s~ and we t&e thii to mm that their ~~n~#~~nsare too0 small compared to those of the other species present. The: stability constants follow a trend which correlates with that of the protonation constants of the ligands, ie.., they are generally higher far the compounds with longer methy~ene chains. The difference is most marked from the first to the second member of the series, as expected from analogous differences in the values of j&z. The distribution of the species is also different in the case of the first member of the series (CEDTA), rejected

which does not farma normal ML compgex, and the dimer M2Lr is more relevant above pH 5 than in the case of the higher homologues-see Fig. 2. The protonated complex MLH is the dominant species between pH 3 and 6 in all systems and it should be noted that the NIL2 complexes correspond to MLH dimers in which two LID:+ groups bonded to an ~rn~od~~~~ group of the &and are bridge-d by two hydroxyi groups. The two protuns wiIl of course attach to the two remaining imino nitragen atoms of the ligands. Between pH 5 and 6 several other minor poiy-

ANALYTICAL

DATA

-; 2-

Table 4. Average difference between the conditional constants of the correSDondinR so&es _-odta and codta _ Ki@KlogK,,” Ligands 0.764 0.070 0.027

EDTA-CEDTA PDTA-DBT BDTA-GTC

nuclear species seem to coexist, but although the calculated constants are reproducible and the fitting parameters satisfactory, precipitates start to be formed in this pH range, depending on the concentration ratios of the metal and ligand, so a detailed discussion of differences and trends in the various systems is not warranted. The stability constants given by the MINIQUAD program for the UO$+/cpdta systems are always higher than those obtained for the corresponding Uq+/pdta systems. This does not indicate that the ability of the new ligands to complex the UO:+ ion is higher for the cpdta than for the pdta, since the basicity and the number of protons of the ligand must be taken into account. For this sort of comparison “conditional constants”, as defined by Bingbom and Schwarzenbach,‘3*‘4 must be used. The values of the conditional constants for the MLH species of CEDTA, DBT and OTC and for the corresponding UO:+ complexes formed with acid (EDTA), nethylenediaminetetra-acetic propanediaminetetra-acetic acid (PDTA) and nbutanediaminetetra-acetic acid (BDTA)*, were calculated by using the expression: Kb,,

lMLH1

= L

WI

ILit

where [Ml’ and [L]’ are the total concentrations of metal and protonated ligand not involved in MLH (i.e. the conditional concentrations). The values of the conditional constants are pHdependent and were calculated for a range of pH in which the MLH species are dominant (pH 3-5). The results are represented graphically in Fig. 3, and it can be seen that the trend is now reversed, i.e., the pdta form stronger complexes than the cpdta with UO$+ in this range of pH. The difference is highest for EDTA and CEDTA, smaller for PDTA and DBT and within the error limits for BDTA and OTC. The decreasing difference on moving forward in the series, and the highest values for EDTA-CEDTA and PDTA-DBT show that the a-carboxyl substituent does not co-ordinate to UOz+ and, furthermore, decreases the stability of the complexes due to its effect on the basicity of the nitrogen atoms of the ligand. This effect is especially important if it is not compensated by a sufficiently long hydrocarbon chain. These findings confirm that the uranyl ion has co-ordination number five in its equatorial plane for complexes formed with this type of ligand.‘5*‘6 Thus

523

-a

,

EDTA

l-

C)-

.z

3

k

3 -1

-2

-3

L

Fig. 3. Variation of log K$,,

with pH for qdta and pdta.

one of the carboxyl groups of the ligand cannot co-ordinate; on the contrary it decreases the stability of the complexes formed, by its negative inductive effect, particularly in the case of the first members of the series since both nitrogen atoms of the ligands are affected. For the higher members the effect is compensated and the presence of the acarboxyl group becomes irrelevant with reference to complexation of the uranyl ion. Acknowledgemenr-The authors thank Prof. J. J. R. Fratisto da Silva for help given in the interpretation and discussion of the results, and to the Spanish M.E.C. for a grant that enabled the research to be Performed. REFERENCES

1. S. Gonzalez Garcia, F. Sanchez Santos and M. F. Morales Ayala, Ars Phurm., 1976, 17, 295. 2. S. Gonzalez Garcia, J. Niclos Gutierrez and A. Matilla Hernandez, An. Q&n., 1983, 79B, 24. 3. J. Niclos Gutierrez, S. Gonzalez Garcia, A. Matilla Hemandez and J. M. Tercero Moreno, ibid., 1983,79B, 517. 4. A. Matilla Hemandez, S. Gonzalez Garcia, J. Niclos Gutierrez and J. M. Tereero Moreno, ibid., 1985, MB, 297. 5.

J. Niclos Gutierrez, S. Gonzalez Garcia, A. Matilla Hemandez and J. M. Tereero Moreno, ibid., 1983,79B, 525. 6. J. J. R. Fradsto da Silva and M. L. Sirnoes Gorqalves, Tuhta, 1968, 15, 609.

524

ANALYTICAL

7. M. L. SimBes Gongalves, A. M. Almeida Mota and

8. 9. 10. 11. 12.

J. J. R. Fralisto da Silva, ibid., 1983, 30, 69. i&m, ibid., 1984, 31, 531. A. Sabatini, A. Vacca and P. Gans, ibid., 1974, 21, 53. P. Gans, A. Sabatini and A. Vacca, Inorg. Chim. Acta, 1976, 18, 237. G. Gran, Rnuiyst, 1952, 77, 661. S. Ahrland, Acta Chem. Stand., 1951, 5, 199.

DATA

13. A. Ringbom, Les complexes en chimie onalytique, Dunot, Paris, 1967. 14. G. Schwarzenbach and H. Flaschka, Complexometric Titration, Methuen, London, 1969. 15. J. J. R. Fratisto da Silva and M. L. Sii5es Gonqalves, .J. Znorg. Chem., 1970, 32, 1313. 16. R. Graziani, B. Zarli, A. Cassol, G. Bombieri, E. Forsellini and F. Tondello, fnorg. Chem., t970,9,2116.