This work has been digitalized and published in 2013 by Verlag Zeitschrift für Naturforschung in cooperation with the Max Planck Society for the Advancement of Science under a Creative Commons Attribution4.0 International License.

Dieses Werk wurde im Jahr 2013 vom Verlag Zeitschrift für Naturforschungin Zusammenarbeit mit der Max-Planck-Gesellschaft zur Förderung derWissenschaften e.V. digitalisiert und unter folgender Lizenz veröffentlicht:Creative Commons Namensnennung 4.0 Lizenz.

Lanthanide(III) Complexes of Oxamic Acid

S. P. Perlepes, Th. F. Zafiropoulos, J. K . Kouinis, and A. G. Galinos* Department of Inorganic Chemistry, University of Patras, 231 Korinthou str., Patras, Greece

Z. Naturforsch. 36b, 697-703 (1981); received February 17, 1981

Oxamato Complexes, Lanthanide(III) Complexes, Primary Amide

The preparation, for the first time, of the deprotonated complexes of oxamic acid with Ce(III), Pr(III), Nd(III), Sm(III), Eu(III), Dy(III), Ho(III ) and Yb(III ) is reported.

Properties, analytical results, conductometric measurements, magnetic moments and spectral data (IR and diffuse reflectance spectra) are discussed in terms of possible struc-tural types and the nature of the bonding. Oxamic acid acts as a bidentate non-bridging ligand.

Introduction

Complexes of both carboxylic acids and primary amides have been studied rather widely; several vibrational analyses have been made on oxalato [1-3] and oxamido complexes [4-6]. Complexes of ligands containing both of these functional groups have not been widely studied. The simplest such ligand is oxamic acid (NH2COCOOH); however, little work has been published on oxamato com-plexes.

Oxamic acid has tumor growth inhibition prop-erties, mainly in its derivatives [7, 8]. This effect is probably due to the inhibition of action of lactic dehydrogenase. Wallace and Wagner [9] assigned and discussed the group vibrations of oxamic acid, sodium oxamate and bis(oxamato) cuprate(II). They also presented a normal coordinate analysis of oxamic acid; the spectral evidence indicated that this acid does not exist as the zwitterion in the solid state. Laser Raman and I R spectra of crystalline oxamic acid have been investigated in the region 4000-40 c m - 1 [10]. The spectral features are con-sistent with hydrogen-bonded chains, which consist of alternate acid and amide dimeric units each having a local center of symmetry.

Oxamic acid has very interesting ligating possi-bilities, because it may coordinate to the metal ions through both O-atoms or one nitrogen and one oxygen. The X - ray structural analysis of [Co(C203NH2)2(0H2)2] • 2H2O showed that Co(II) was chelated by the amidic oxygen and one of the carboxylic oxygens as donor atoms [11]. A complex

of oxamic acid, in which the amidic hydrogen was ionized, was reported very early [12] as K2[Cu(C203NH)2] . W e prepared and studied chelates of oxamic acid with the first-row transition, I I b and I H b metal ions [13-15]. The synthesis of a binuclear copper(II) complex with a ^-oxamato-bridge was reported recently [16]. The product of the reaction between K2[PtCLi] and oxamic acid, the so-called "platinum oxamate blue" , is thought to consist of platinum-containing polyanions with a resultant P t - P t interaction along the polymeric or oligomeric chain [17].

In recent years interesting developments in the field of the complex chemistry of the lanthanides have been made. Not only the complexes with various donor atoms have been synthesized [18-20], but also a huge amount of structural work is now available. The coordination chemistry of the lan-thanide ions is dominated by ligands that are pure oxygen-atom donors or mixed oxygen- and nitrogen-atom donors. A survey of the literature on lan-thanide(III) coordination compounds reveals that ligands involving amide groups as donor sites have received little attention.

As a continuation of our interest in the study of the metal-amide interactions [13-15, 21-23], the present paper deals with the preparation of the Ce(III) , Pr(III ) , Nd(III ) , Sm(III ) , Eu(III ) , Dy( I I I ) , H o ( I I I ) and Y b ( I I I ) anionic chelates with oxamic acid.

Experimental Materials

Oxamic acid was obtained from Pfaltz and Bayer, Inc. The rare earth(III) chlorides as hydrated products were prepared by dissolving the appro-

priate oxide (99.9%, Merck) in 5 N HCl. The product was crystallized under vacuum. The com-position of these salts has been controlled by determining the lanthanide and chloride contents. DMSO was obtained from Merck (99 .5%) and used without further purification.

Preparation of the chelates A solution (50 ml) of 0.03 mol of oxamic acid in

DMSO was mixed with 0.01 mol of metal salt dissolved in DMSO (25 ml). The solutions were heated at 50-60 °C for 45 min. To the solutions obtained an aqueous solution (6-8 ml) of 0.05 mol of K O H was added dropwise, under vigorous stir-ring at room temperature. Precipitates were im-mediately obtained. The reaction mixtures were refluxed at 60-70 °C for several hours (exact time not critical). The precipitates were filtered off after cooling to room temperature and washed with anhydrous acetone (15 X 5 ml). They were refluxed again in DMSO for 1 h to be freed of unreacted oxamic acid, then filtered and washed with dry acetone. This step was repeated two or three times. After the final filtration the compounds were washed several times with acetone (ca. 300 ml), dried in vacuum over P4O10 and again dried in vacuum for 12 h at 105-110 °C.

Analyses and physicochemical measurements Microanalyses for C and H were performed by the

microanalytical unit of the Steel Corporation, Almiros, Volos, Greece. The determination o f potas-sium and the other metals was carried out b y fusing an accurately weighed amount of the compound

with five times the amount of Na2C03 in a platinum crucible at ca. 900 °C. The fusion product, after cooling, was dissolved in the minimum volume of 6 N HCl and diluted with water to a definite volume. From this solution potassium was determined by the Na+[B(C6H5)4]- method [24]. For the determina-tion of the other metals, the above solution was neutralized slowly to p H 5 and the metal ion was titrated with a standard 0.05 M E D T A solution, using an acetate buffer and Xylenol Orange as indicator [25].

The physicochemical measurements were carried out as previously reported [22, 23, 26]. The conduct-ances of the solutions were measured within 2 min o f dissolution.

Results and Discussion General properties of the complexes

Under completely aqueous conditions it is almost impossible to prepare the complexes in pure form, both because oxamic acid decomposes partially and because the species obtained are contaminated with ligand formed by the hydrolysis of the complexes. Therefore, we have prepared the oxamato com-plexes using DMSO as a solvent (oxamic acid is insoluble in the common organic solvents).

In our previous work on oxamato complexes [13, 14], the quantity of K O H added in each case was equivalent for the precipitation of the metal hydroxide from the metal chloride in solution plus 5 0 % excess. In this work we have used a larger

Table I. Analytical results, yields, colors, Am values and decomposition temperatures of the oxamato complexes.

Decomp. Compound [%] C H M K Yielda

[%] Color [S cm-

temp. -2mol-i] [°C]

K3[Ce(C203NH)3(H20)3] Calcd Found

12.59 12.43

1.57 1.50

24.47 25.23

20.49 21.07

98 off-white

349 260

K3[Pr(C203NH)3(H20)3] Calcd Found

12.57 12.51

1.57 1.49

24.58 24.99

20.46 20.97

96 pale green

i 240--245

K3[Nd(C203NH)3(H20)3] Calcd Found

12.50 12.39

1.56 1.60

25.01 24.78

20.34 20.49

57 pale lilac

390 c

K3[Sm(C203NH)3(H20)3] Calcd Found

12.36 12.23

1.54 1.60

25.80 25.01

20.13 20.80

94 cream i 238--241

K3 [Eu(C203NH)3 (H20)3 ] Calcd Found

12.33 12.30

1.54 1.52

26.00 26.73

20.07 20.37

75 white 405 e

K3[Dy(C203NH)3(H20)2] Calcd Found

12.49 12.37

1.21 1.20

28.17 27.66

20.33 21.02

21 white 371 270

K3 [Ho(C203NH)3 (H20)2 ] Calcd Found

12.44 12.53

1.21 1.21

28.47 28.90

20.25 20.94

88 pale yellow

i 240--250

K3 [Yb(C203NH)3 (H20)2 ] Calcd Found

12.27 12.20

1.19 1.15

29.46 28.99

19.97 20.42

43 white 386 280--283

M = metal; a based on the metal; b values of molar conductance at 28 °C for ca. 10 -3 M solutions in H2O; the complexes were assumed monomeric. i = insoluble; c no visible thermal change till 285 °C.

excess in order to improve the yields o f the com-plexes. When the ratio mol K O H / m o l MCI3 was larger than 5.3, hydroxo-derivatives with poor analytical results were obtained; in addition am-monia wras detected in solution.

The prepared complexes are crystalline and stable in the normal laboratory atmosphere over a long period of time. They are insoluble in polar organic solvents and moderately soluble in H2O, but are decomposed by dilute mineral acids. The lanthanide complexes are of relatively high thermal stability. The high temperature loss of H2O may indicate that the wrater molecules are coordinated. However, a recent study by Nuttall and Stalker [27] concluded that water elimination above 200 °C is not a priori indicative of coordinated water.

The analytical results and some physicochemical data are given in Table I. The molar conductances of the Ce(III), Nd(III ) , Eu(III ) , Dy(I I I ) and Y b ( I I I ) complexes in water are in good agreement with those reported for 3 : 1 electrolytes [28]. The con-ductances of the solutions increase slightly with time.

Magnetic moments and diffuse reflectance spectra

Table II gives the magnetic moments (295 K ) of the new complexes.

Table II. Magnetic moments of the oxamato com-plexes.

M iMeff [B. M.]

Ce 2.56 Pr 3.46 Nd 3.61 Sm 1.65 Eu 3.59 Dy 10.79 Ho 10.14 Y b 4.51

The paramagnetic behavior of lanthanide(III) ions is due to the presence of unpaired 4 f electrons. Since these 4 f electrons are well shielded by the 5 s 2 5 p 6 octet, the magnetic susceptibility of a lanthanide(III) complex should indicate whether these 4 f electrons are affected by bond formation or not. The magnetic moments of the present chelates show little deviation from Van Vleck values [29], indicating thereby that 4 f electrons do not partici-pate in bond formation. Thus the magnetic suscep-tibilities of the new lanthanide complexes of oxamic

acid are within the range predicted and observed in compounds of paramagnetic lanthanide(III) ions [30, 31].

Tables I I I - V I I give details of the diffuse re-flectance spectra (350-800 nm) of some complexes. The spectra of the Ce(III) and Y b ( I I I ) complexes show no maxima in the 350-800 nm spectral region, while the spectrum of K 3 [Dy(C20 3NH)3(H 20)2] exhibits weak bands at 351, 363, 427, 454 and 477 nm.

Table III . Internal 4 f 2 transitions in the Pr(III) complex at room temperature.

^max Assignment0

[nm] 3H4 ->

448 3P2 473 3Pi + He 487 3Po 594 !D2 604 sh V

sh = shoulder; a from ref. [32, 33, 35, 36].

Table IV. Internal 4 f 3 transitions in the Nd(III) com-plex at room temperature.

^max Assignment4 ^max Assignment1

[nm] 4Ig/2 -> [nm] 4l9/2 -»•

357 4D3/2 574 sh, 5771 . 4G5/2, 2G7/2 428 2Pl/2 582, 593 shj 463 4GH/2 640 2HH/2 470 (2D, 2P)3/2 684,693 4F 9/2 477 2G9/2 735, 745 4F7/25 4S3/2 510 V 786 ? 515 4"G9/2 790 sh ? 523 4G7/2 800 4F5/2, 2H9/2 534 sh 2KI3/2

sh = shoulder; * from ref. [32-36]; t> the 4I 9 / 2 -> 4G5 /2 , 2G7/2 transition is hypersensitive.

Table V. Internal 4 f 5 transitions in the Sm(III) com-plex at room temperature.

^•max Assignment8

[nm] 6H5 / 2 -

363,376 ? 403 6'P3/2 416 V 421 6P 5 /2 443 1 464 4 I l 3 / 2 478 ? 528 4F 3/2 561 4G 5 /2

» From ref. [32, 33, 35, 36].

Table VI. Internal 4f6 transitions in the Eu(III) com-plex at room temperature.

Amax Assignment [nm] 7F o

361 sd4 364 ? 375 5G4/5) SG6 393 s l 6 414 5D3 424 ? 464 sd2

a From ref. [33] and [36].

Table VII. Internal 4f10 transitions in the Ho(III) complex at room temperature.

/max Assignment81 >b

[nm] 5 I 8 - >

361 3 G 6

386 ? 417 5 ' G 5

424 sh ? 450, 453 sh\ 5 G 6

459 sh, 468/ 474 ? 486 5 F 3

537 5 S 2 , 5 F 4

551 sh. 641 sh ? 644 5 " F 5

660 sh 1

sh = shoulder; * from ref. [32-36]; b the 5I8 -> 3G6 and 5l8 5Ge transitions are hypersensitive.

The reflectance spectra of rare earth complexes of oxamic acid may involve intraligand, charge transfer and f - f transitions which are found at decreasing energy (and intensity). The f - f transitions are observed in the visible region; in these transi-tions the bands are extremely sharp, because of the shielding effect of 5s2 and 5p 6 shells.

Sinha has reported [37] that the red shift of bands in the complexes of various ligands with respect to the aquo complexes is related to covalency in the metal-ligand bond. Our data indicate that the energy of f - f transitions in the complexes is slightly reduced as compared to that of the corresponding aquo ions [34, 38-40], either because of the slight covalent interaction of the 4f orbitals with vacant orbitals, leading to some delocalization with conse-quent reduction in interelectronic repulsion, or more normally by increased nuclear shielding of the f-orbitals due to slight covalent ligand-metal elec-tron drift [40]. It is in any case clear that the 4 f orbitals do not strongly take part in the bonding; if

they did, vibronic coupling would lead to marked changes in position and sharpness of the f - f transi-tions [41].

Several investigators have noted that, although most of the Laporte forbidden electronic transitions within the f n configuration are affected only slightly by environmental changes, the fine structure and intensities of a few, e.g. the 4Ig/2 ->4Gs/2, 2G?/2 tran-sition of Nd(III) (Table IV) and the 5 I 8 -> 3 G 6 . 5 I 8 -> 5 G 6 transitions of Ho(III) (Table VII), are extremely sensitive ("hypersensitive"). J0rgensen and Judd [42] have suggested that these transitions are probably pseudoquadrupolar in nature. If an asymmetric distribution of dipoles is induced by the electromagnetic field in the environment about the ion, the variation of the electric vector across the ion is much greater than for a symmetric distribu-tion of dipoles. As a result, the hypersensitive pseudoquadrupolar transitions, which are normally very weak, are enhanced. Alternatively, Judd [43] has assigned the hypersensitivity principally to a change of point symmetry, the explanation being presented within the framework of Crystal Field Theory.

The hypersensitive transition 4l9/2->4Gs/2, 2G?/2

(near 580 nm), shows differences in shape and intensity with variation in coordination number [33, 34, 44]. The band is more intense and much sharper with fewer shoulders for Nd(III) compounds with coordination number 10 or more. The oxamato chelate of Nd(III) has a complex spectrum in the 574-594 nm region; thus it has coordination number smaller than 10 [44], Since reflectance spectra were used band shapes but no intensities were considered.

IR spectral studies Tables VIII and I X give details of the IR spectra

together with assignments for most bands. The I R spectra of all the complexes in the

4000-450 cm - 1 region are similar, being independent of either the number of water molecules or the specific metal ion.

In the r ( O H ) w a t e r region the spectra exhibit a strong, single and sharp band at 3380 cm - 1 , attribut-able to the presence of coordinated aqua ligands [45]. The absence of a weaker broad continuous absorp-tion, covering the whole 3600-3150 cm - 1 region, indicates that lattice water (in addition to the aqua ligands) is not present [45]. The presence of co-

Table VIII. Characteristic IR frequencies of diagnostic value (4000-450 cm - 1 ) of oxamic acid and oxamato complexes with their tentative assignments.

Assignment» NH2COCOOH K 3 [M(C 20 3NH) 3 (H 20)„]

f(OH)water 3380 ss l 'as(NH2) 3350 s v8(NH2) 3242 s v(NH) 3240 m, br v(OH)acid 2880 w

2760 w 2685 w 2540 w 2450 w

*(C = 0)acId 1740 m v(C = 0)amideb 1675 m, br v(C = 0) 1690 m 1625 m, br <5(NH2)C 1595 m <50H( + vC-0) 1470 m i>(C-0 + CN) 1440 s 1320 vs vCN( + <5NH2) 1363 s 1365 w vC-0( + <50H) 1240 vs, br 7l(OH) 985 vs v(CC) 839 m 865 m 855 m W(NH2) 819 s jr(NH) 790 m T(NH2) 753 vs CO(H20) ~ 540 m

a From ref. [9, 10, 15]; b amide I ; c amide II. n = 2 or 3. ss = strong and sharp, vs = very strong, s = strong, m = medium, w - - weak, br = broad, vas = antisymmetric stretching, v8 — symmetric stretching, ö = in-plane deformation, n = out-of-plane bending, co = wagging, r — torsion.

ordinated water is also confirmed by the appearance of the w(H 2 0) mode [46].

The characteristic vibrational modes of the O H and NH 2 groups are absent from the spectra of the complexes. The I R spectrum of oxamic acid shows two strong absorption bands at 3350 and 3242 c m - 1 , which are due to the asymmetric and symmetric N - H stretching vibrations of the - N H 2 group [9,10]. In the spectra of the lanthanide(III) complexes these two bands are replaced by a medium single band at ca. 3240 cm - 1 . This change is an indication

of the ionization of the amidic hydrogen, since a similar change occurs when a primary amide is converted to a secondary amide [13, 14]. Usually, the non-hydrogen bonded secondary amides absorb in the 3460-3420 c m - 1 region. The coordination with the metal ions causes a considerable shift to lower frequencies.

The v(OH) frequency of acids in the monomeric state is expected in the 3500-3000 c m - 1 region, but strong hydrogen bonding in the solid state can cause lowering and broadening of the band. The O H stretch appears in the spectrum of oxamic acid as the usual very broad absorption covering the 2900-2400 c m - 1 region with several sub-maxima, which are also typical of OH stretching modes. These bands disappear after chelation, for the complexes examined.

The carbonyl stretching region in the spectra of the complexes is somewhat unusual. The motions of the six carbonyl groups are very likely coupled. The bands at ca. 1690 and 1625 c m - 1 are probably due to these modes and are assigned simply as v ( C = 0 ) . However, Wallace and Wagner [9] observed only one very strong broad absorption band at 1606 cm-i in the spectrum of K 2 [Cu(C 2 0 3 NH) 2 ] ; they assigned this band to an overlapping of both of the I R active v ( C = 0 ) modes. Nakamoto and Armendarez [5] observed a strong broad doublet in the 1660-1600 c m - 1 region in the spectra of some oxamido chelates. The appearance of two bands at ca. 1690 and 1625 c m - 1 in our spectra rules out the existence of oxamato-bridges [16].

The i '(C-O) and <5(OH) vibrations are strongly coupled in acids. A very strong broad band, which appears at 1240 c m - 1 in the spectrum of oxamic acid, is absent or diminshes in intensity very considerably after complexation.

The region of the free ligand's spectrum between 450 and 250 c m - 1 has only two absorption bands at

Compound Observed absorptions

NH2COCOOH 332 ma 317 w b

K3[Ce(C203NH)3(H20)3] 447 sh 369 m 338 m 275 sh 259 w K3[Pr(C203NH)3(H20)3] 449 m 330 m 260 m K3[Nd(C203NH)3(H20)3] 449 w 330 m 265 m K3[Sm(C203NH)3(H20)3] 450 m 380 w 330 m 285 w 260 m K3[Eu(C203NH)3(H20)3] 380 w 328 m 305 w 287 w 258 m K3 [Dy(C203NH)3 (H20)2 ] 397 m 334 w 269 m 260 m K3[Ho(C203NH)3(H20)2] 445 w 328 m 305 w 285 w 260 w K3 [Yb(C203NH)3 (H20)2 l 389 m 337 m 301 m 281 w

Table IX . IR data for oxamic acid and oxamato complexes in the 450-250 cm - 1

region.

a g(C02); b <w(skel). sh = shoulder, m = medium, w - weak, o = rocking, u> = wagging.

332 and 317 cm - 1 . This would indicate that other bands observed in this spectral region (Table I X ) would be assignable to metal-ligand stretching frequencies. Metal-ligand vibrations are difficult to assign on an empirical basis since their frequencies are sensitive to the metal and often they couple with other low frequency modes in metal chelates. A comparison of I R spectra between free ligand and its metal complexes fails to give clear cut assignment since some ligand vibrations, activated by complex formation, may appear in the same region as the metal-ligand vibrations. Of the several bands ob-served in the far I R spectra, some of them may be assigned to v(M-N), v(M-O) and r(M-OH2) vibra-tions [40, 47].

Final conclusions The lanthanide ions are typical " A " type cations

in the Ahrland-Chatt-Davies sense, or "hard" acids in the Pearson sense. The majority of the complexes that can be isolated contain pure oxygen donors or mixed oxygen-nitrogen donors [19].

On the basis of available evidence a tentative coordination number of nine in the lighter lanthani-des (Ce-Eu) and eight in the heavier lanthanides (Dy, Ho, Yb) can be assigned. In the complexes

oxamic acid acts as a bidentate N , 0 non-bridging ligand. On these grounds 9- and 8-coordinated arrangements of the [LnOßNa] and [LnOsNs] types respectively can be suggested for the complexes in the solid state. The decrease in the H 2 0 :M ratio for heavier lanthanides is obviously a consequence of decrease in ionic radii due to the well known "lanthanide contraction".

Our results are consistent with the reports of Panyushkin and his co-workers [20]. They have reported that for lanthanide(III) chelates the num-ber of members in the metal-containing rings is connected with the coordination number; thus, five-membered rings lead to complexes with constant coordination numbers (9 for light lanthanides and 8 for heavy lanthanides).

A final point is the absence of either coordinated or lattice DMSO molecules, in spite of the fact that the synthesis was carried out in a DMSO environ-ment. Since DMSO is a solvent with a higher electron-donating activity than that of H 2 0 , the absence of DMSO from the complexes was un-expected.

The authors thank Mr. M. Papazisis for C and H microanalyses.

[1] H. Murata and K. Kawai, J. Chem. Phys. 25, 598 (1956).

[2] M. J. Schmelz, T. Miyazawa, S. Mizushima, T. J. Lane, and J. V. Quagliano, Spectrochim. Acta 9, 51 (1957).

[3] J. Fujita, A. E. Martell, and K. Nakamoto, J. Chem. Phys. 86, 324 (1962).

[4] Y. Kuroda, M. Kato, and K. Sone, Bull. Chem. Soc. Jpn. 34, 877 (1961).

[5] P. X . Armendarez and K. Nakamoto, Inorg. Chem. 5, 796 (1966).

[6] T. J. Thamann and T. M. Loehr, Spectrochim. Acta, Part A 36, 751 (1980).

[7] B. P. Baker and W. W. Lecote, J. Med. Pharm. Chem. 1960, 633.

[8] D. T. Plummer and J. H. Wilkinson, Biochem. J. 81, 38 (1961).

[9] F. Wallace and E. Wagner, Spectrochim. Acta, Part A 34, 589 (1978).

[10] G. N. R. Tripathi and J. E. Katon, Spectrochim. Acta, Part A 35, 401 (1979).

[11] M. A. Pellingelli, A. Tiripicchio, and M. Tiri-picchio-Camellini, Acta Crystallogr., Part B 28, 998 (1972).

[12] K. A. Hofmann and U. Ehrhardt, Ber. Dtsch. Chem. Ges. 46, 1457 (1913).

[14] J. K. Kouinis, J. M. Tsangaris, and A. G. Galinos, Z. Naturforsch. 33b, 987 (1978).

[15] S. P. Perlepes, Th. F. Zafiropoulos, J. K. Kouinis, and A. G. Galinos, Inorg. Nucl. Chem. Lett. 16, 475 (1980).

[16] K. Nonoyama, H. Ojima, K. Ohki, and M. Nonoyama, Inorg. Chim. Acta 41, 155 (1980).

[17] J. H. Burness, Inorg. Chim. Acta 44, L 49 (1980). [18] T. Moeller, D. F. Martin, L. C. Thompson, R.

Ferrüs, G. R. Feistel, and W. J. Randall, Chem. Rev. 65, 1 (1965).

[19] T. Moeller, in MTP International Review of Science, Inorganic Chemistry, Series One, Vol. 7 (Edited by K. W. Bagnall), pp. 275-298, Butter-worths, London 1972.

[20] Y. T. Panyushkin, Yu. A. Afanas'ev, A. D. Garnovskii, and O. A. Osipov, Russ. Chem. Rev. 46, 1109 (1977).

[21] J. M. Tsangaris, S. Perlepes, and A. G. Galinos, Z. Naturforsch. 34b, 456 (1979).

[22] A. G. Galinos, S. P. Perlepes, Th. F. Zafiropoulos, P. V. Ioannou, and J. K. Kouinis, Monatsh. Chem. 1980, in press.

[23] Th. F. Zafiropoulos, S. P. Perlepes, P. V. Ioannou, J. M. Tsangaris, and A. G. Galinos, Z. Naturforscb. 36b, 87 (1981).

[24] A. I. Vogel, A Text-Book of Quantitative In-organic Analysis, 3rd ed., p. 564, Longman, London 1961.

[25] J. E. McDonald and T. Moeller, J. Inorg. Nucl. Chem. 40, 253 (1978).

[26] A. G. Galinos, J. K. Kouinis, P. V. Ioannou, Th. F. Zafiropoulos, and S. P. Perlepes, Z. Natur-forsch. 34b, 1101 (1979).

[27] R. H. Nuttall and D. M. Stalker, J. Inorg. Nucl. Chem. 40, 39 (1978).

[28] N. K. Dutt and K. Nag, J. Inorg. Nucl. Chem. 30, 3273 (1968).

[29] J. H. Van Vleck and A. Frank, Phys. Rev. 34, 1494, 1625 (1929).

[30] R. Didchenko and F. P. Gortsema, J. Phys. Chem. Solids 24, 863 (1963).

[31] N. M. Karayannis, C. M. Mikulski, J. V. Minkie-wicz, L. L. Pytlewski, and M. M. Labes, J. Less-Comm. Met. 20, 29 (1970).

[32] J. L. Ryan and C. K. J0rgensen, J. Phys. Chem. 70, 2845 (1966).

[33] D. G. Karraker, Inorg. Chem. 6, 1863 (1967). [34] D. G. Karraker, Inorg. Chem. 7, 473 (1968). [35] J. L. Burmeister, S. D. Patterson, and E. A.

Deardorff, Inorg. Chim. Acta 3, 105 (1969).

[36] K. B. Yatsimirskii and N. K. Davidenko, Coord. Chem. Rev. 27, 223 (1979).

[37] S. P. Sinha, Spectrochim. Acta 22, 57 (1966). [38] S. P. Sinha, J. Inorg. Nucl. Chem. 27, 115 (1965). [39] J. L. Ryan, Inorg. Chem. 8, 2053 (1969). [40] M. C. .Tain, R. K. Sharma, A. K. Srivastava, and

P. C. Jain, J. Inorg. Nucl. Chem. 41, 1305 (1979). [41] D. R. Cousins and F. A. Hart, J. Inorg. Nucl.

Chem. 29, 1745 (1967). [42] C. K. J0rgensen and B. R. Judd, Mol. Phys. 8, 281

(1964). [43] B. R. Judd, J. Chem. Phys. 44, 839 (1966). [44] J. I. Bullock and H-A. Tajmir-Riahi, Inorg.

Chim. Acta 38, 141 (1980). [45] A. N. Speca, L. S. Gelfand, L. L. Pytlewski,

C. Owens, and N. M. Karayannis, Inorg. Chem. 15, 1493 (1976).

[46] I. Nakagawa and T. Shimanouchi, Spectrochim. Acta 20, 429 (1964).

[47] E. W. Bensen, R. E. Kohrman, R. A. Pickering, and D. X . West, J. Inorg. Nucl. Chem. 42, 143 (1980).