Structural and IR-spectroscopic characterization of cadmium and lead(II) acesulfamates

1 Introduction

Acesulfame-K, the potassium salt of 6-methyl-1,2,3-oxathiazin-4(3H)-one-2-2-dioxide, is one of the most widely used low-calorie artificial sweeteners [1], [2] and its general chemical and biological properties have been thoroughly investigated [1], [2], [3].

The coordination capabilities of the acesulfamate anion, which resembles those of saccharin (1,2-benzothiazole-3(2H)-one-1,1-dioxide) (see Fig. 1) [4], have often been exploited and a large number of metal complexes containing this ligand have been reported during the last decade (cf. for example [5] and references therein).

Fig. 1:

Schematic drawings of the structures of saccharin (A) and acesulfamic acid (B).

Concerning simple salts of the anion, apart of the potassium salt [1], [6], well known for more than 40 years, some others (NH₄⁺ [7], Na⁺, Rb⁺, Cs⁺ [8], Mg²⁺ [9], Ca²⁺ [10], Sr²⁺, Ba²⁺ [11] and Tl⁺ [12]) have been prepared and characterized only very recently.

In order to advance in the knowledge of this type of compounds, we have now prepared and investigated the crystal structures and vibrational-spectroscopic behavior of the corresponding Cd(II) and Pb(II) salts.

2 Results and discussion

Cadmium acesulfamate has been obtained by the reaction of CdCO₃ with an aqueous solution of acesulfamic acid, using a general procedure recently reported by us for the synthesis of simple acesulfamate salts [7], [8], [9], [12] and which is described in the Experimental section. Lead(II) acesulfamate was prepared in the same way, starting from the basic carbonate, 2PbCO₃·PbO·H₂O. Cd(ace)₂·(H₂O)₂ crystallizes in the monoclinic space group P2₁/c and Pb(ace)₂ in the triclinic space group P1̅.

Ortep [13] drawings of the two acesulfamate salts are shown in Figs. 2 and 3. Bond lengths within the acesulfamate anions for the two salts are detailed in Table 1 and short contact distances around the metals are listed in Table 2.

Fig. 2:

View of the one-dimensional polymeric cadmium acesulfamate salt showing the labeling of the non-H atoms and their displacement ellipsoids at the 30% probability level. Cadmium-ligand bonds are indicated by full lines and hydrogen bonds by dashed lines. Symmetry operations: (i) x, 3/2−y, ½+z; (ii) x, 3/2−y, −½+z; (iii) x, ½−y, ½+z.

Fig. 3:

View of the two-dimensional polymeric lead acesulfamate salt. Symmetry operations: (i) 1−x, 1−y, 1−z; (ii) 1−x, −y, 1−z; (iii) x−1, y, z; (iv) 2−x, 1−y, 1−z.

The observed bond lengths and angles within the acesulfamate anions in both salts agree with one another and with values reported for other divalent metal salts, namely Mg(ace)₂·6H₂O [9] and Ca(ace)₂ [10], and also for the ammonium salt [7] and the M-ace (M: Na, K, Rb, Cs) isomorphic series of alkali metal salts [6], [8]. Particularly, for the better refined cadmium salt, the short Ci3–Ci4 (i: 1,2) bond lengths of 1.322(6) and 1.321(5) Å confirm the formal double bond character expected for this link. The carbonyl >C=O double bond lengths are 1.255(5) and 1.248(4) Å and sulfoxide S=O bond lengths are in the range 1.406(3)–1.427(3) Å. The other ring single bond lengths are: C–O bond lengths of 1.389(4) Å, O–S bond lengths of 1.599(2) and 1.620(3) Å, S–N bond lengths of 1.585(3) and 1.577(3) Å, C–N bond lengths of 1.354(4) and 1.347(4) Å, and C–C bond lengths of 1.439(5) and 1.443(5) Å. These bond lengths can be compared with those of the neutral crystalline acesulfamic acid molecule [14]. Major changes in the bonding structure of the acesulfamate ion in the cadmium salt occur at the S–N bond, which upon deprotonation suffers an average shortening of about 0.048 Å (about 16 times the standard error σ). A smaller shortening (−0.03 Å=−10σ) is observed in the N–C bond length. Other major bond length changes when going from neutral acesulfamic acid to the acesulfamate anion are observed for the ring S–O and O–C bonds (+0.017 Å=+5.5σ and −0.016 Å=−5σ, respectively).

Cd(ace)₂(H₂O)₂: The Cd(II) ion is in a seven-fold bipyramidal environment, equatorially coordinated to two acesulfamate anions acting as bidentate ligands through their carbonyl oxygen atoms [Cd–O bond lengths of 2.646(2) and 2.423(2) Å] and nitrogen atoms [Cd–N bond lengths of 2.295(3) and 2.409(3) Å] and to the carbonyl oxygen atom of a neighboring acesulfamate [d(Cd–O)=2.303(2) Å] which gives rise to a one-dimensional polymeric structure that extends along the c axis. The apical positions are occupied by water molecules [Cd–Ow bond lengths of 2.280(3) and 2.314(3) Å]. As already appreciated in Fig. 2 and detailed in Table 3, the polymer is further stabilized by hydrogen bonding where the w1 water molecule bridges neighboring monomers through OwH···O(sulf) [d(O1w···O21ⁱ)=2.938(4) Å, ∠(O1w–H···O21ⁱ)=158(4)°] and OwH···O(carb) [d(O1w···O24ⁱⁱ)=2.780(3) Å; ∠(O1w– H···O24ⁱⁱ)=166(4)°] medium-strength interactions. The other coordination water molecule (w2) provides an additional intra-chain O2wH···O(sulf) bond [d(O2w···O22ⁱⁱⁱ)=3.016(4) Å; ∠(O2w–H···O22ⁱⁱⁱ)=155(5)°].

Pb(ace)₂: The larger ionic radius of the Pb(II) ion (1.19 Å) as compared with the Cd(II) ion (0.95 Å) affords a metal coordination number larger than for cadmium, because the crowding of ligands (L) around the metal limits M–L short contacts. In fact, Pb(II) is in an eight-fold polyhedral coordination with two independent acesufamate anions acting as bidentate ligands through their carbonyl oxygen [Pb–O(carb) bond lengths of 2.516(5) and 2.699(6) Å] and nitrogen atoms [Pb–N bond lengths of 2.750(6) and 2.580(6) Å], with one sulfoxide oxygen atom each of two neighboring symmetry-related acesulfamate ligands [Pb–O(sulf) bond lengths of 2.835(5) and 2.914(5) Å]. Bonding of carbonyl oxygen atoms of two other acesulfamate groups [Pb–O(carb) bond lengths of 2.484(5) and 2.730(5) Å] generates a two-dimensional polymeric structure. This gives rise to an electrically neutral layered arrangement parallel to the (a, b) plane. Neighboring layers are separated from each other by the unit cell vector c and bonded to each other mainly through van der Waals interaction. This explains both the crystal growing habit as thin plates parallel to (001) and the observation of this being an easy cleavage plane.

Being lead a member of group 14 in the periodic table, it is to be expected some kind of sp hybridization giving rise to directional bonding. However, it has been recognized that relativistic effects are much more important for the high Z-valued lead than for the other members of the group down to tin [15], [16]. As a consequence, these relativistic effects produce a radial shrinkage of the 6s and 6p atomic orbitals of lead, which are now more effective in shielding the nucleus attraction felt by d and f orbitals, which therefore expand outwardly. The net effect is that the valence s and p electrons are thus buried in the atomic electronic structure and therefore less prone for bonding by hybridization. This effect is called ‘inert lone-pair effect’ [17].

On account of published crystallographic information and own MO calculations, Shimoni-Livny et al. [18] proposed a classification to characterize this and other effects on metal-ligand bonding in lead compounds, based on the stereochemistry of the coordination sphere around the metal: (1) holodirected, when Pb-ligand bonds present a distributed orientation throughout all the sphere and (2) hemidirected, when Pb-ligand bonds are only found directed throughout part of the sphere, hence leaving sizable orientation voids which are considered a signature of ‘stereochemically active lone pair of electrons’. Besides this effect, the authors also considered the competing influence on the environment geometry due to the coordination number, interaction between neighboring ligands, etc. to conclude that, among other factors, holodirected geometry around Pb(II) is favored by high coordination numbers (crowding) and repulsive interaction among ligands, which lead to relatively long and uniform values for Pb-ligand bond lengths. This is the case of our Pb(ace)₂ salt, with a coordination number of 8 and negatively charged acesulfamate ligands. In fact, our average (dispersion) values for Pb–N and Pb–O bond lengths of 2.67(9) and 2.70(15) Å, respectively, are in agreement with corresponding values reported for a whole series of Pb(II) compounds [2.62(11) and 2.53(15) Å] with holodirected stereochemistry [14].

It seems interesting to make some comparisons of the two newly reported structures with those of some related complexes. The acesulfamate complex of Zn(II), the lighter analog of Cd(II) in the same group of the periodic table, crystallizes in the triclinic space group P1̅ (Z=1) as a centro-symmetric [Zn(ace)₂(H₂O)₄] octahedral complex, where acesulfamate acts as a monodentate ligand through its carbonyl oxygen atom [19].

In the case of Cd(II) saccharinate, a similar monoclinic (P2₁/c) complex of composition [Cd(sac)₂(H₂O)₄]·2H₂O was described. In this case coordination occurs through the N atom of saccharin [20]. The complex is isostructural with the saccharinato complexes of the first row transition metal cations M(II) [4], [21]. On the other hand, the lead(II) ion in the monoclinic (P2₁/c) Pb(sac)₂(H₂O) salt is also in an eight-fold polyhedral environment that includes two independent saccharinate anions acting as bidentate ligands through their –N–(C=O)– ‘bite’, structurally closely related to the acesulfamate –N–(C=O)– ‘bite’ (cf. again Fig. 1). The eight-fold coordination is completed with a water molecule and the oxygen atoms of a carbonyl and two sulfoxide groups of a neighboring saccharinate ligand [22].

The FTIR spectra of both compounds show practically identical patterns and band intensity distributions, confirming the close similarities in bonding characteristics. The spectra are also very similar to those of previously investigated acesulfamate salts [8], [9], [12]. The spectrum of the Cd(II) salt is shown in Fig. 4 and the spectral assignments proposed for both salts are presented in Table 4. They were made on the basis of a recent experimental and theoretical study of potassium acesulfamate [23], additionally supported by our previous studies [7], [8], [9], [12], and are briefly discussed as follows:

Fig. 4:

FTIR spectrum of Cd(acesulfamate)₂·2H₂O in the spectral range between 4000 and 400 cm⁻¹.

• The O–H stretching band in the case of Cd(ace)₂·2H₂O can be seen as a relatively broad and strong band. The corresponding deformational modes, δ(H₂O), are probably overlapping with the bands found in the 1590–1660 cm⁻¹ region. No evidences for librational H₂O modes could be found in the low frequency spectral range. From the spectroscopic point of view the hydrogen bonds generated by these water molecules can be classified as medium strength [24], [25].

• As in the previously investigated cases, vibrational modes related to ν(C=O) and ν(C–C) stretches are strongly coupled and result in two very strong IR bands.

• The 1289 cm⁻¹ band (1293 cm⁻¹ in the Pb(II) salt) is dominated by the ν(CN) stretching vibration, coupled with some other motions.

• Bands related to the SO₂ vibrations are found practically at the same frequencies in both salts, despite the fact that this group participates in bonding in the case of Pb(ace)₂. This behavior is in agreement with the structural results, which show only a very slight lengthening of the S–O bonds of this moiety in the case of the Pb(II) salt (cf. again Table 1).

3 Experimental section