Portal de Periódicos da CAPES

Surface‐activated chemical ionization versus electrospray ionization in the study of selected aluminium(III)/ligand solution equilibria

2006; Wiley; Volume: 20; Issue: 4 Linguagem: Inglês

10.1002/rcm.2354

1097-0231

Valerio Di Marco, Martina Ranaldo, G. Giorgio Bombi, Pietro Traldi,

Molecular Sensors and Ion Detection

Electrospray ionization mass spectrometry (ESI-MS) is being increasingly used to study metal/ligand equilibria in solution.1 The number and stoichiometry of the complexes can be generally determined from m/z values, MSn spectra, and isotopic ratio analysis. The relative abundances of the complexes, hence their stability constants, can be calculated under favourable conditions from the intensity of the related peaks. In recent years, other MS techniques have been developed which allow the detection of species directly from solution samples: APCI (atmospheric pressure chemical ionization),2 APPI (atmospheric pressure photoionization),3 SACI (surface-activated chemical ionization),4 and APLI (atmospheric pressure laser ionization).5 These techniques, and especially the most recently developed SACI and APLI, demonstrate improved sensitivity and reduced chemical noise with respect to ESI,4, 5 and they can be expected to be employed like ESI in the study of metal/ligand equilibria in solution. Moreover, as good-quality SACI spectra can be obtained in pure water (whereas ESI generally requires a water/organic solvent mixture), equilibrium data could be obtained in conditions more similar to those encountered in physiological and environmental solutions. The solution equilibria between aluminium(III) and either 2,3-dihydroxypyridine (1) (Fig. 1), 3,4-dihydroxybenzoic acid (2), citric acid (3), or ethylenediaminetetramethylenephosphonic acid (4), have been recently studied by means of ESI (V. B. Di Marco, M. Ranaldo, G. G. Bombi, P. Traldi, to be published). In this work, SACI was used to study the same metal/ligand solution equilibria, with the purpose of comparing SACI and ESI performance. Structure of 2,3-dihydroxypyridine (1). Solutions for the SACI-MS and ESI-MS measurements were prepared by dissolving in water (purified with a Milli-Q/plus apparatus, Millipore) weighed amounts of ligand (1: Aldrich; 2: Ega-Chemie; 3: Baker; 4: product of synthesis6) and AlCl3 · 6H2O (Fluka). Concentration values are given in molality (m, mol/kg). Solutions for ESI analysis were added with the proper amount of methanol (Prolabo, >99.8% purity) to obtain 70:30 w/w water/methanol mixtures. The pH of the solutions was adjusted by adding concentrated NaOH (Fluka, minimal purity 98%), and it was measured with a combined glass electrode (BDH 309/1025/02) previously calibrated against standard buffer solutions (pH 4 and 7). The instrument employed for both SACI and ESI was a Thermoquest (Finnigan MAT) LCQ mass spectrometer with an upper mass spectral limit of m/z 2000. Solutions were infused directly via a syringe pump, using N2 as nebulising gas. For ESI measurements, the flow rate was 5 µL min−1, the capillary temperature was set at 250°C, the spray voltage was 4 kV, and the vaporization temperature was 300°C. SACI is based on the simple vaporization of the analyte solution, performed by the same heating device employed for APCI experiments. Typical vaporization temperature are in the range 300–400°C and, in the present investigation, this parameter was fixed at 300°C. Considering that the source employed for the present investigation has a ‘linear’ configuration (in other words the vaporizer and the entrance capillary are on the same axis), a metallic surface (stainless steel) was placed close to the vaporizing device with a small angle with respect to the vaporizer axis. In this condition the ion production is the results of two different phenomena:7 (i) vaporization of the ions already present in solution and their better focalization with respect to the entrance capillary of the MS analyzer; (ii) ionization phenomena occurring on the surface. Relevant to this approach are some instrumental parameters, among which the surface area, the nebulized solution flow and the voltage applied on the surface are the most important ones. For these reasons a careful evaluation of these parameters was performed. The stainless steel surface area was kept constant (1 cm2) and the best values of solution flow and applied voltage were found to be 30 µL/min and 50 V, respectively. The sheath gas flow was also optimized. No auxiliary gas was used. SACI and ESI spectra for solutions containing Al(III) and 1 at several pH values are shown in Figs. 2 and 3, respectively. The recognised peaks are summarised in Table 1. Two peaks were obviously observed only by ESI as they are methanol adducts, but all the other peaks can be identified both in SACI and in ESI spectra. In other words, both SACI and ESI suggest the same speciation data, i.e. that the metal/ligand complexes formed in solution are AlL (detected at m/z 154, 172, and 186), AlL2 (m/z 247, 269, and 287), AlL3 (m/z 380), Al2L3 (m/z 400, 418, and 432), Al2L4 (m/z 511), Al2L5 (m/z 626), and Al2L6 (m/z 737) (the proton content cannot be determined by MS techniques1). Table 2 summarises the relative intensity of main peaks in SACI and in ESI spectra: values are very similar. As peak intensity is related to the concentration of the corresponding species, both SACI and ESI suggest a very similar quantitative distribution of the complexes at each pH value. SACI+-MS spectra of aqueous solutions containing Al(III) and ligand at several pH values. The spectrum labelled ‘Sol 1:1’ was obtained from a solution containing 9.7 × 10−4 m AlCl3 and 1.0 × 10−3 m ligand 1, the other spectra from a solution containing 2.1 × 10−4 m AlCl3 and 1.0 × 10−3 m ligand. ESI+-MS spectra of H2O/CH3OH (70:30, w/w) solutions containing Al(III) and ligand at several pH values. The spectrum labelled ‘Sol 1:1’ was obtained from a solution containing 3.6 × 10−4 m AlCl3 and 3.6 × 10−4 m ligand 1, the other spectra from a solution containing 8.1 × 10−5 m AlCl3 and 4.2 × 10−4 m ligand. The speciation obtained by SACI and ESI (Table 1) differs from that indicated by available thermodynamic data,8 both in the number and stoichiometry of the species which should be present at equilibrium, and in their quantitative distribution (Table 3). Therefore, equilibrium perturbations occurred in the MS analysis, as will be discussed in work to be published. Our results indicate that equilibrium was modified in the same (or in a very similar) way in SACI and in ESI. Among the several phenomena which in MS analysis may perturb the equilibrium state,1 only gas-phase reactions and differences in ion response factors should be significant for Al(III) complexes.9 The similarity of gas-phase reactions and of ion response factors in SACI and in ESI are therefore suggested. SACI spectra of adequate quality could not be obtained for solutions containing Al(III) and either 2, 3, and 4 (corresponding spectra are therefore not reported). With ESI, on the other hand, no or little problems were encountered in the study of the same solutions (V. B. Di Marco, M. Ranaldo, G. G. Bombi, P. Traldi, to be published). Despite the possible advantages of SACI above ESI, at the present time this latter technique appears to be more successful in the study of metal/ligand solution equilibria. Valerio B. Di Marco [email protected]*, Martina Ranaldo* , G. Giorgio Bombi*, Pietro Traldi , * Università degli Studi di Padova, Dipartimento di Scienze Chimiche, via Marzolo 1, 35131 Padova, Italy, CNR, Istituto di Scienze e Tecnologie Molecolari, Corso Stati Uniti 4, 35100 Padova, Italy.