Print version ISSN 0103-5053
J. Braz. Chem. Soc. vol.15 no.2 São Paulo Mar./Apr. 2004
Luciana A. da Silva; Jailson B. de Andrade*
Instituto de Química, Universidade Federal da Bahia, Campus de Ondina, 40170-290 Salvador - BA, Brazil
Although the first mixed valence double sulfite, Cu2SO3.CuSO3.2H2O, was prepared in the early 19th century by M. Chevreul, interest in this type of compound was only rekindled in the mid-1960s, when the crystalline structure of Chevreul's salt was determined and its infrared spectrum characterized. Studies of this type of compound have been intensified in recent years, especially after the discovery that the isomorphic Cu(II) in Chrevreul's salt could be replaced by a divalent metal ion, forming an isomorphic series whose properties are strongly dependent on the nature of the M(II) cation. Because of their interesting properties, these mixed valence systems can be used as models to identify intermediates in atmospheric corrosion processes, and to evaluate the role of transition metals as catalysts of S(IV) autoxidation in the conversion of SO2 in the atmosphere.
Keywords: Chevreul's salt, double sulfites, mixed valence systems, isomorphic series
Embora o primeiro sulfito duplo de valência mista, Cu2SO3.CuSO3.2H2O, tenha sido preparado no início do século XIX por M. Chevreul, o interesse por compostos dessa natureza ressurgiu só depois da metade do século XX com a determinação da estrutura cristalina deste composto e a caracterização de seu espectro de absorção na região do infravermelho. Nos últimos anos, os estudos referentes a este tema têm se intensificado, especialmente após a descoberta da possibilidade de substituição isomórfica do Cu(II) no sal de Chevreul por diversos cátions com o mesmo estado de oxidação, gerando uma série de compostos cujas propriedades são fortemente dependentes da natureza do cátion M(II). Devido às suas propriedades interessantes, esses sistemas de valência mista podem ser usados como modelos para identificar intermediários em processos de corrosão atmosférica, bem como para avaliar o papel de metais de transição como catalisadores de autooxidação de S(IV) na conversão do SO2 na atmosfera.
Mixed valence systems are of considerable interest in chemistry because their properties rarely represent only the sum of the ions separately. The first known mixed valence sulfite complex, Cu2SO3.CuSO3.2H2O, was prepared by M. Chevreul in 1812.1 However, it was only in 1965 that its crystalline structure was determined by Kierkegaard and Nyberg in terms of coordination polyhedra (Figure 1).2,3 The structure consists of [SO3] trigonal pyramids, [CuIO3S] tetrahedra and [CuIIO4(H2O)2 ] octahedra linked together, giving a three-dimensional network pertaining to the P21/n spatial group. The structure is monoclinic, with the unit cell containing two formula units and dimensions of: a = 5.5671 Å; b = 7.7875 Å; c = 8.3635 Å and b = 91.279o.
The first optical approach to identify Chevreul's salt was published in 1973, when Nyberg and Larson4 studied sulfite ion coordination types using fundamental active infrared modes. X-ray diffraction results indicate that the free sulfite ion is pyramidal (C3v symmetry), with the sulfur at the apex.5 This model gives rise to four fundamental active infrared modes: n1 (symmetric stretch), n2 (symmetric bend), n3 (asymmetric stretch), and n4 (asymmetric bend). The two asymmetric modes are doubly degenerated. Because the Na-O bonds in Na2SO3 are very weak, this compound can be used as a free sulfite ion standard for comparison with other metal sulfites. The infrared spectrum of the Na2SO3 in the solid state presents the following bands (in cm-1): n1 = 1010 (medium), n2 = 633 (strong), n3 = 961 (strong), and n4 = 469 (strong).6
Assuming that the 3d orbitals of sulfur participate in bonding, the S-O bond shows a partially double bond character and two of the resonance structures of the sulfite ion can be written as shown in the Figure 2.
In an oxygen-coordinated compound (Figure 2a), the symmetry of the group changes to Cs and the number of fundamental active infrared modes increases to six, due to the removal of degeneracy from n3 and n4. This is a consequence of the decrease in the bond order (S-O) following the decrease in the stretching frequency. If the sulfite group is coordinated with the metal through sulfur (Figure 2b), the C3v symmetry is preserved, but the stretching modes should shift to higher frequencies than those of the free ion because of the higher bond order. The sulfite group can coordinate to the metal as a monodentate sulfur (I) or oxygen (II) donor or, in bidentate form, via oxygen (III) or oxygen and sulfur (IV) (Figure 3).7
According to Nyberg and Larson,4 the structure of sulfite compounds can be divided into three groups after coordination of the sulfite ion: (i) compounds without sulfur coordination (under 975 cm-1); (ii) compounds with both sulfur and oxygen coordination (over and under 975 cm-1); and (iii) compounds with dominant sulfur coordination (over 975 cm-1). Chevreul's salt spectrum shows stretching frequencies of high intensity over and under 975 cm-1, suggesting that the compound belongs to group (ii), with both metal-sulfur bonds and metal-oxygen bonds.
The presence of both Cu(I) and Cu(II) in the structure of Chevreul's salt was confirmed by the presence of two distinct 2p1/2 and 2p3/2 copper bands with their associated satellites in the photoelectron spectrum.8 The binding energy of the 2p1/2,3/2 sulfur band confirmed the expectation that sulfur is present in only one oxidation state.8 The magnetic susceptibility of the Cu2+(Cu+)2(SO32-) 2(H2O)2 formula gives the magnetic moment 1.90 M.B. for the paramagnetic Cu2+ ion. The latter value agrees with the magnetic moment calculated from EPR g factors determined by Inoue et al.9 These magnetic properties support the formula proposed by Chevreul.1
2. Isomorphic Series
After Chevreul first prepared his salt, several simple sulfites of transition metals Zn, Ni, Ag, Pd, Co, and Cd and double sulfites consisting of transition metals and an alkali metal or ammonium ion were prepared and characterized.4,10-13 Although Cu(II) sulfite (CuSO3) seems never to have been isolated, the literature contains references14 to two simple sulfites of Cu(I), commonly distinguished as Étard's salt, Cu2SO3.1/2H2O, and Rogojski's salt, Cu2SO3.H2O. The color, diamagnetism, infrared spectrum and elemental analysis of Étard's salt confirm its formulation.15 However, Dasent and Morrison15 proved that Rogojski's salt is an approximately equimolar mixture of Cu2SO3.CuSO3.2H2O and metallic copper. They suggested that its preparation method involved a partial disproportionation of the Cu(I) compound:
followed by the immediate combination of the resulting Cu(II) sulfite with unchanged Cu2SO3 to form Chevreul's salt:
The resulting red mixture of Cu and Cu2SO3.CuSO3.2H2O has the same composition as 2Cu2SO3.H2O.15
The accidental discovery of a Cu, Fe, Mn triple sulfite in the mid 60s, when a hydrometallurgical study was designed to obtain precious metals from deep-sea manganese nodules, gave rise to preliminary studies involving complex sulfites, which resulted in preparation of non-stoichiometric compound mixtures, including the Chevreul's salt.16-18 Almost two centuries after Chevreul's salt preparation, Cipriano,18 Miguel and de Andrade19 studied the chemistry and preparation of double sulfites involving Cu(I) and a divalent metal ion, M(II), establishing the best conditions to obtain the complete replacement of Cu(II) by transition metal ions such as Fe(II), Mn(II) and Cd(II) ions. In this case, their X-ray diffraction data were similar to those of Chevreul's salt, with minor changes in "d" spacings, forming an isomorphic series: Cu2SO3.CuSO3.2H2O, Cu2SO3.FeSO3.2H2O, Cu2SO3.MnSO3.2H2O, and Cu2SO3.CdSO3.2H2O. 20-22
Sghyar et al.22 determined the structure and crystallographic parameters of isomorphic series of doubles sulfite such as Cu2SO3.MSO3.2H2O (M = Cu, Fe, Mn, and Cd):
The infrared spectra of Cu2SO3.FeSO3.2H2O, Cu2SO3.MnSO3.2H2O and Cu2SO3.CdSO3.2H2O are similar to that of Chevreul's salt and may be assigned in a similar way.20,21 The vibrational bands of the sulfite group were assigned by comparing them with data reported in the literature. The four fundamental bands of the sulfite ion vibrations were identified.
Copper(II) and copper(I) have different bonding characteristics. Copper(I) is a class b metal or soft base and follows the trend of having increased polarizability over Cu(II) and of having a decreased number of coordination sites (four, instead of five or six). Copper(I) preferentially complexes with soft sulfur-containing ligands. Copper(II) is sometimes classified as class b, but it can show characteristics of both classes (a or b).23-25 In CuI2SO3.CuII SO3.2H2O, Cu(I) is bonded to SO32- through both oxygen and sulfur bonds, while Cu(II) is bonded through oxygen bonds and its hard character prevails. Thus, considering M(II) cations of the isomorphic series, the hardness order is: Cu(II) > Fe(II) > Mn(II) > Cd(II). Pearson26 proposed that a hard acid prefers to combine with a hard base, and a soft acid prefers to combine with a soft base; hence, since M(II) is bonded only through oxygen, which is a hard acid, a preference is expected for the formation of Cu2SO3.CuSO3.2H2O.
Selectivity evaluation studies20 of the incorporation of M(II) during the formation of salt have shown that, contrary to expectation, the replacement of Cu(II) by M(II) takes place preferentially in decreasing order of hardness. Sghyar et al.22 calculated the unit cell volumes of several double sulfites, considering isomorphic substitution (Figure 4). It is worth noting that, in the incorporation of M(II), the unit cell expands proportionally to the increase in ionic radius, confering a supplementary stability on the sulfite network. This fact may justify the preference for Cu(II) replacement by the soft cation. Nevertheless, this replacement only occurs if there is an excess of M(II) cations.
3. Considerations About Synthesis
Chevreul's salt has been synthesized by various procedures. Up to the 1960s, the most common preparation method described for this substance in the literature consisted of passing sulfur dioxide through an aqueous solution of Cu(II) sulfate at about 70 oC for two hours,2,27 after which the gas stream was interrupted and a brick red solid precipited. In 1967, Scarbo et al.18 described a new synthesization procedure to obtain Chevreul's salt from diluted Cu(II) solution, replacing the prolonged heating time by adding an alkali solution. In that protocol, Chevreul's salt is prepared by precipitation from a copper sulfate solution (1 g/100 mL) by the passage of SO2 at ambient temperature. After saturation with sulfur dioxide, the solution temperature is raised to 78 ± 2 oC and the pH value is adjusted to 3.0 3.5 with the addition of a hydroxide or carbonate solution, at which point Chevreul's salt precipites.
The isomorphic series was prepared according to Skarbo's procedure, adding, besides Cu(II) sulfate, an excess of M(II) sulfate to each of the initial solutions. The compositions of the starting solutions are given in Table 1. Further details of the steps involved in isomorphic series synthesis are discussed elsewhere.18-22
When carbonate or hydroxide solution is used as a neutralizing agent, a substantial increase of the pH occurs at the point where the drop reaches the solution, causing the cation, usually Na+, K+ or NH4+, to become incorporated as an impurity in the sulfite structure. In the specific case when alkali is a sodium carbonate solution, Na+ can replace Cu+ in the crystalline structure because these two cations have the same charge, +1, and their ionic radii are very similar (Na+, 97 pm and Cu+, 96 pm).
To avoid incorporation of the base cation, Silva et al.20 suggested two alternative methods to raise the pH, replacing the addition of alkali with the removal of excess SO2: i) by reducing the internal pressure of the system associated with the ultrasonic bath at 60 oC; ii) by burbling N2 in the sulfur dioxide saturated solution.
In both cases, the salts obtained displayed high degrees of purity. While the traditional method (the addition of Na2CO3) leads to a contamination by Na+ of about 0,8% (m/m), the use of the new procedure results in sodium concentrations at blank levels.20
4. Redox Chemistry of Cu(II)-S(IV) Complexes
Transition metal complexes are important in aqueous atmospheric systems (e.g., cloud droplets and haze aerosols), where trace metal catalysis of the autoxidation of S(IV) has been postulated as an important pathway for SO2 conversion.28-41 The Cu(II)-S(IV) system was proposed by Conklin and Hoffmann29 for use as a model for transition metals that are active catalysts for SO2 autooxidation. This system is similar to the preparation of Chevreul's salt.
According to Conklin and Hoffmann,30 copper(II) forms transient complexes with S(IV) in solution. These complexes react to form a mixture of Cu(I), Cu(II), SO42- and SO32- in apparent equilibrium with Chevreul's salt (CuI2SO3.CuII SO3.2H2O). However, the reduction of Cu(II) by S(IV), in terms of the predominant Cu(II) and S(IV) species, is thermodynamically unfavorable:
The unfavorable DGo for this reaction indicates that other processes must be taken into account. These processes include the stabilization of the products and the energy released by precipitation.
A mechanism for the anoxic reaction between Cu(II) and S(IV) in the formation of Chevreul's salt was suggested by Conklin and Hoffmann30 observed that positively charged dimeric Cu(II)-S(IV) complexes are formed upon mixing solutions of Cu(II) and S(IV); the formation of Cu(II)-S(IV) complexes is pH-dependent; the rate of disappearance of the complexes follows biphasic first-order kinetics; two-electron transfer appears to occur with the formation of S(VI); an equilibrium mixture of Cu(II), Cu(I), S(IV), and S(VI) with precipitated CuI2SO3.CuII SO3.2H2O is established.
A mechanism that is consistent with the above observations is:30
5. Oxidation by Air
There is evidence that a significant portion of the airborne sulfur species emitted into the atmosphere may be particulate SO32-. Some studies indicate that solid metal sulfite and sulfite complex in solution may play a role as intermediaries in atmospheric corrosion processes.38 The kinetics of surface oxidation in particles of Cu2SO3.CuSO3.2H2O, Chevreul's salt, was monitored by X-ray photoelectron spectroscopy (XPS) and used as a model compound for characterizing sulfite stability on particle surfaces exposed to air.31 Several surface reactions of Chevreul's salt exposed to air, including possible oxidation of S(IV) into S(VI) and Cu(I) to Cu(II), were considered. Since no SO42- was detected by XPS in the final products, the reaction below is most likely to occur:
This is consistent with the conclusions of Eatough et al.38 that atmospheric S(IV) species, such sulfites, may have long residence times in the environment.
6. Optical Properties
The isomorphic species present distinct colors. The intense red of Chevreul's salt changes to yellow-brown after Cu(II) is replaced by Fe(II). If the substitution involves Mn(II) instead of Fe(II), a gradual transition to yellow is observed. On the other hand, if Cu(II) is replaced by Cd(II), a faint yellow color is obtained.
The spectroscopic properties of Chevreul's salt were investigated recently by Inoue et al.9 based on EPR measurements and diffuse reflectance absorption spectra. A more complete study was made by Silva et al.,42 involving the electronic interactions between the constituents of Chevreul's salt derivatives, particularly insofar as the types of electronic transition responsible for their different colors are concerned, with emphasis on the theoretical evaluation employing molecular modeling calculations.43-46
The spectrum of Chevreul's salt and those of the corresponding derivatives with Fe(II), Mn(II) and Cd(II) ions are shown in Figure 5.
The electronic spectrum of Chevreul's salt consists of a charge-transfer band around 425 nm associated with the [CuI2(SO3)2(SO3)2] 6- chromophore,42 and two ligand field transitions at 785 and 1000 nm involving the Jahn-Teller splitting of the Cu(II) levels, an effect that is associated with d9 ions. An additional intervalence-transfer band, responsible for its red color, can be found at 500 nm. The replacement of the Cu(II) ions by Fe(II), Mn(II) and Cd(II) does not eliminate the absorption band at 425 nm, supporting its assignment as a charge-transfer transition centered on the Cu(I) sites. The less intense band at 860 nm for the Fe(II) derivative is consistent with a Laporte forbidden ligand field transition, while the bands at 820 nm in the case of Mn(II) and at 725 nm in that of Cd(II) have been tentatively attributed to charge-transfer transitions in the [CuI2(SO3)2(SO3)2]6- chromophore, considering the lack of characteristic electronic transitions for these ions in this region.
7. Thermal Properties
To gain a better understanding of the chemical nature of mixed valence compounds, Silva et al.47 investigated several thermal properties, such as thermal diffusivity (a), heat capacity (rcP), and thermal conductivity (k), as well as the non-radiative relaxation time (tR) and the characteristic diffusion time (tb) of the excited state of ions in the double sulfites of the isomorphic series, using photoacoustic spectroscopy (PAS) and the rising temperature method.48-52
The results in Table 3 show the thermal properties determined by photothermic techniques.47 The thermal diffusivities have values one order of magnitude higher than some semiconductors and porous silicon.53-55
The overall behavior of the thermal parameters seems to reflect the change in the sample structure. It is well known that an amorphous phase is characterized by a slight thermal diffusivity. On the other hand, the high crystallinity of these compounds results in their high diffusivity. In this case, as expected, the heat flows faster through crystalline materials. It is noteworthy that the thermal conductivity changes according to the ionic radius of M(II) cations. Their behavior is related to increases in the unit cell volume (Figure 4). Figure 6 shows the results of thermal conductivity as a function of the ionic radius.
The trend indicated in the graph seems to indicate a linear dependence, with the thermal conductivity decreasing as the ionic radii of M(II) cations [Cu(II) < Fe(II) < Mn(II) < Cd(II)], except Cu2SO3.CuSO3.2H2O, increase. The electronic spectrum of Chevreul's salt discussed previously42 shows an intervalence transfer transition band at 500 nm (see Figure 6) from the tetrahedral Cu(I) ions to the empty 3dx2-y2 level of eg symmetry of the Cu(II) ions. This band, in addition to being responsible for the salt's red color, is evidence of the strong Cu(I)-Cu(II) interaction and explains the linear correlation (R = 0.9992) observed when one considers the weighted average (88 pm) of the ionic radius of Cu(I) and Cu(II) (Figure 7).56
The information concerning the non-radiative relaxation time (tR) and characteristic diffusion time (tb) was obtained from measurements of the PA phase angle as a function of the modulation frequency of the maximum absorbance bands (see electronic spectra Figure 6) using the thermal diffusivity model51 for a thermally thick sample.47
The photoacoustically measured relaxation time does not necessarily constitute an appropriate representation of the level initially excited in the absorption process, but rather of the average lifetime of a variety of states before energy is lost in the form of heat. Thus, in general, there is a non-unique relaxation pathway consisting of several successive steps, and the measured lifetime is an average lifetime of heat production. For a powder sample, tR also contains a contribution from the heat exchange time between the powder particle and the transducing gas of the PA cell.
8. Thermal Decomposition
The isomorphic series has been subjected to thermal treatment in helium, nitrogen, air and reductive gaseous mixtures (H2/Ar) in dynamic atmospheres. The studies revealed that double sulfites are thermally stable up to 200oC under all conditions.21 Based on the characterization of solid intermediary products and to TG and DSC curve analyses, Silva et al.21 suggested the thermal decomposition reactions of the double sulfites in air and nitrogen atmospheres presented in Scheme 1:
The authors21 found that the sulfite ion coordination structures strongly influence the course of these salts' thermal decomposition (Figure 8). Sulfite species coordinated to the metal through the oxygen atom (a) are more easily oxidized into sulfate than sulfur coordinated species (b),7,21 leading to the preferential formation of MIISO4 and Cu2O at the onset of thermal decomposition.21
The presence of Cu2O in the solid residue indicates that the evolution of SO2 during the first step of thermal decomposition takes place through the break of Cu(I)-S and Cu(I)O-S bonds.21
Silva et al.21 have reported that the TG curves of all the double sulfites investigated show a weight gain in the second step of thermal decomposition in inert (N2 and He) and oxidative atmospheres (air) and that they disappear when the salts are heated in a reductive atmosphere (H2/Ar). This weight gain has been ascribed to the oxidation of Cu2O to CuO and/or oxidation of MIISO3 to MIISO4, probably by a self-generated atmosphere composed of dense volatile products released in the former step. The oxidation reaction is inhibited in a reductive atmosphere.
9. Final Considerations
Although interest in these types of compounds dates back to the early 19th century, most of the information about double sulfites found in the literature was published after the 1960s. However, these studies have been intensified in recent years, especially in our group, which has drawn a profile of the isomorphic series Cu2SO3.MSO3.2H2O [M = Cu, Fe, Mn and Cd] in order to gain a better understanding of the chemistry of these compounds.
CNPq and FAPESB for financial support.
1. Chevreul, M. E. ; Ann. Chim. 1812, 83, 187. [ Links ]
2. Kierkgaard, P.; Nyberg, B.; Acta Chem. Scand. 1965, 19, 2189. [ Links ]
3. Dunaj-Jurco, M.; Ondujovic, G.; Melnic, M.; Coord. Chem. Rev. 1988, 83, 1. [ Links ]
4. Nygerg, B.; Larson, P.L.O.; Acta Chem. Scand. 1973, 27, 63. [ Links ]
5. Zachariasen, W. H.; Buckley, H. E.; Phys. Rev., 1931, 37, 1291. [ Links ]
6. Évans, J. C.; Bernstein, H. J.; Can. J. Chem. 1955, 33, 1270. [ Links ]
7. Hall, J. P.; Griffith, W. P.; Inorg. Chim. Acta. 1981, 48, 65. [ Links ]
8. Brant, P.; Fernado, Q.; J. Inorg. Nucl. Chem. 1978, 40, 235 [ Links ]
9. Inoue, M.; Grijalva, H.; Inoue, M. B.; Fernando, Q.; Inorg. Chim. Acta 1999, 295, 125. [ Links ]
10. Larsson, L. O.; Acta Chem. Scand. 1969, 23, 581. [ Links ]
11. Nyberg, B.; Kierkgaar, P.; Acta Chem. Scand. 1968, 22, 581. [ Links ]
12. Hjerten, I.; Nyberg, B.; Acta Chem. Scand. 1973, 27, 345. [ Links ]
13. Lutz, H. D.; El Suradi, S.; Z. Anorg. Allg. Chem. 1976, 425, 134. [ Links ]
14. Sneed, M. C.; Maynard, J. L.; Brasted, R. C.; Comprehensive Inorganic Chemistry, Van Nostrand: New York, 1954, vol. 2. [ Links ]
15. Dasent, W. E.; Morrison, D. ; J. Inorg. Nucl. Chem. 1964, 26, 1122. [ Links ]
16. Stromeyer, A.; Achwefligsaure Kupferoxydul, Liebigs Analen 1959, 109, 237. [ Links ]
17. Skarbo, R. R.; US Pat. 3.515.512 1970. [ Links ]
18. Cipriano, C.; MSc. Dissertation, Universidade Federal do Rio de Janeiro, Brazil, 1981. [ Links ]
19. Miguel, A. H.; de Andrade, J. B.; Cot, L.; Duran, J.; Proceedings of the II Simpósio Nacional de Química Inorgânica, São Paulo, Brazil, 1984. [ Links ]
20. Silva, L. A.; Correia, S. J.; Martins, C. R.; de Andrade, J. B.; Quim. Nova 1998, 21, 151. [ Links ]
21. Silva, L. A.; Matos J. R.; de Andrade, J. B.; Thermochim. Acta 2000, 360, 17. [ Links ]
22. Sghyar, M.;, Durand, J.; Miguel, A. H.; Cot, L.; Rev. Chim. Miner. 1984, 21, 710. [ Links ]
23. Gray, H. B.; Chemical Bonds: An Introduction to Atomic and Molecular Structure, 1st ed., W.A. Benjamin Inc.: New York, 1973 [ Links ]
24. Huheey, J. E.; Keiter, E. A.; Keiter, R.; Inorganic Chemistry: Principles of Struture and Reactivity, 4th ed., Harper Collins College Publishers: New York, 1993. [ Links ]
25. Porterfield, W. W.; Inorganic Chemistry: A Unified Approach, 1st ed., Addison-Wesley: Reading Massachusetts, 1984. [ Links ]
26. Pearson, R. G.; J. Am. Chem. Soc. 1963, 85, 3533. [ Links ]
27. Pascal, P.; Nouveau Traité de Chimie Minerale 1957, 111, 298. [ Links ]
28. Conklin, M. H.; Hoffmann, M. R.; Environ. Sci. Technol. 1988, 22, 883. [ Links ]
29. Conklin, M. H.; Hoffmann, M. R.; Environ. Sci. Technol. 1988, 22, 891. [ Links ]
30. Conklin, M. H.; Hoffmann, M. R.; Environ. Sci. Technol. 1988, 22, 899. [ Links ]
31. Cox, X. B.; Linton,R. W.; Miguel, A. H.; de Andrade, J. B.; Atmos. Environ. 1986, 20, 1139. [ Links ]
32. Hoffmann, M. R.; Atmos. Environ. 1986, 20, 1145. [ Links ]
33. Radojevic, M.; Tyler, B. J.; Wicks, A. J.; Gay M. J.; Choularton, T. W.; Atmos. Environ. 1990, 24A, 323. [ Links ]
34. Xue, H.; Gonçalves, M. L. S.; Reutlinger, M.; Sigg, L.; Stumm, W.; Environ. Sci. Technol. 1991, 25, 1716. [ Links ]
35. Grgic, I.; Hudnik, V; Bizjak, M.; Levec, J.; Atmos. Environ. 1991, 25A, 1591. [ Links ]
36. Martin, L. R.; Good, T. W.; Atmos. Environ. 1991, 25A, 2395. [ Links ]
37. Sedlak D. L.; Hoigné, J.; Environ. Sci. Technol. 1994, 28, 1898. [ Links ]
38. Eatough, D. J.; Christensen, J. I.; Eatough, N. L.; Hill, M. W.; Major, T. D.; Mngelson, N. F.; Post, M. E.; Ryder, J. F.; Hansen, L. D.; Atmos. Environ. 1982, 16, 1001. [ Links ]
39. de Andrade, J. B.; Pinheiro, H. C.; Andrade, M. V. S.; J. Braz. Chem. Soc. 1995, 6, 287. [ Links ]
40. Martins, C. R.; Cabral Neto, C. A.; Alves, J. J. F.; de Andrade, J. B.; J. Braz. Chem. Soc. 1999, 10, 45. [ Links ]
41. Martins, C. R.; de Andrade, J. B.; Quim Nova 2002, 25, 259. [ Links ]
42. Silva, L. A.; de Andrade, J. B.; Toma E. H.; J. Braz. Chem. Soc. 2002, 13, 624. [ Links ]
43. Zerner, M. C.; Loew, G. H.; Kirchner, R. F.; Mueller-Westerhoff, U. T.; J. Am. Chem. Soc. 1980, 102, 589. [ Links ]
44. Bacon, A. D.; Zerner, M. C.; Theoret. Chim. Acta 1979, 53, 21. [ Links ]
45. Ridley, J. E.; Zerner, M. C.; Theoret. Chim. Acta 1976, 42, 223. [ Links ]
46. Ridley, J. E.; Zerner, M. C.; Theoret. Chim. Acta 1973, 32, 111. [ Links ]
47. Silva, L. A.; de Andrade J. B.; Moysés Araújo, C.; Nakamura, O.; Ferreira da Silva, A.; Massunaga, M. S. O.; Vargas, H.; Phys. Chem. Chem. Phys. 2001, 3, 4800. [ Links ]
48. Lima, G. A. R.; Baesso, M. L.; Arguello, Z. P.; da Silva, E. C.; Vargas, H.; Phys. Rev. B 1987, 36, 9812. [ Links ]
49. Baesso, M. L.; Mansanares, A. M.; da Silva, E. C.; Vargas, H.; Phys. Rev. B 1989, 40, 1880. [ Links ]
50. Abritta, T.; Cella, N.; Vargas, H.; Chem. Phys. Lett. 1989, 161, 12. [ Links ]
51. Rosencwaig, A.; Gersho, A.; J. Appl. Phys. 1976, 47, 46. [ Links ]
52. Alexandre, J.; Saboya, F.; Marques, B. C., Ribeiro; M. L., Salles, C.; da Silva, M. G., Sthel, M. S.; Auler, L. T.; Vargas, H.; Analyst 1999, 124, 1209. [ Links ]
53. Calderon, A; Alvarado-Gil, J. J.; Garevich, Y. G.; Cruz-Orea, A.; Delgadilho, I.; Vargas, H.; Miranda, L. C. M.; Phys. Rev. Lett. 1997, 79, 5022. [ Links ]
54. Silva, T. S.; Alves, A. S.; Pepe, I.; Tsuzuki, H.; Nakamura, O.; d'Aguiar Neto, M. M. F.; Ferreira da Silva, A.; Veissid, N.; An, C. Y.; J. Appl. Phys. 1998, 83, 6193. [ Links ]
55. Ferreira da Silva, A.; Souza da Silva, T.; Nakamura, O.; d'Aguiar Neto, M. M. F.; Pepe, I.; Roman, L. S.; Veje, E.; Mater. Res. 2001, 4, 23. [ Links ]
56. Silva, L. A.; Vargas, H.; Massunaga, M. S. O.; Ferreira da Silva, A.; de Andrade J. B.; Abstracts of the 26a Reunião Anual da Sociedade Brasileira de Química, Poços de Caldas, Brazil, 2003. [ Links ]
Received: June 17, 2003
Published on the web: April 6, 2004