Design of an NO photoinduced releaser xerogel based on the controlled nitric oxide donor trans-[Ru(NO)Cl(cyclam)](PF6)2 (cyclam=1,4,8,11-tetraazacyclotetradecane)

This article was originally published in a journal published byElsevier, and the attached copy is provided by Elsevier for the

author’s benefit and for the benefit of the author’s institution, fornon-commercial research and educational use including without

limitation use in instruction at your institution, sending it to specificcolleagues that you know, and providing a copy to your institution’s

administrator.

All other uses, reproduction and distribution, including withoutlimitation commercial reprints, selling or licensing copies or access,

or posting on open internet sites, your personal or institution’swebsite or repository, are prohibited. For exceptions, permission

may be sought for such use through Elsevier’s permissions site at:

http://www.elsevier.com/locate/permissionusematerial

Autho

r's

pers

onal

co

py

Journal of Colloid and Interface Science 300 (2006) 543–552www.elsevier.com/locate/jcis

Design of an NO photoinduced releaser xerogel based on the controllednitric oxide donor trans-[Ru(NO)Cl(cyclam)](PF6)2

(cyclam = 1,4,8,11-tetraazacyclotetradecane) ✩

Kleber Queiroz Ferreira a, José F. Schneider b, Pedro A.P. Nascente c,Ubirajara Pereira Rodrigues-Filho d, Elia Tfouni a,∗

a Departamento de Química, Faculdade de Filosofia, Ciências e Letras de Ribeirão Preto, Universidade de São Paulo, Av. dos Bandeirantes 3900,14040-901 Ribeirão Preto, SP, Brazil

b Insituto de Física de São Carlos, Universidade de São Paulo, CP 369, 13560-970 São Carlos, SP, Brazilc Departamento de Engenharia de Materiais, Universidade Federal de São Carlos, 13565-905 São Carlos, SP, Brazil

d Instituto de Química de São Carlos, Universidade de São Paulo, CP 780, 13560-970 São Carlos, SP, Brazil

Received 21 February 2006; accepted 28 March 2006

Available online 6 April 2006

Abstract

The immobilization and properties of the nitric oxide donor trans-[Ru(NO)Cl(cyclam)](PF6)2, Ru–NO, entrapped in a silica matrix by the sol–gel process is reported herein. The entrapped nitrosyl complex was characterized by spectroscopic (UV–vis, infrared (IR), X-ray photoelectron,and 13C and 29Si MAS NMR) and electrochemical techniques. The entrapped species exhibit one characteristic absorption band in the UV–visregion of the electronic spectrum at 354 nm and one IR νNO stretching band at 1865 cm−1, as does the Ru–NO species in aqueous solution. Ourresults show that trans-[Ru(NO)Cl(cyclam)](PF6)2 can be entrapped in a SiO2 matrix with preservation of the molecular structure. However, ina SiO2/SiNH2 matrix, the complex undergoes a nucleophilic attack by the amine group at the nitrosonium. Irradiation of the complex, entrappedin the SiO2 matrix, with light of 334 nm, resulted in NO release. The material was regenerated to its initial nitrosyl form by reaction with nitricoxide.© 2006 Elsevier Inc. All rights reserved.

Keywords: Nitric oxide; Sol–gel; Xerogel; Ruthenium; Nitrosyl; Controlled; Photochemistry; Cyclam; Silica; Aminopropylsilica; Nucleophilic; Attack

1. Introduction

In mammalian species, nitric oxide (NO) plays key roles inalmost every function [1], where high or low NO concentrationscan be either beneficial or harmful and could accompany nu-merous pathological states [1]. For this reason, there has been agrowing interest in NO donors and scavengers aiming at thera-peutic applications [2–17]. Ruthenium nitrosyl complexes haveshown to be very promising NO donors [5,6,9,10,12,14,17–25],and some of them have shown biological activity [6,22,24–27].

✩ Taken in part from K.Q. Ferreira, Ph.D. thesis, Departamento de Químicada Faculdade de Filosofia, Ciências e Letras de Ribeirão Preto, Universidade deSão Paulo, 2004.

* Corresponding author. Fax: +55 16 3602 4838.E-mail address: [email protected] (E. Tfouni).

In these quite stable complexes, the coordinated NO has a ni-trosonium character, and it can be released photochemically orvia one-electron reduction [3,12,17–20,22,26,28].

Our laboratories have directed efforts toward the synthesisof ruthenium complexes as NO donors and scavengers [6,14,23,24,26]. The trans-[Ru(NO)Cl(cyclam)](PF6)2 complex releasesNO photochemically or upon reduction [14,29,30], and it isless toxic than nitroprusside, a well-known vasodilator [6,9,14].Furthermore, it is as effective at reducing blood pressure as ni-troprusside, but with a longer effect [6]. The blood pressureeffects were interpreted in terms of the reactivities of the com-plexes involved in NO release. The longer blood pressure re-duction effect of trans-[Ru(NO)Cl(cyclam)]2+ was interpretedas a result of the much lower rate of NO release from trans-[Ru(NO)Cl(cyclam)]2+ than from similar tetraammine nitrosyl

0021-9797/$ – see front matter © 2006 Elsevier Inc. All rights reserved.doi:10.1016/j.jcis.2006.03.081

Autho

r's

pers

onal

co

py

544 K.Q. Ferreira et al. / Journal of Colloid and Interface Science 300 (2006) 543–552

ruthenium complexes [6,9,14]. Aiming at extending the rangeof potential applications, there is an interest in designing carri-ers or supporters for such complexes. Conceivably, the immo-bilization of these complexes could result in materials that maybe used in association with optical fibbers to provide the op-portunity for controlled NO release at specific target sites usinglaser photoexcitation [31,32].

In this regard, novel strategies using NO donors other thanmetal nitrosyl complexes have also been investigated. Nitricoxide-releasing diazeniumdiolates are successfully being im-mobilized in polymers, silica-gel, and metal surfaces, aimingat biological applications [33]. Recently, the preparation, char-acterization, and preliminary biomedical application of variousnitric oxide (NO)-releasing fumed silica particles with aminegroups has been reported [34]. These amine groups were thenconverted into the corresponding N-diazeniumdiolate groupsvia reaction with NO(g) at high pressure in the presence ofmethoxide bases. The N-diazeniumdiolate moieties attached tothe silica surface underwent a primarily proton-driven dissoci-ation to NO under physiological conditions, and they also un-derwent slow thermal dissociation to NO. These resulting NO-releasing fumed silica particles could be embedded into poly-mer films to create thromboresistant coatings, via NO release atfluxes that mimic healthy endothelial cells (EC), making thema very interesting system. The NO-addition efficiency for thisdirect reaction, however, was found to be 12% in an acetonitrilesuspension of Sil-2N [6] particles. This NO-loading capacity islower than the one observed for various free amines, which leadto a typical yield of 30–90% [35]. The immobilization of diaze-niumdiolates in sol–gel to yield NO releasing materials has alsobeen reported [36]. More recently, a ruthenium salen nitrosylcomplex has been copolymerized with ethyleneglycol dimethy-lacrylate to form a material which is photolabile for NO release[37]. Toma et al. have also recently reported ruthenium hexaac-etate clusters incorporated in polyvinyl alcohol films that aresensitive to daylight [38].

The use of ruthenium nitrosyl immobilized species as NOdonors has some advantages in relation to other species, sinceruthenium complexes can deliver and recover NO at milder con-ditions than those associated with the diazeniumdiolate system.Moreover, it should be noted that the NO delivery from ruthe-nium amine (or ammine) nitrosyl complexes can be tuned pho-tochemically, through their different UV–vis spectra or by theirdifferent reduction potentials [3,12,14,17,20,21,24,26,39,40],and thus, the choice of complex can be made based on thetarget and conditions. These NO donors attached to solid statematrices can be achieved by two different approaches: (a) graft-ing or physical adsorption of the complex on a matrix alreadyprepared; (b) occlusion, where the complex is mixed with pre-cursors of the matrix which is formed around the complex. Thematrix must be inert toward the complex and the NO released,and it should also be chemically stable. Furthermore, in orderto keep the photochemical NO release, the matrix should notabsorb in the same wavelength range of the complex. Althoughan organic matrix [37] can be envisaged, a xerogel matrix hasthe advantage of exhibiting higher chemical and physical iner-tia, as well as displaying lower or zero absorbance in the near

UV and visible region. Sanchez et al. have already reviewed theadvantages and challenges of using hybrid xerogels for opti-cal applications, and they clearly showed the feasibility of theiruse in such applications [41]. The first attempt to immobilize aruthenium complex with potential ability to act as NO donorwas achieved by Franco et al. They used the chemisorptionof trans-[Ru(NH3)4(SO2)(H2O)]2+ on 3-(L-imidazolyl)propylorganomodified silica gel to prepare a ruthenium complexmodified silica gel [–Si(CH2)3imN-Ru(NH3)4SO2] [42]. Thismethod has the advantage of leading to a chemical bond be-tween the ruthenium complex and the silica gel, possibly lead-ing to a more stable material from the recycling point of view.However, the ruthenium loading in these materials is low evenfor relatively small complexes like the ruthenium ammine com-pounds because they are not able to diffuse inside the innersilica pores. Therefore, other methodologies should be used forthe preparation of heavily loaded ruthenium nitrosyl silicas.

As suggested before, the immobilization by sol–gel entrap-ment/occlusion in silica matrices can be a better choice [43–46].The mild characteristics offered by the sol–gel process allowthe introduction of inorganic complexes inside an inorganic net-work [47]. The sol–gel methodology has so far been used inthe context of inorganic catalysts, as part of the matrix [48],as supports for dispersed metal particles [49], and for copoly-merization with suitable silicon-containing ligands [50]. Theintroduction of a host molecule is obtained by adding its so-lution to the polymerizing mixture. When the polymerization iscomplete, the dopant molecules are entangled in the inorganicpolymeric network. The nature of the entrapment is still notfully understood, and it is really remarkable to see how manyapplications of the entrapment have been developed, without afull understanding of the process at the molecular level [51].The entrapment of ruthenium nitrosyls can conceivably lead tochanges in kinetic properties, such as rate of release of NO,which would probably lower than in solution. Similarly, Avnirand Frenkel-Mullerad [52] recently studied the unusual stabi-lization of alkaline and acid phosphatases occluded in xerogels.Remarkably, the enzymes kept their activity even at pH as lowas 0.9. The explanation proposed for such a high stability tookinto consideration the porous microenvironment in xerogels at amolecular level. The restricted space inside these pores seems tochallenge the classical meaning of thermodynamic parameterslike pH, and a nanoscopic view of the interactions inside thepores among the surface groups, i.e., silanols, adsorbed waterand entrapped species, seems to be more appropriate. There-fore, the study of chemical reaction on largely restricted media(LRM) needs a different approach from that used in bulk solu-tion chemistry.

In this context and considering that ruthenium nitrosyl com-plexes allow the possibility of tuning the NO donor properties[6,14,23,24,26,30], we have extended our investigations to theirimmobilization for possible therapeutic applications. In this re-gard, initial studies on the immobilization of the rutheniumnitrosyl complex, [Ru(salen)(OH2)(NO)]+ (salen = N,N ′-bis-(salicylidene)ethylenediaminato), impregnated into a silica sol–gel have indicated that it releases NO under irradiation withlight and it can also be regenerated [53].

Autho

r's

pers

onal

co

py

K.Q. Ferreira et al. / Journal of Colloid and Interface Science 300 (2006) 543–552 545

In this paper, we describe the immobilization and character-ization of the controlled NO donor trans-[Ru(NO)Cl(cyclam)]-(PF6)2 entrapped in xerogels containing tetraethylorthosilicate(TEOS), and 3-aminopropyltriethoxysilane (3-APTS), by thesol–gel process. The reaction of 3-aminopropyltriethoxysilanewith the coordinated nitrosonium of the complex, the photo-chemical release of NO from the SiO2 material, and the re-generation of the ruthenium nitrosyl complex entrapped in thematrix are also described.

2. Experimental

2.1. Chemicals and reagents

Ruthenium trichloride (RuCl3·nH2O) (Strem) was the start-ing material for the synthesis of the ruthenium complexes.Acetone, acrylonitrile, chloroform, and ethanol were purifiedaccording to procedures published in the literature [54]. Dou-bly distilled water was used throughout. Tetraethylorthosilicate(TEOS) and 3-aminopropyltriethoxysilane (3-APTS) (Aldrich)were used without further purification. All other materials werereagent grade and were used without further purification.

2.2. Complex syntheses

Trans-[RuCl(tfms)(cyclam)](tfms) (Rutfms), trans-[RuCl2-(cyclam)]Cl (RuCl), and trans-[Ru(NO)Cl(cyclam)](PF6)2(RuNO) were synthesized by using a published procedure [14].

2.2.1. Preparation of the entrapped complextrans-[Ru(NO)Cl(cyclam)](PF6)2

The ruthenium complex (25 mg; 3.8 × 10−5 mol) was dis-solved in a hydrolytic solution containing TEOS (2.5 mL) inethanol:water (4:1 v/v) and 0.1 M HCl (0.4 mL). TEOS hydrol-ysis resulted in the formation of silanol groups (Si–OH) [51].These silanol moieties reacted further among them to formsiloxane (Si–O–Si) oligomers in a condensation reaction, lead-ing to the formation of a colloidal suspension (sol). Finally,the solvents were removed from the interconnected porous net-work during a three-day drying process at 50 ◦C, leading to2.7 g of a dry vitreous material, the xerogel SiO2/RuNO. Thexerogel with amino groups, SiO2/SiNH2/RuNO, was obtainedusing the desired amount of Ru complex and TEOS (1.7 mL,8.5 mmol) mixed with 3-APTS (0.83 mL, 4.0 mmol) insteadof the pure TEOS solution. The xerogels were characterizedby optical and scanning electronic microscopy; X-ray photo-electron, UV–visible, infrared, and 29Si and 13C MAS NMRspectroscopies; and electrochemical techniques.

2.2.2. Reaction of trans-[Ru(NO)Cl(cyclam)](PF6)2 with3-aminopropyltrietoxysilane

The reaction of the complex with aminopropyltrietoxysilanewas monitored by three spectroscopic techniques: (1) diffusereflectance infrared spectroscopy, using the decrease of the in-tensity of the νNO band at 1865 cm−1, and the νNO/νSiO peaksareas ratios; (2) electronic, using the increase of the absorbanceof the absorption band at 484 nm; (3) 13C NMR, using various

proportions of complex, aminopropyltrietoxysilane and tetrae-toxysilane. The 13C NMR spectroscopy studies were carriedout in the solid state and in solution by bubbling NO(g) inaminopropyltrietoxysilane dissolved in CDCl3. Also, 13C NMRspectra of the free complex and in the presence of aminopropyl-trietoxysilane were obtained.

2.2.2.1. Electrochemical measurements Cyclic voltammetryand differential pulse voltammetry experiments were taken witha model 273 PARC potentiostat/galvanostat, using a conven-tional three-electrode cell consisting of a modified carbon paste,an Ag/AgCl, and a platinum wire as the working, reference andauxiliary electrodes, respectively. The voltammetric spectra ofthe complex on the matrix (carbon paste mixture) was hinderedby the work potential range of the carbon paste (−1.1 to 1.1 Vversus Ag/AgCl) [55]. The measurements were carried out at25 ◦C. E′

1/2 values for the redox process of the nitrosyl ligandin the immobilized complex were determined. The E′

1/2 valueswere the arithmetic means of the Epa and Epc values.

2.2.2.2. Electronic, vibrational, and NMR spectra Electronicabsorption spectra of immobilized complex were recorded us-ing a Hewlett–Packard model 8452A recording spectropho-tometer. Since the refraction index of carbon tetrachloride andthe silica matrix are nearly the same [56], the supported com-plex on the silica gel surface was immersed in a spectra gradecarbon tetrachloride and the spectra of the suspension wereobtained using a quartz cell of 1 mm path length. Diffuse re-flectance infrared spectra were obtained on a MB-102 Bomemspectrophotometer.

13C and 29Si solid-state high resolution NMR spectra wereobtained on a Varian Unity INOVA 400 (9.4 Tesla) spectrometerand a CP/MAS 7 mm Varian probe. The magic angle spinningtechnique (MAS) was used with a spinning frequency of 5 kHz.29Si NMR spectra were measured by applying a single radiofrequency π/2 pulse of 4.5 µs. A recycle time of 400 s wassufficient to ensure complete magnetization recovery for all thesilicon species. For the 13C spectra, the cross-polarization (CP)1H–13C NMR technique was applied with a π/2 1H pulse of4 µs, contact time of 1 ms and recycle time of 5 s. Tetramethyl-silane was used as reference for the obtention of 13C and 29Sichemical shifts.

2.2.2.3. Scanning electron microscopy The xerogel wasspread over a double-side carbon tape and the sample wascoated with a carbon film using a BALTECMed 020 sputterunit. Scanning electron microscopy of the xerogel was recordedusing a LEO 440 microscopy equipped with an Oxford EDS de-tector.

2.2.2.4. X-ray photoelectron spectroscopy The powder sam-ples were prepared by spreading the powder over a carbondouble-side adhesive tape. The charge correction was done us-ing the internal C–Hx C 1s peak at 285 eV and the Si 2p peakof SiO2 at 103.4 eV [57]. A take-of angle of 90◦ was used.The pressure in the analyzer chamber was in the 5 × 10−7 to1 × 10−6 Pa range. A non-monochromatic MgKα radiation

Autho

r's

pers

onal

co

py

546 K.Q. Ferreira et al. / Journal of Colloid and Interface Science 300 (2006) 543–552

(a) (b)

Fig. 1. (a) Topographic, backscattered electron, image of trans-[Ru(NO)Cl-(cyclam)](PF6)2 entrapped in the SiO2 matrix. (b) EDAX mapping based onthe Si κα X-ray emission line.

(1253.6 eV; 180 W), as the X-ray power source, and a KratosXSAM spectrometer were used. The Xp-spectra were fitted toa Gaussian–Lorentzian set of peaks as described before [58].

2.2.2.5. Photochemistry Monochromatic irradiations at 334nm were carried out using a 150 W Xenon lamp in a model6253, Oriel Universal Arc Lamp Source. The irradiation wave-length was selected with an Oriel interference filter for photol-ysis at the appropriate wavelengths. The interference filters hadan average band pass of 10 nm and the collimated beam inten-sities ranged from 1 × 10−9 to 4 × 10−8 einstein−1 cm−2, asdetermined by ferrioxalate actinometry. A sample of the xero-gel in CCl4 was placed in the cuvette and deaerated by bubblingargon before irradiation. The progress of the photoreactions wasmonitored spectrophotometrically on a MB Bomem 102 FTIRSpectrometer, using a ZnSe ATR crystal, or on HP8452A diodearray spectrophotometers, in the cases of in situ vibrational andelectronic spectroscopy, respectively.

3. Results and discussion

3.1. Characterization of the material

The material containing trans-[Ru(NO)Cl(cyclam)](PF6)2entrapped in the SiO2 matrix exhibits a homogeneous colorand smooth surface as judged by optical microscopy. The ho-mogeneous distribution of the Ru–NO complex was verifiedby scanning electronic spectroscopy (SEM) using the EDS de-tector to generate an elemental mapping of Si, and Cl X-rayemission. Unfortunately, it was not possible to collect enoughsignal to study the distribution of trans-[Ru(NO)Cl(cyclam)]2+,so we had to use the Cl κα line to do so. The topographic im-age and EDAX Si and Cl mapping of the SiO2/RuNO xerogel,which is similar to that of SiO2/SiNH2/RuNO, are shown inFig. 1. The Si mapping (Fig. 1b) confirms the siliceous natureof the xerogel. The Cl mapping, not shown, shows an homoge-neous distribution for the Ru complex in the material.

3.2. Elemental analysis

The concentration of ruthenium in the xerogels, as esti-mated from the Cl X-ray emission, is 0.6 × 10−2 mmol g−1,corresponding to 43% of the theoretical maximum value of

Fig. 2. Electronic spectrum in CCl4 of trans-[Ru(NO)Cl(cyclam)](PF6)2 en-trapped in the SiO2 matrix (100 mg of SiO2/RuNO in 10 mL).

Fig. 3. Electronic spectrum in CCl4 of trans-[Ru(NO)Cl(cyclam)](PF6)2 en-trapped in the SiO2/SiNH2 matrix (100 mg of SiO2/SiNH2/RuNO in 10 mL).Solid line: experimental. Dotted line: spectrum fitted with Gaussian compo-nents.

1.4 × 10−2 mmol g−1. It was not possible to estimate the Ncontent by EDS.

3.3. Electronic and infrared spectra

Figs. 2–4 show the electronic and vibrational spectra oftrans-[Ru(NO)Cl(cyclam)](PF6)2 entrapped in the SiO2 andSiO2/SiNH2 matrices.

The electronic and vibrational spectra of trans-[Ru(NO)-Cl(cyclam)](PF6)2 entrapped in the silica matrices indicate thepresence of coordinated nitrosonium (NO+). The electronicspectrum of the free complex in aqueous solution displays twobroad absorption bands in the near UV–visible region. One bandis centered around 352 nm (ε = 190 cm−1 M−1); the other,much weaker (ε = 54 cm−1 M−1), is located around 452 nm.The UV–vis spectra for the complex entrapped in the two dif-ferent matrices are fairly similar to that of the free complexin solution. The absorption band at 352 nm for the free com-plex is shifted to 355 and 350 nm in the case of the SiO2

and SiO2/SiNH2 matrices, respectively. The band at 452 nm

Autho

r's

pers

onal

co

py

K.Q. Ferreira et al. / Journal of Colloid and Interface Science 300 (2006) 543–552 547

Fig. 4. Vibrational spectrum of trans-[Ru(NO)Cl(cyclam)](PF6)2 entrapped inthe SiO2 matrix (SiO2/RuNO).

is barely seen in the matrices, and by fitting the spectra an ab-sorption band can be located at 484 nm for the SiO2/SiNH2matrix. The absence of the former band and the presence of thelatter indicate the absence of the NO+ group. However, it hasbeen observed that nitrosonium can undergo nucleophilic attackfrom an amine in solution [59–61]. As a matter of fact, a rela-tively more intense band appears at 484 nm in the spectrum of amixture of the complex and 3-aminopropyltrietoxysilane duringgel formation as a result of such reaction (see ahead). Thus, therelatively weak absorption at 484 nm in the SiO2/SiNH2 ma-trix may indicate that there is also some amount of a modifiedcomplex in the matrix. This is consistent with the color changefrom yellow to orange in this matrix and with the infrared re-sults.

The νNO band in the infrared spectra of ruthenium nitrosylcomplexes is medium-dependent and can split as a result ofsolid-state effects [30]. For trans-[Ru(NO)Cl(cyclam)](PF6)2,this band appears at 1875 cm−1 in KBr pellet [14,26,30]. In nu-jol mulls this band is split into two bands, 1881 and 1867 cm−1,and in acetonitrile and water it appears as a single absorption at1889 and 1899 cm−1, respectively [30]. For the complex en-trapped in the SiO2 and SiO2/SiNH2 matrices, the νNO bandappears at 1870 and 1854 cm−1, respectively. These differencescould also be conceivably due to possible different microenvi-ronments in the matrix, as already observed for other systems[62–64]. The infrared spectrum of pure silica displays a weakband near 1870 cm−1 assigned to a combination of fundamen-tal silica skeleton vibrations, which may also contribute to ab-sorption in this region in the entrapped complex. However, forSiO2/SiNH2, the νNO band decreases in intensity and there isan increase in the absorption at 1600 cm−1. These changes areconsistent with a nucleophilic attack of the –NH2 group on thecoordinated nitrosonium.

3.4. Magic angle spinning nuclear magnetic resonance

3.4.1. 13C and 29Si solid-state NMRThe 29Si NMR spectrum of the SiO2/RuNO matrix is shown

in Fig. 5. Three broad NMR peaks centered at −111.0 ppm,−102.0 ppm and −91.5 ppm can be observed. They can be

Fig. 5. 29Si NMR spectrum of trans-[Ru(NO)Cl(cyclam)](PF6)2 entrapped inthe SiO2 matrix, SiO2/RuNO. Dotted curves: least-square fittings to the ob-served peaks.

readily attributed to silicon species with different condensa-tion degrees: Q4, Q3 and Q2, respectively [65]. As usual, Qn

indicates the degree of condensation of a given SiO4 tetrahe-dron, being n the number of O involved in Si–O–Si bridges.The silicon species ratios can be obtained from the integratedintensities of the NMR lines. A least square fitting of Gauss(Q4) and Lorentz (Q3 and Q2) functions was carried out toquantify these intensities. From these results, we can concludethat 55% of the silicon present in the matrix correspond toQ4 species, where the polymerization occurred in three dimen-sions, not leaving bound silanol groups. Also, 43% correspondto Q3 (only one silanol group) and less than 2% to Q2 (gemi-nated silanol).

These observations have led us to think that the condensa-tion of the hydrolyzed tetraethylorthosilicate was not hinderedby the presence of the complex in the sol medium, resultingin a highly interconnected three-dimensional network of SiO2.This is very important for the final properties of the mater-ial, since the presence of large numbers of geminal groups oreven of hydrolyzed tetraethylorthosilicate would mean a loss inthe thermal and mechanical stability of the xerogel or wouldeven prevent its formation. Also, the presence of 43% of iso-late silanol groups (Q3) is typical of high surface energy andhigh polar surface. So, we could expect a polar environmentsurrounding the complex, similar to that felt by the complex ina polar solvent like ethanol [66]. This polar environment couldstabilize charge dissociation reaction pathways and facilitate in-duced NO dissociation from the entrapped complex. A similarmatrix effect was observed for pentacyanoferrates anchored onorganomodified silica gel [67] and for molybdenum carbonylsin the intrazeolyte cavity of the NaY zeolite [68].

Figs. 6a and 6b show the 13C CP-MAS NMR spectra oftrans-[Ru(NO)Cl(cyclam)](PF6)2 entrapped in the SiO2 ma-trix and in the SiO2/SiNH2 matrix, respectively. The 13C MASNMR spectrum of SiO2/RuNO display peaks in the 10–70 ppmregion, as observed for the RuNO complex in solution. Thesepeaks are due to the –CH2 carbon of the cyclam ring. However,besides the peaks at 10–70 ppm region, the 13C MAS NMRspectrum of SiO2/SiNH2/RuNO displays a peak at 177 ppm,whose origin is not clear (see ahead). This is consistent withthe UV–vis and infrared results, which give evidence of a pos-

Autho

r's

pers

onal

co

py

548 K.Q. Ferreira et al. / Journal of Colloid and Interface Science 300 (2006) 543–552

Fig. 6. (a) Representative 13C CP-MAS NMR spectrum of trans-[Ru(NO)Cl-(cyclam)](PF6)2 entrapped in the SiO2 matrix, SiO2/RuNO. (b) 13C CP-MASNMR spectrum of trans-[Ru(NO)Cl(cyclam)](PF6)2 entrapped in the SiO2/SiNH2 matrix.

Fig. 7. In situ electronic spectrum of a mixture of trans-[Ru(NO)-Cl(cyclam)](PF6)2 (1 × 10−3 mol L−1) and 3-aminopropyltrietoxysilane(0.1 mol L−1) in acetonitrile. Spectra taken after 5, 10, 15, 60, 90, 100 and130 min of reaction.

sible nucleophilic attack of the –NH2 group on the coordinatednitrosonium [59–61]. This process occurs in the ammino mod-ified xerogel but not in the SiO2 xerogel, because of the nucle-ophilic character of the –NH2 groups.

3.4.2. Nucleophilic attack of 3-aminopropyltrietoxysilane totrans-[Ru(NO)Cl(cyclam)](PF6)2

Figs. 7 and 8 show the in situ UV–vis and infrared ab-sorption spectra of the reaction between the trans-[Ru(NO)Cl-(cyclam)]2+ complex and 3-aminopropyltrietoxysilane. Thespectroscopic monitoring of this reaction shows alterations inthe electronic and vibrational spectra of the complex as indi-cated by an increase in the absorbance of the bands at 350and 484 nm (Fig. 7) due to the reaction products, and a de-crease in the intensity of the stretching band of the coordi-nated NO at 1875 cm−1, accompanied by an increase around1600 cm−1. These spectral changes can be attributed to the nu-cleophilic attack of the –NH2 group to the NO+ coordinated

Fig. 8. In situ vibrational spectrum of a mixture of trans-[Ru(NO)-Cl(cyclam)](PF6)2 (1 × 10−3 mol L−1) and 3-aminopropyltrietoxysilane(0.1 mol L−1) in acetonitrile. Time between successive spectra is 5 min.

to the ruthenium ion, with an estimated half-life of 10 min.Furthermore, the 13C NMR spectrum in d3-acetonitrile of theproduct of the reaction of trans-[Ru(NO)Cl(cyclam)](PF6)2 and3-aminopropyltrietoxysilane displays a peak at 180 ppm, whichis not present in the 13C NMR spectrum obtained in solutionby bubbling NO(g) in 3-aminopropyltrietoxysilane dissolvedin CDCl3 or in the 13C NMR spectrum of the free complexin acetonitrile-d3. These results are consistent with the reac-tion between trans-[Ru(NO)Cl(cyclam)](PF6)2 and the aminegroup, and also with the infrared and UV–visible spectral dataof the complex immobilized in the SiO2/SiNH2 matrix. Theresidual absorption peak at 1875 cm−1 in the infrared spec-trum can be attributed to an incomplete reaction. However,the reaction product of this reaction is not clear. The reac-tion of some nitrosyl complexes with aromatic amines hasbeen reported to be a diazotation reaction [59–61], result-ing in infrared bands around 2000 cm−1 and UV–vis bandsaround 300 nm in the spectra of the products. In the case oftrans-[Ru(NO)Cl(cyclam)](PF6)2, there are no infrared bandsat ∼2000 cm−1 in the xerogel and the band at 484 nm presentin the electronic spectrum of the free complex in solution is notobserved in the xerogel either, possibly because only a partialreaction occurs.

3.5. X-ray photoelectron spectroscopy

In order to study the xerogels by XPS, it was necessary tofirst characterize the non-entrapped Ru-cyclam compounds bythe same spectroscopy. The Ru 2p3/2, N 1s, Cl 2p3/2, and theseparation between N 1s (NO) and O 1s (NO) of the pure trans-[Ru(NO)Cl(cyclam)](PF6)2 and trans-[RuCl2(cyclam)]Cl com-pounds are shown in Table 1. Those of the entrapped complexesare shown in Table 2. The �(N–O) difference observed for thecomplex is about the same as that observed in other M–NOcomplexes [69,70].

From Table 1, it is possible to see that the Ru 2p3/2 peaksof Ru–Cl have higher binding energy than those of trans-

Autho

r's

pers

onal

co

py

K.Q. Ferreira et al. / Journal of Colloid and Interface Science 300 (2006) 543–552 549

Table 1XPS binding energies in eV for the ruthenium nitrosyl complexes synthesized herein and for complexes reported in the literature

Sample Ru 3p3/2 N 1s(cyclam)

N 1s(NO)

Cl 2p3/2(bound)

�(N 1s–O 1s) Ref.

Trans-[RuCl2(cyclam)]+ 463.1 398.7 196.5 a

Trans-[RuCl(NO)(cyclam)]2+ 462.5 399.4 400.9 198.6 131.9 a

[Ru(NO)2(Pφ3)2]4+ 464.5 400.6 132.0 [69][Ru(NO)Cl5]2− 464.6 402.8 [70]

a This work.

Table 2XPS binding energies in eV for the xerogel matrices (SiO2 and SiO2/SiNH2) and for the xerogel matrices containing the entrapped complexes (SiO2/RuNO,SiO2/SiNH2/RuNO)

Sample Si 2p C 1s N 1s O 1s Ru 3d �(N 1s–O 1s)NO Ref.

Silica gel 103.4 [57,69]

Trans-[Ru(NO)Cl(cyclam)]2+ (RuNO) 399.4 532.8 281.6 131.9 a

400.9

SiO2 101.9 284.7 530.2 a

103.4 286.4 531.5532.7534.2

SiO2/SiNH2 101.6 283.3 398.9 530.5 a

103.4 284.2 400.8 531.9285.6 533.1287.5 534.5

SiO2/RuNO 101.4 284.0 397.6 530.0 282.1 a

103.4 285.2 399.2 531.5400.8 532.8

(39)534.0

SiO2/SiNH2/RuNO (2:1) 101.5 284.8 398.2 530.2 283.1 a

103.4 286.3 399.8 531.7402.0 533.1

534.8

SiO2/SiNH2/RuNO (1:1) 101.6 284.8 399.0 530.1 283.5 a

103.4 286.2 400.8 531.3532.6534.1

a This work.

[Ru(NO)Cl(cyclam)](PF6)2. This higher binding energy can beinterpreted as a result of a lower electronic density on the Ru–Clcomplex. This is in agreement with the oxidation state of theRu atoms in the different complexes. In Ru–Cl the Ru atom isformally +3, and in the trans-[Ru(NO)Cl(cyclam)](PF6)2 com-plex it is +2. It is also possible to see that the Cl 2p3/2 andN 1s (cyclam) peaks in the trans-[Ru(NO)Cl(cyclam)](PF6)2complex are at higher binding energies than those for the trans-[RuCl2(cyclam)]Cl complex. Therefore, it seems that the coor-dinated nitrosonium, NO+, is drifting electronic density fromthe chloro ligand in trans and from the nitrogen of the cy-clam. This explanation is in agreement with the previouslyreported decrease in the pKa of the coordinated water in trans-[Ru(NH3)4(NO)(H2O)]3+ compared to [Ru(NH3)5(H2O)]3+[23,26]. Indeed, the authors explained this decrease in pKa interms of electronic density drifting from water to NO throughthe H2O–Ru–NO+ axis [23,26]. So, the chloro ligand in trans-

[Ru(NO)Cl(cyclam)](PF6)2 is probably acting as a σ - and π -Lewis base toward Ru and, therefore, it transfers electronicdensity to the Ru–NO moiety.

The Ru 3d5/2 and N 1s XP-peaks of SiO2/RuNO are quitesimilar to those of the bulk trans-[Ru(NO)Cl(cyclam)]2+, asshown in Table 2. These findings agree with the FT-IR results.

The N 1s XP-peaks of SiO2/SiNH2 display an asymmetricpeak fitted with 2 Gaussian–Lorentzian components assigned toNH2 and NH+

3 species at 398.9 and 400.8 eV, respectively [71].The ratio between these species is ca. 3:1.

The Ru 3d5/2 XPS peaks obtained for the trans-[Ru(NO)-Cl(cyclam)](PF6)2 complex are at a lower energy than thoseobserved for all the encapsulated complexes (Table 2), sug-gesting an electron-withdrawing effect for the xerogel matrix.A greater difference is observed for the Ru 3d5/2 binding en-ergy of the complex encapsulated in the SiO2/SiNH2/RuNOmatrix. The binding energies for this complex lie at ca. 283.1

Autho

r's

pers

onal

co

py

550 K.Q. Ferreira et al. / Journal of Colloid and Interface Science 300 (2006) 543–552

and 283.5 eV for the xerogel made with 3-APTS and trans-[Ru(NO)Cl(cyclam)](PF6)2 complex with molar ratios of 2:1and 1:1. This higher shift seems to corroborate the UV–vis,FTIR, and 13C NMR results for these xerogels.

3.6. Electrochemistry

The immobilization of trans-[Ru(NO)Cl(cyclam)](PF6)2 inthe SiO2 matrix does not significantly affect the redox po-tentials, as in the case of trans-[Ru(NH3)4(imN)(SO2)] [42].In the potential range studied (−1.0–1.0 V), where the mate-rials are not electro-active, the cyclic voltammetry of trans-[Ru(NO)Cl(cyclam)](PF6)2 entrapped in SiO2 shows the occur-rence of two electrochemical processes (vs Ag/AgCl): Epc1 at−0.39 V, Epc2 at −0.30 V (E′

1/2 = −0.34 V; �E = 0.09 V;Ipa/Ipc = 0.8). The Epc1 value is close to that of the complexin solution (−0.33 V vs Ag/AgCl) [14], and is assigned like-wise to the reduction of the nitrosonium ligand (NO+/0) in theimmobilized complex.

3.7. Photochemical studies

In view of the previous results, we focused our attentionon SiO2/RuNO. Exposure of SiO2/RuNO, in CCl4, in deaer-ated conditions, to irradiation with light of 334 nm resultsin decrease in the intensity of the νNO band at 1870 cm−1,(Fig. 9). This decrease undoubtedly suggests the photochem-ical labilization of NO. Similar changes in the infrared spec-trum had already been observed with a 10−4 mol L−1 trans-[RuCl(NO)(cyclam)](PF6)2 solution at pH 7 (0.1 mol L−1,phosphate buffer) [29]. Furthermore, after photolysis, the colorof the solid material changed from yellow to green, with abroad absorption increase in the 300–340 nm region of theUV–vis spectrum, which is consistent with the formation ofa RuIII(cyclam) complex photoproduct in the intra-pores ofthe xerogel, thus yielding an NO depleted material, SiO2/Ru(Fig. 10), as observed for trans-[RuCl(NO)(cyclam)](PF6)2

Fig. 9. Vibrational spectra, in the mid-infrared region, of SiO2/RuNO underirradiation at 334 nm. Arrows indicate the decrease in the NO peak as a functionof the irradiation time. Elapsed time between each line is 60 min.

[41] and other [Ru(NH3)4(X)(NO)]n+ [8,20,72] and trans-[RuCl(NO)([15]ane)]2+ [73] complexes in solution, which ren-ders the respective Ru(III) aqua species and NO.

Despite the undoubted infrared results, confirmation of thephotolabilization of NO from the entrapped complex in thecase of SiO2/RuNO was achieved by a trapping technique us-ing a quartz cuvette topped with a glass reservoir. The xerogelwas placed in the cuvette and 3 mL of a 7.4 × 10−5 mol L−1

solution of the trapping agent, trans-[RuCl(cyclam)(OH2)]2+,was placed in the upper reservoir under argon atmosphere. Thetrans-[RuCl(cyclam)(OH2)]2+ complex reacts with NO (kon =0.2 M−1 s−1 at pH 1 and 25 ◦C) [29] to form the correspondingnitrosyl complex trans-[RuCl(NO)(cyclam)]2+. Upon photol-ysis of the solid material in the cuvette, the released NO wasthus trapped by trans-[RuCl(cyclam)(OH2)]2+ to give trans-[RuCl(NO)(cyclam)]2+ as evidenced by the new absorptionband at 266 nm, typical of the latter species (Fig. 11).

Fig. 10. UV–vis spectra of SiO2/RuNO (a) before irradiation at 334 nm and(b) after 60 min of irradiation.

Fig. 11. UV–vis spectra of the NO sequestering complex, trans-[RuCl(cyclam)-(OH2)]2+, during the NO trapping experiment, as a function of time of irradi-ation of SiO2/RuNO.

Autho

r's

pers

onal

co

py

K.Q. Ferreira et al. / Journal of Colloid and Interface Science 300 (2006) 543–552 551

Fig. 12. Infrared spectral changes during reaction between the photolysis prod-uct, at 334 nm, of trans-[Ru(NO)Cl(cyclam)](PF6)2 entrapped in SiO2 andNO(g). (a) Before reaction (in Web version green solid) and (b) after reaction(in Web version yellow solid).

The NO depleted xerogel (SiO2/Ru) could be reloaded to thenitrosyl form, SiO2/RuNO, by reaction with bubbling NO(g),at 25 ◦C, for 60 min in deaerated conditions. The regenerationof SiO2/RuNO was confirmed by observing the reappearanceof its typical yellow color and the νNO stretching band at1870 cm−1 in the FTIR spectrum (Fig. 12). Therefore, a load-ing–depleting–reloading cycle can be delineated as illustratedin Fig. 13.

4. Summary

Our results show that the controlled NO donor trans-[Ru(NO)Cl(cyclam)](PF6)2 can be entrapped in a SiO2 matrixwith preservation of its molecular structure and properties. Ina SiO2/SiNH2 matrix, the complex undergoes a nucleophilic

attack by the amine group at the nitrosonium, thus indicatingthat the use of amine functionalized silicas for metal nitrosylcomplexes should be avoided. Like the complex in solution, ir-radiation of the complex entrapped in a SiO2 matrix with lightof 334 nm results in the release of NO, as evidenced by changesin the UV–vis and IR spectra of the solid as well as by trappingthe released NO. This material was regenerated to the nitro-syl form by reaction with nitric oxide. Thus, this system hasthe potential to serve as a model to design regenerable pre-cursors for photochemical NO delivery to various biologicaltargets, and it also contributes to the design of materials coat-ings such as stents. Experiments concerning further materialscharacteristics as well as quantitative data concerning chemicaland photochemical release of NO and chemical regeneration ofthe material are under schedule in our lab.

Acknowledgments

The authors thank the Brazilian agencies FAPESP, CNPq,and CAPES for financial support, Prof. Thiery Gacoin forhelpful suggestions and Dr. Cynthia Maria de Campos PradoManso, for the English revision of the manuscript.

Supplementary material

The online version of this article contains additional supple-mentary material.

Please visit DOI: 10.1016/j.jcis.2006.03.081.

References

[1] M.J. Ignarro, Nitric Oxide: Biology and Pathobiology, Academic Press,New York, 2000.

[2] H. Kunkely, A. Vogler, Inorg. Chim. Acta 346 (2003) 275.[3] M.G. Sauaia, R.G. de Lima, A.C. Tedesco, R.S. da Silva, J. Am. Chem.

Soc. 125 (2003) 14718.

Fig. 13. Schematic representation of a reversible NO photolabilization and NO loading in SiO2/RuNO materials. The colors are representative of chromatic transi-tions between the NO loaded and the NO-depleted material.

Autho

r's

pers

onal

co

py

552 K.Q. Ferreira et al. / Journal of Colloid and Interface Science 300 (2006) 543–552

[4] Y. Chen, R.E. Shepherd, J. Inorg. Biochem. 68 (1997) 183.[5] A.S. Torsoni, B.F. de Barros, J.C. Toledo, M. Haun, M.H. Krieger, E.

Tfouni, D.W. Franco, Nitric Oxide Biol. Chem. 6 (2002) 247.[6] F.G. Marcondes, A.A. Ferro, A. Souza-Torsoni, M. Sumitani, M.J. Clarke,

D.W. Franco, E. Tfouni, M.H. Krieger, Life Sci. 70 (2002) 2735.[7] P.C. Ford, I.M. Lorkovic, Chem. Rev. 102 (2002) 993.[8] K. Szacilowski, W. Macyk, G. Stochel, Z. Stasicka, S. Sostero, O. Tra-

verso, Coord. Chem. Rev. 208 (2000) 297.[9] E. Tfouni, M. Krieger, B.R. McGarvey, D.W. Franco, Coord. Chem.

Rev. 236 (2003) 57.[10] A.K. Patra, M.J. Rose, K.A. Murphy, M.M. Olmstead, P.K. Mascharak,

Inorg. Chem. 43 (2004) 4487.[11] S. Wecksler, A. Mikhailovsky, P.C. Ford, J. Am. Chem. Soc. 126 (2004)

13566.[12] R.G. de Lima, M.G. Sauaia, D. Bonaventura, A.C. Tedesco, R.F.V. Lopez,

L.M. Bendhack, R.S. da Silva, Inorg. Chim. Acta 358 (2005) 2643.[13] P.G. Wang, M. Xian, X.P. Tang, X.J. Wu, Z. Wen, T.W. Cai, A.J. Janczuk,

Chem. Rev. 102 (2002) 1091.[14] D.R. Lang, J.A. Davis, L.G.F. Lopes, A.A. Ferro, L.C.G. Vasconcellos,

D.W. Franco, E. Tfouni, A. Wieraszko, M.J. Clarke, Inorg. Chem. 39(2000) 2294.

[15] S.P. Fricker, E. Slade, N.A. Powell, O.J. Vaughan, G.R. Henderson, B.A.Murrer, I.L. Megson, S.K. Bisland, F.W. Flitney, Br. J. Pharmacol. 122(1997) 1441.

[16] J. Bourassa, W. DeGraff, S. Kudo, D.A. Wink, J.B. Mitchell, P.C. Ford,J. Am. Chem. Soc. 119 (1997) 2853.

[17] M.G. Sauaia, F.D.S. Oliveira, R.G. de Lima, A.D.L. Cacciari, E. Tfouni,R.S. da Silva, Inorg. Chem. Commun. 8 (2005) 347.

[18] J. Bordini, D.L. Hughes, J.D.D. Neto, C.J. da Cunha, Inorg. Chem. 41(2002) 5410.

[19] C.F. Works, C.J. Jocher, G.D. Bart, X.H. Bu, P.C. Ford, Inorg. Chem. 41(2002) 3728.

[20] R.M. Carlos, A.A. Ferro, H.A.S. Silva, M.G. Gomes, S.S.S. Borges, P.C.Ford, E. Tfouni, D.W. Franco, Inorg. Chim. Acta 357 (2004) 1381.

[21] F.D. Oliveira, V. Togniolo, T.T. Pupo, A.C. Tedesco, R.S. da Silva, Inorg.Chem. Commun. 7 (2004) 160.

[22] J.C. Toledo, L.G.D. Lopes, A.A. Alves, L.P. da Silva, D.W. Franco,J. Inorg. Biochem. 89 (2002) 267.

[23] C.W.B. Bezerra, S.C. da Silva, M.T.P. Gambardella, R.H.A. Santos,L.M.A. Plicas, E. Tfouni, D.W. Franco, Inorg. Chem. 38 (1999) 5660.

[24] B.F. de Barros, J.C. Toledo, D.W. Franco, E. Tfouni, M.H. Krieger, NitricOxide Biol. Chem. 7 (2002) 50.

[25] J.A. Silva, J.C. Toledo, M.E.R. Guzzi, D.W. Franco, D. Wilhelm, R.C.Pedrosa, T.B. Creczynski-Pasa, Free Radical Biol. Med. 33 (2002) S244.

[26] B.R. McGarvey, A.A. Ferro, E. Tfouni, C.W.B. Bezerra, I. Bagatin, D.W.Franco, Inorg. Chem. 39 (2000) 3577.

[27] D. Bonaventura, F.S. Oliveira, R.S. da Silva, L.M. Bendhack, Clin. Exp.Pharmacol. Physiol. 32 (2005) 478.

[28] V. Mori, J.C. Toledo, H.A.S. Silva, D.W. Franco, M. Bertotti, J. Elec-troanal. Chem. 547 (2003) 9.

[29] K.Q. Ferreira, Ph.D. thesis, Universidade de São Paulo, 2004.[30] E. Tfouni, K.Q. Ferreira, F.G. Doro, R.S. da Silva, Z.N. da Rocha, Coord.

Chem. Rev. 249 (2005) 405.[31] G. Stochel, A. Wanat, E. Kulis, Z. Stasicka, Coord. Chem. Rev. 171 (1998)

203.[32] P.C. Ford, J. Bourassa, K. Miranda, B. Lee, I. Lorkovic, S. Boggs, S. Kudo,

L. Laverman, Coord. Chem. Rev. 171 (1998) 185.[33] L.K. Keefer, Annu. Rev. Pharmacol. Toxicol. 43 (2003) 585.[34] H.P. Zhang, G.M. Annich, J. Miskulin, K. Stankiewicz, K. Osterholzer,

S.I. Merz, R.H. Bartlett, M.E. Meyerhoff, J. Am. Chem. Soc. 125 (2003)5015.

[35] J.A. Hrabie, J.R. Klose, D.A. Wink, L.K. Keefer, J. Org. Chem. 58 (1993)1472.

[36] S.M. Marxer, A.R. Rothrock, B.J. Nablo, M.E. Robbins, M.H. Schoen-fisch, Chem. Mater. 15 (2003) 4193.

[37] J.T. Mitchell-Koch, T.M. Reed, A.S. Borovik, Angew. Chem. Int. Ed. 43(2004) 2806.

[38] H.E. Toma, A.M. Alexiou, A.L.B. Formiga, M. Nakamura, S. Dovi-dauskas, M.N. Eberlin, D.M. Tomazela, Inorg. Chim. Acta 358 (2005)2891.

[39] J.C. Toledo, H.A.S. Silva, M. Scarpellini, V. Mori, A.J. Camargo, M.Bertotti, D.W. Franco, Eur. J. Inorg. Chem. 9 (2004) 1879.

[40] J.M. Slocik, M.S. Ward, R.E. Shepherd, Inorg. Chim. Acta 317 (2001)290.

[41] C. Sanchez, B. Lebeau, F. Chaput, J.P. Boilot, Adv. Mater. 15 (2003) 1969.[42] S.M.C. Neiva, J.A.V. Santos, J.C. Moreira, Y. Gushikem, H. Vargas, D.W.

Franco, Langmuir 9 (1993) 2982.[43] J. Blum, A. Rosenfeld, N. Polak, O. Israelson, H. Schumann, D. Avnir,

J. Mol. Catal. A Chem. 107 (1996) 217.[44] Y. Kurusu, D.C. Neckers, J. Org. Chem. 56 (1991) 1981.[45] A. Rosenfeld, J. Blum, D. Avnir, J. Catal. 164 (1996) 363.[46] C. Sanchez, F. Ribot, B. Lebeau, J. Mater. Chem. 9 (1999) 35.[47] H. Schmidt, A. Kaiser, H. Patzelt, H. Scholze, J. Phys. 43 (1982) 275.[48] D. Bianchi, M. Lacroix, G.M. Pajonk, S.J. Teichner, J. Catal. 68 (1981)

411.[49] T. Lopez, P. Bosch, M. Asomoza, R. Gomez, J. Catal. 133 (1992) 247.[50] U. Schubert, New J. Chem. 18 (1994) 1049.[51] D. Avnir, Acc. Chem. Res. 28 (1995) 328.[52] H. Frenkel-Mullerad, D. Avnir, J. Am. Chem. Soc. 127 (2005) 8077.[53] J. Bordini, P.C. Ford, E. Tfouni, Chem. Commun. (2005) 4169.[54] A.I. Vogel, Química Orgânica, Ao Livro Técnico, Rio de Janeiro, 1980.[55] K. Kalcher, J.M. Kauffmann, J. Wang, I. Svancara, K. Vytras, C. Neuhold,

Z. Yang, Electroanalysis 7 (1995) 5.[56] Y. Gushikem, C.R.M. Peixoto, U.P. Rodrigues, L.T. Kubota, E. Stadler,

J. Colloid Interface Sci. 184 (1996) 236.[57] S.S. Chao, Y. Takagi, G. Lucovsky, P. Pai, R.C. Custer, J.E. Tyler, J.E.

Keem, Appl. Surf. Sci. 26 (1986) 575.[58] A. Proctor, D.M. Hercules, Appl. Spectrosc. 38 (1984) 505.[59] S. Sinha, A.K. Banerjee, B.K. Ghosh, Trans. Met. Chem. 23 (1998) 151.[60] F. Doctorovich, C. Trápani, Tetrahedron Lett. 40 (1999) 4635.[61] W.L. Bowden, W.F. Little, T.J. Meyer, J. Am. Chem. Soc. 99 (1977) 4340.[62] Z.J. Wu, H. Joo, I.S. Ahn, J.H. Kim, C.K. Kim, K. Lee, J. Non-Cryst.

Solids 342 (2004) 46.[63] M.S. Ahola, E.S. Sailynoja, M.H. Raitavuo, M.M. Vaahtio, J.I. Salonen,

A.U.O. Yli-Urpo, Biomaterials 22 (2001) 2163.[64] M. Zayat, R. Pardo, D. Levy, J. Mater. Chem. 13 (2003) 2899.[65] G. Engelhardt, E.D. Michel, High Resolution Solid-State NMR of Sili-

cates and Zeolites, Wiley, Norwich, 1987.[66] C. Rottman, G. Grader, D. Avnir, Chem. Mater. 13 (2001) 3631.[67] U.P. Rodrigues, Y. Gushikem, F.Y. Fujiwara, E. Stadler, V. Drago, Struct.

Chem. 5 (1994) 129.[68] H.O. Pastore, G.A. Ozin, A.J. Pöe, J. Am. Chem. Soc. 115 (1993) 1215.[69] D.T. Clark, I.S. Woolsey, S.D. Robinson, K.R. Laing, J.N. Wingfield, In-

org. Chem. 16 (1977) 1201.[70] V.I. Nefedov, Y.V. Salyn, A.P. Sadovsky, L. Beyer, J. Electron Spectrosc.

Relat. Phenom. 12 (1977) 121.[71] J.L. Magalhaes, L.M. Moreira, U.P. Rodrigues, M.J. Giz, M.A. Pereira da

Silva, R. Landers, R.C.G. Vinhas, P.A.P. Nascente, Surf. Interface Anal. 33(2002) 293.

[72] E. Tfouni, Coord. Chem. Rev. 196 (2000) 281.[73] D. Bonaventura, F.D. Oliveira, V. Togniolo, A.C. Tedesco, R.S. da Silva,

L.M. Bendhack, Nitric Oxide Biol. Chem. 10 (2004) 83.