Nitric Oxide as an Activation Agent for Nucleophilic Attack in trans-[ Ru

Neste trabalho foi investigado o ataque nucleofílico sofrido no estado sólido pelo complexo trans-[Ru(NO)(NH 3 ) 4 {P(OEt) 3 }](PF 6 ) 3 , no éster de fósforo coordenado, originando como produto a espécie trans-[Ru(NO)(NH 3 ) 4 {P(OH)(OEt) 2 }](PF 6 ) 3 . A reação foi monitorada e os produtos caracterizados utilizando ressonância magnética nuclear (P{H} CP-MAS NMR e P{H} NMR), espectroscopia no infravermelho (FTIR), ressonância paramagnética eletrônica (EPR), voltametria cíclica (CV), espectroscopia eletrônica (UV-Vis) e análise elementar. De acordo com os dados experimentais e cálculos de mecânica quântica (DFT), a reação ocorre no estado sólido pelo ataque nucleofílico no éster de fósforo coordenado devido à forte polarização no eixo P–Ru–NO induzida pela presença do ligante nitrosilo. A reação segue o mecanismo tipo Michaelis-Arbusov para hidrólise de ésteres de fósforo. Em solução (pH 7,0), o ataque nucleofílico ocorre nos ligantes NO e P(OEt) 3 , gerando como produtos os complexos trans-[Ru(H 2 O)(NH 3 ) 4 {P(OEt) 3 }] e trans-[Ru(NO)(H 2 O)(NH 3 ) 4 ] em proporções comparáveis.


Introduction
2][3] In the absence of oxygen, i.e., hypoxia situation of living organisms, the NOS enzymes are inhibited. 1Ruthenium nitrosyl complexes of the type trans-[Ru(NO)(NH 3 ) 4 (L)] 3+ (L = N-heterocyclic and triethylphosphite P(OEt) 3 ) are very robust in aqueous solutions, but when activated by one electron reduction, they Vol.21, No. 7, 2010   release NO 0 at rates that may be controlled depending on the nature of L. 4,5 Therefore, these compounds would provide NO 0 under circumstances where NOS are not operational.
A systematic approach aiming to use ruthenium nitrosyl complexes as NO carriers is being developed. 6,7For this, trans-[Ru(NO)(NH 3 ) 4 {P(OEt) 3 }](PF 6 ) 3 is found to be interesting since k -NO is 0.98 s -1 at 25 ºC, the compound is water soluble, its metabolites have very low cytotoxicities, and the E 0 NO+/NO0 = −0.18 vs. SCE is accessible to biological reductors such as cysteine, gluthatione and ascorbic acid. 8urthermore, this compound exhibits a quite promising behaviour in biological experiments as tripanomicyde and leishmanicyde agent. 9hosphorus esters are known to be quite reactive. 10In acidic solutions, triethylphosphite, P(OEt) 3 , undergoes acid catalyzed hydrolysis yielding diethylphosphite (P(OH)(OEt) 2 ) and diethylphosphonate (P(H)(O)(OEt) 2 ) by the Michaelis-Arbusov reaction mechanism. 11Furthermore, in the presence of oxidants such as hydrogen peroxide or bromine, the oxidation of P III to P V is observed to take place easily.The formation of the stable P=O bond is considered to be the driving force for this reaction. 12However, when coordinated to Ru II , P(OEt) 3 is stable with respect to oxidation.For example, trans-[Ru(NH 3 ) 4 {(P(OEt) 3 } 2 ] 2+ is stable in solid state for long periods of storage in a vacuum desiccator and in the absence of light.Also, the complexes trans-[Ru(H 2 O)(NH 3 ) 4 {P(OEt) 3 }] 2+ and trans-[Ru(NH 3 ) 4 {(P(OEt) 3 } 2 ] 2+ are stable in acid aqueous solution even on addition of excess Br 2 or Ce IV , leading only to the metal center oxidation. 8itrogen monoxide binds strongly to the Ru II center and, in this case, the back-bonding in the [Ru II NO + ] 3+ fragment is so strong that the metal center exhibits properties of Ru III . 13,14The Ru II center in trans-[Ru(NO)(H 2 O)(NH 3 ) 4 ] 3+ is not oxidized by Ce IV or Br 2 even in the presence of an excess of these oxidants. 15For example, the pK a value of the coordinated water molecule in trans-[Ru(NO)(H 2 O) (NH 3 ) 4 ] 3+ is 3.1, thus being a stronger acid than trans-[Ru(H 2 O)(NH 3 ) 5 ] 2+ , pK a 4.5, by one order of magnitude. 15hus, the presence of the nitrosyl ligand in trans-[Ru(NO) (NH 3 ) 4 {P(OEt) 3 }](PF 6 ) 3 induces not only changes in the metal center but also promotes the activation of the trans positioned phosphorus ester for nucleophilic attack.

Results and Discussion
Samples of trans-[Ru(NO)(NH 3 ) 4 {P(OEt) 3 }](PF 6 ) 3 have shown to undergo gradual degradation in the solid state, even when the complex was stored in a vacuum desiccator under controlled air moisture.Figure 1 illustrates changes in the vibrational spectrum of the complex in solid samples kept for 8 months in such conditions (p = 0.035 bar, air moisture 20 ± 3%) and protected from light at room temperature.The dashed line is the spectrum of freshly prepared trans-[Ru(NO)(NH 3 ) 4 {P(OEt) 3 }](PF 6 ) 3 , which exhibits the characteristic band for coordinated nitrosyl (ν(NO + ) = 1915 cm -1 ). 5 The band assigned to ν(NO + ) was found to shift to a lower wavenumber (1885 cm -1 ), indicating changes in the phosphorus ester ligand as a function of time.Concomitantly, changes in the intensity of the band relative to d(CH 3 ) at 1390 cm -1 (Figure 1, a) were also observed in the aged sample.The area calculated for this band in the degradation product is 35% smaller than that determined for the starting material.This suggests the loss of an ethyl group (C 2 H 5 ) from the original complex.Two new bands assigned as ν(P−O−H) 16 are observed to appear at 1140 and 940 cm -1 (Figure 1, b and c), corroborating the nucleophilic attack on the P(OEt) 3 ligand.The spectral characteristics suggest that the phosphorus ligand stays coordinated after the hydrolysis process.The NO + ligand was not found to suffer any reaction during the period of time under analysis, as indicated by the FTIR spectra.
To estimate the amount of trans-[Ru(NO) (NH 3 ) 4 {P(OEt) 3 }](PF 6 ) 3 that suffered nucleophilic attack, deconvolution analyses of the bands relative to ν(NO + ) at 1915 and 1885 cm -1 were carried out.Results suggest that, after 8 months of storage, approximately 95% of the original complex was converted into the degradation product.Furthermore, evidence of participation of air moisture in the degradation process was observed.Samples of trans-[Ru(NO)(NH 3 ) 4 {P(OEt) 3 }](PF 6 ) 3 were also kept under argon in ampoules, where the degradation process was found to J. Braz.Chem.Soc.1268 take place at the beginning and then ceased after 3 months, suggesting that the process is air-moisture dependent.FTIR experiments carried out periodically showed that 20% of the initial amount of trans-[Ru(NO)(NH 3 ) 4 {P(OEt) 3 }](PF 6 ) 3 was converted into the degradation product.For these samples, no additional changes were observed in the FTIR spectra during a period of one year.Also, samples of trans-[Ru(NO) (NH 3 ) 4 {P(OEt) 3 }](PF 6 ) 3 stored in Schlenk tubes under dry argon did not undergo any decomposition after three months.
Elemental analysis of the degradated samples of trans-[Ru(NO)(NH 3 ) 4 {P(OEt) 3 }](PF 6 ) 3 showed a loss of 35% in the carbon content, but no changes in the nitrogen content.This is consistent with the loss of one ethyl group (C 2 H 5 ), suggesting strongly that the P(OEt) 3 ligand suffered nucleophilic attack.
Figure 2 shows the UV-Vis spectrum of the degradation product (solid line) obtained after 8 months of storage.The figure clearly depicts bands detected at 242 nm (ε = 1,429 L mol -1 cm -1 ), 312 nm (ε = 912 L mol -1 cm -1 ) and 474 nm (ε = 38 L mol -1 cm -1 ), indicating that the decomposition product has NO + and a phosphorus ester in the coordination sphere.These absorption coefficient values were calculated considering 95% of conversion of the original complex into the new one.The electronic spectrum of freshly prepared trans-[Ru(NO)(NH 3 ) 4 {P(OEt) 3 }](PF 6 ) 3 (dashed line) has three bands at 260 nm (ε = 1,291 L mol -1 cm -1 ), 316 nm (ε = 258 L mol -1 cm -1 ) and 522 nm (ε = 35 L mol -1 cm -1 ) attributed to ligand field (260 and 316 nm) and charge transfer (522 nm) transitions. 8This spectrum and band assignments are consistent with the described for other ruthenium nitrosyl tetraammines. 17,18No shift in the bands at 242 and 312 nm in the electronic spectra of the degradation product was observed in acetonitrile, suggesting the predominant ligand field character of the these bands.Due to the microsymmetry similarity exhibited by these two compounds, the electronic absorption bands of the product of nucleophilic attack were tentatively attributed to the same transitions assigned to trans-[Ru(NO)(NH 3 ) 4 {P(OEt) 3 }](PF 6 ) 3 . 8he 31 P{ 1 H} CP-MAS NMR analysis of the decomposition product was also carried out, confirming changes in the coordination sphere of trans-[Ru(NO) (NH 3 ) 4 {P(OEt) 3 }](PF 6 ) 3 , as suggested by FTIR and UV-Vis data.After the storage period of 8 months, more than 95% of the resonance peak recorded at 80 ppm for trans-[Ru(NO)(NH 3 ) 4 {P(OEt) 3 }](PF 6 ) 3 (Figure 3, top) was observed to disappear, being replaced with a new one at 67 ppm (Figure 3, bottom).According to the literature, 8,19 this chemical shift region for 31 P can be attributed to coordinated P III , and this indicates that the phosphorus ester remains coordinated after the degradation process.The change in the chemical shift by 13 ppm suggests the formation of a new P III species coordinated to the Ru II center, but with a weaker π acidity character than that presented by the P(OEt) 3 ligand.This would explain the observed displacement of the peak to a lower chemical shift value as a consequence of the small back-donation between the Ru II 4d π orbitals and P III 3d π orbitals.The splitting of the 31 P signals in both spectra may be due to polymorphism in the solid state samples. 20he 31 P{ 1 H} NMR spectrum of the degradation product in acid aqueous solution (C H+ = 1.0×10 -1 mol L -1 ) shows only one peak at 67 ppm, which is the same chemical shift value found in the 31 P{ 1 H} CP-NMR spectrum.No additional peaks were observed in the NMR spectra for a period of one hour, suggesting that the new complex is stable in this experimental condition.(E 0 1/2 a1 = 0.315V).Furthermore, as described previously 5,8 only one reduction wave attributable for the NO + /NO 0 process is present in the voltammogram of trans-[Ru(NO) (NH 3 ) 4 {P(OEt) 3 }] 3+ at −0.18V (Figure 4, b2), whereas for the degradation product it is possible to observe two waves: one at −0.36V (b1) and another at −0.56V (c) (Figure 4).From the stand point of the observed peak current intensity, the first wave may be assigned to the reduction process NO + /NO 0 (a1; E 0 = −0.36V).The shift of the half-wave potential of the NO + /NO 0 couple to more negative values is coherent with the reduced π acidity of the phosphorus ester ligand, corroborating the 31 P{ 1 H} CP-NMR and FTIR spectral data.
Until now, we do not have a good explanation for the second peak at −0.56V.A more detailed electrochemical investigation is being carried out.
All samples analysed for 8 months were EPR silent, suggesting the presence of only Ru II after the degradation process.
Taking in account all data obtained so far, it is likely that the nucleophilic attack on the coordinated phosphorus ester leads, in the solid state, to trans-[Ru(NO)(NH -in acidic media have been unsuccessful so far. 22On the other hand, a recent report describes the synthesis and reactions of the complex [Mo(CO) 5 {P(OH)(OEt) 2 }], in which the metal center is also in a d 6 low spin configuration.In solution, this complex undergoes nucleophilic attack on the diethylphosphite ligand, yielding [Mo(CO) 5 {P(OH) 3 }]. 23,24xperiments in aqueous media confirm that trans-[Ru(NO)(NH 3 ) 4 {P(OEt) 3    These products were identified through differential pulse voltammetry (DPV) studies (Figure 5), in which the presence of trans-[Ru(H 2 O)(NH 3 ) 4 {P(OEt) 3 }] 2+ and trans-[Ru(NO)(H 2 O)(NH 3 ) 4 ] 3+ was observed at +0.51 (peak c) and −0.48 V (peak b) vs. SCE, respectively.Consistently, the FTIR spectrum 25  The reaction of phosphorus esters in acid media is well known and recognized to take place following the Michaelis-Arbusov type mechanism, in which the nucleophilic attack occurs at the carbon atom of the P(OEt) 3 ligand.The P(OEt) 3 coordinated to the ruthenium in trans-[Ru(NO)(NH 3 ) 4 {P(OEt) 3 }] 3+ can be activated  for nucleophilic attack, with the [Ru(NO)(NH 3 ) 4 ] 3+ moiety playing the proton role in the Michaelis-Arbusov mechanism. 25In the solid state, water molecules are likely to be the nucleophilic agent, but in solution the behavior is quite different.Only solutions of the ion complex trans-[Ru(NO)(NH 3 ) 4 {P(OEt) 3 }] 3+ with pH > 6.0 suffered nucleophilic attack, suggesting that in solution OH -is the nucleophilic agent.In this perspective, it is likely that the reactions take place in the solid state and in solution following two different pathways (Scheme 1).

trans-[Ru(NO)(NH
When coordinated to ruthenium in the [Ru(NH 3 ) 5 ] 2+ or [Ru(H 2 O)(NH 3 ) 4 ] 2+ complexes, P(OEt) 3 is remarkably stabilized towards hydrolysis and thus stable over a wide pH range. 5,8This could be a consequence of an electron density enrichment on the phosphorus and carbon atoms by the strong back-bonding from the filled Ru II 4d π orbitals to the empty P III 3d π ligand orbitals. 5,8,13,21Under the influence of NO + , the back-bonding is significantly attenuated due to the competition between P(OEt) 3 and NO + for the Ru II 4d π electrons.The nearly linear Ru II −NO + angle (175º.10') in this complex 5,8 also offers a favorable symmetry for the overlapping between the empty p yπ* and p zπ* orbitals (in the NO + ligand) and the electron rich Ru d xz and d yz orbitals. 16he changes in the NO + ligand as a consequence of the trans influence of the phosphorus ligand can be envisaged on the plot of E 0 (NO + /NO 0 ) (SCE) vs. ν(NO + ) (cm 1 ).As can be observed, the trans-[Ru(NO)(NH 3 ) 4 {P(OH)(OEt) 2 }] 3+ ion follows the linear relationship (r 2 = 0.94) observed 5,13 for other tetraamines (Figure 6).
The electronic density of the phosphorus and carbon atoms of the phosphite ligand is also expected to decrease in the nitrosyl complex.According to DFT calculations performed for the trans-[Ru(NO)(NH 3 ) 4 {P(OEt) 3 }] 3+ and trans-[Ru(NO)(NH 3 ) 4 {P(OH)(OEt) 2 }] 3+ ions, the composition of the LUMO orbitals is predominantly characteristic of the NO + and P(OEt) 3 ligands.From the calculated NBO charges for trans-[Ru(NO) (NH 3 ) 4 {P(OEt) 3 }] 3+ and for the free P(OEt) 3 (Table 1), it is possible to infer that, upon coordination to the metal center, remarkable changes occur in the P(OEt) 3 ligand, which becomes more electrophilic.According to the NBO charge values, the more positive is the NBO charge value, more susceptible is the atom to nucleophilic attack. 26rom the NBO data (Table 1) it is possible to infer that four atoms, i.e., the metal center, the nitrogen of the nitrosyl ligand and the phosphorus and carbon atoms of the CH 2 groups, are more susceptible to the nucleophilic attack.According to X-ray diffraction data 27 and quantum mechanical calculations, the metal center and phosphorus atoms, but not the carbon atoms, are expected to exhibit steric hindrance, therefore making the reaction difficult.
The solid state 13 C NMR spectra of the nitrosyl complexes containing P(OEt) 3 and P(OH)(OEt) 2 exhibited chemical shifts of 63 and 70 ppm for the CH 2 groups respectively, thus suggesting that the CH 2 group in the P(OEt) 3 complex is more eletrophilic than in the P(OH)(OEt) 2 species, being consequently more susceptible to nucleophilic attack.Corroborating the NBO data on Table 1, the 13 C chemical shifts determined for the CH 3 groups of the diethyl/triethylphosphite ligands did not show any expressive difference (d(CH 3 ) 16.0 and 16.7 ppm respectively).On this perspective, the DFT calculations, together with the 13 C NMR data, strongly suggest a nucleophilic attack on the CH 2 group of P(OEt) 3 , following the Michaelis-Arbusov type mechanism of phosphorous ester hydrolysis and leading to P(OH)(OEt) 2 .

Experimental
All reagents were used as purchased from Aldrich and Merck.P(OEt) 3 was purified as described before. 8Solvents, purchased from Mallinckrodt Baker and Merck, were purified following literature procedures. 29Five samples of the trans-[Ru(NO)(NH 3 ) 4 {P(OEt) 3 }](PF 6 ) 3 complex were periodically analysed by spectroscopic techniques during a period of 8 months.Samples were stored in a vacuum desiccator (p = 0.035 bar), in which air moisture was controlled with a drying agent (CaCl 2 •H 2 O, air moisture 20 ± 3%). 29Air moisture was continuosly measured using a hygrometer (Barigo, Vilingen-Schwenningen, Germany) and the samples homogenized every 15 days to avoid only superficial degradation.All manipulations of air sensitive compounds were carried out as described in literature, 30 using serum caps, teflon tubes and glass  syringes.The complex trans-[Ru(NO)(NH 3 ) 4 {P(OEt) 3 }] (PF 6 ) 3 was synthesized as described in the literature. 5UV-Vis measurements were performed on a Hitachi U-3501 spectrophotometer using a 1.0 cm path length quartz cell.Degassed solutions of the degradation product were transferred into quartz cells using teflon tubes.The spectra were recorded from 225 to 600 nm.FTIR spectra were recorded using a Bomem MB-102 series spectrometer, using KBr pellets, in the region of 2000-400 cm -1 .Elemental analysis was performed on an EA 1110 CHNS-O Carlo Erba Instrument. 31P{ 1 H} CP-MAS NMR measurements were carried out with a Varian Inova 400 MHz spectrometer using fine powdered samples packed into a zirconia CP rotor.The counter ion PF 6 − (from NH 4 (PF 6 ), d 31 P = −144 ppm) was used as reference.EPR spectra were recorded on a Bruker ESP 300E X-band spectrometer at 77K.DPPH• (g = 2.0037) was used as the field calibrant.Electrochemical measurements (CV and DPV) were performed on an EG&G PAR model 264A equipment, using a glass carbon electrode as a working electrode, a saturated calomel electrode (SCE) as reference and a platinum plate as auxiliary electrode.Solutions of CF 3 COOH/CF 3 COONa (µ = 0.1 mol L -1 ) were used as electrolyte.
DFT calculations for trans-[Ru(NO)(NH 3 ) 4 {P(OEt) 3 }] 3+ and trans-[Ru(NO)(NH 3 ) 4 {P(OH)(OEt) 2 }] 3+ were performed with the Gaussian 03 package. 31Optimized molecular geometries were calculated using the hybrid B3LYP exchange-correlation functional with the GDIIS algorithm, 32,33 and tight self-consistent field convergence criteria were used for calculations.Solvent effects were evaluated using the Polarizable Continuum Model (PCM). 34Single-point calculations with the B3LYP and the standard 6-311+G* basis sets for H, C, N, P and O, and the LANL2DZ basis set for Ru were performed.A dielectric constant of ε = 78.39 was used for the complexes in water.Natural atomic charges 35 from natural population analysis (NPA) were obtained with the NBO 3.0 program as implemented in the Gaussian 03 software.All the fully optimized geometries were characterized by vibrational frequency calculations, which showed only real frequencies.Zero-point vibrational energies were estimated based on the B3LYP frequency calculation (unscaled) using the same basis set as for geometry optimization.

Conclusions
The data presented here strongly suggest that, in solid state, a nucleophilic attack at the CH 2 group of P(OEt) 3 coordinated to the Ru II center in trans-[Ru(NO) (NH 3 ) 4 {P(OEt) 3 }](PF 6 ) 3 occurs, leading to the formation of trans-[Ru(NO)(NH 3 ) 4 {P(OH)(OEt) 2 }](PF 6 ) 3 .After 8 months of storage, the conversion was found to be higher than 95% and the diethylphosphite complex was stable for a period of at least one year.The degradation process was followed with five independent samples and shown to be reproducible.The nucleophilic attack follows the Michaelis-Arbusov type-reaction mechanism.In solution, the nucleophilic attack occurs on the NO + and P(OEt) 3 ligands competitively, yielding the trans-[Ru(H 2 O)(NH 3 ) 4 {P(OEt) 3 }] 2+ and trans-[Ru(NO)(H 2 O)(NH 3 ) 4 ] 3+ complex ions.The presence of the nitrosyl on the Ru coordination sphere appears to be the driving force for the nucleophilic attack, since no reaction was observed in the same experimetal conditions for the synthetic precursors of the nitrosyl complex.The strong π electron polarization generated by NO + may be useful as catalysis strategy to activate molecules for nucleophilic attack in biological pathways where NO is involved.