Study of Interaction between NO Radicals and Martin's Spirosilane by

Article pubs.acs.org/JPCA

Study of Interaction between NO Radicals and Martin’s Spirosilane by Means of IR Spectroscopy E. L. Zins,*,† L. Krim,† H. Lenormand,‡ J.-P. Goddard,‡ and L. Fensterbank‡ †

Laboratoire de Dynamique, Interactions et Réactivité (LADIR, UMR CNRS 7075), Université Pierre et Marie Curie Case 49, 4 place Jussieu, 75252 Paris cedex 05, France ‡ Institut Parisien de Chimie Moléculaire (IPCM, UMR CNRS 7201) Fédération de Chimie Moléculaire (FR 2769), Université Pierre et Marie Curie Case 229, 4 place Jussieu, 75252 Paris cedex 05, France ABSTRACT: The matrix isolation method is used to record the IR spectrum of C18H8O2F12Si in the 4000−500 cm−1 range. To gain an IR spectrum with a sufficient resolution, this technique was used with neon as the dilution medium at 5 K. The generated species were characterized by in situ fourier transform infrared (FT-IR) spectroscopy. Once the Martin’s spirosilane 1 (C18H8O2F12Si) was characterized, its reactivity toward NO was investigated under the same experimental conditions (i.e., using neon as a dilution medium at 5 K). In this case, the use of neon at very low temperature leads to the formation of a chemically inert matrix in which the species are trapped and isolated from one another, thus hindering consecutive reactions. As a consequence, intermediates can be observed. This approach allowed us to characterize the NO adduct, leading to the formation of 1-(NO). Concentration effects as well as annealing experiments were carried out. In addition to this experimental approach, products were identified by using reference spectra. Our results proved that, in the dilute phase, the reaction between 1 and NO radicals leads to the formation of an adduct. This stable species can further react with NO to form a more stable compound: 1-(NO)2. This proves the ability of such species to trap NO.

■

INTRODUCTION

NO radicals play a vital role in atmospheric chemistry and in biochemistry. Indeed, they are present in the upper layer of the Earth’s atmosphere1 and they play a crucial biological role.2,3 As a consequence, its characterization in different environments is needed. To this end, one approach is to find molecules that will react with NO radicals to form a stable complex. To some extent, spirosilanes are good candidates to form such unpaired electron species. Indeed, it is known that some silanes can form free radicals with five- or six-coordinated silicon either with equivalent or nonequivalent ligands.4−9 Since organo-silicon compounds can form radicals and biradicals with six coordinate silicon atoms, tetravalent organo-silicon compounds may react with either one or two NO radicals. The Martin’s spirosilane 1 was chosen to investigate such reactivity (Figure 1).10−12 The products formed were isolated and characterized thanks to the matrix isolation infrared absorption spectroscopy. Matrix isolation infrared absorption spectroscopy is a powerful method to isolate and characterize species.13,14 In such experiments, gaseous species are introduced simultaneously with a chemically inert gas, and this mixture is condensed on a cold surface. Infrared spectroscopy is further used to characterize the sample thus generated.15 This approach limits the formation of aggregates and hydrated © 2013 American Chemical Society

Figure 1. Structure of the 3,3,3′,3′-tetrakis(trifluoromethyl)1,1′(3H,3′H)-spirobi-[2,1-benzoxasilole] C18H8O2F12Si (CAS registry number 70091-69-9) spirosilane 1 studied.

species, and as a consequence, the absorption bands observed are often narrow and easily assigned. In the context of reactivity studies, the matrix isolation technique provides a unique way to study reactions step by step and to isolate intermediates. Indeed, the reagents are isolated from one another in the inert matrix, thus hindering consecutive reactions. Additionally, in such experiments, the temperature is strictly controlled and kept low, and annealing experiments can be carried out, providing valuable information on the relative stability of products. Received: July 11, 2012 Revised: March 27, 2013 Published: March 27, 2013 3296

dx.doi.org/10.1021/jp401177c | J. Phys. Chem. A 2013, 117, 3296−3303

The Journal of Physical Chemistry A

Article

−78 °C to freshly distilled SiCl4 (10.21 g, 60.08 mmol, 0.59 equiv.). The reaction was stirred at 25 °C for 12 h, then quenched with 7 mL of water, dissolved in diethyl ether (100 mL), and washed with aqueous 0.5 M HCl (4 × 100 mL) and water (100 mL). The organic phase was dried over MgSO4, filtered, and evaporated under reduced pressure to afford an orange oil. The residue was crystallized in hexane to give 13.77 g of a beige solid and then a recrystallization of the crude product with hexane afforded white solid 1 (10.45 g, 20.40 mmol, 41%). mp: 132 °C. 1H NMR (400 MHz, acetone-d6) δ 7.98−7.86 (m, 6H, arom.), 7.80 (m, 2H, arom.). 13C NMR (100 MHz, acetone-d6): δ = 142.0 (Cquat, arom.), 135.3 (CH, arom.), 134.2 (CH, arom.), 132.5 (CH, arom.), 130.0 (Cquat, arom.), 126.0 (m, CH, arom.), 124.9 (quad, J = 284.6 Hz, Cquat, CF3), 83.2 (m, Cquat, C−CF3). 19F NMR (377 MHz, acetoned6): −76.5 (s, 12F, CF3). The spectroscopic data were consistent with those reported in the literature. 3. Experimental Setup. a. Ne Matrix Experiments. The experimental method and setup have been previously described in detail (Zins et al.).29 Nevertheless, we shall recall the main points and the specific experimental conditions used for the present study. In a first series of experiments, pure neon was introduced in a flask containing 1 and heated up to 40 °C, in order to increase the vapor pressure of the reagent. As a consequence, the gas mixture escaping this flask is a Ne/1 mixture. This mixture was further condensed on the cryogenic metal mirror maintained at 5 K during 30 min. Since Ne is the major component present in the gas mixture reaching the mirror, the sample obtained is a neon matrix in which spirosilane molecules are trapped. In a second series of experiments, the Ne gas was replaced by a NO/Ne 1:100 mixture. For concentration effect studies, experiments were also carried out with NO/Ne 0.5:100 and 2:100 mixtures. Six mirrors are stuck on the head of a cryostat (Cryomech PT 405) maintained at 5 K. Each of these mirrors can be used to study a sample. Indeed, this part of the device is movable, so that each of the mirrors can be exposed to the gas coming from the flask, and then experimentally probed by IR spectroscopy (Figure 2). As a consequence, it is possible to carry out simultaneously six experiments with six samples characterized by different concentrations.

Previous reactivity studies between atomic oxygen and either SiH4, Cl3SiOH, or Me2Si(OH)2 proved that matrix isolation experiments can lead to the synthesis of molecules containing SiOH and SiO functional groups .16−20 The formation of methyl- and dimethylsilanone from methyl-substituted silane and oxygen atoms was also spectroscopically identified.21−23 In order to attribute the main signals observed, theoretical calculations were carried out. To this end, the density functional theory was applied to fully optimize the structure of the reagent. Once the most stable structure was obtained, frequency calculations were carried out, in order to calculate the IR spectrum, and to check that the geometry obtained is indeed compatible with a minimum on the potential energy surface. A temperature as low as 5 K was preferred, in order to characterize the reactions step by step and to avoid side reactions. Moreover, the low temperature facilitated the minimization the mobility of cold NO radicals and further avoids the secondary reactions. Neon was chosen as a dilution medium for three reasons, (1) it allows the study of the reactivity of the reagents in their fundamental states, (2) it allows the formation of a rare gas matrix thus isolating the products from one others, and (3) compared with other rare gases, the matrix formed from Ne slightly interacts with the molecules. As a consequence, the IR spectrum obtained is closer to the one corresponding to the gas phase.

■

EXPERIMENTAL TECHNIQUE 1. Theoretical Calculations. In order to assign the main signals observed on the IR spectra for the isolated compound, theoretical calculations based on the density functional theory were carried out. To this end, different structures were considered and fully optimized by analytic gradient methods. All calculations were performed using Gaussian 03 software.24A hybrid method based on the Becke (B) exchange functional along with the Lee−Yang−Parr (LYP) nonlocal correlation functional was used for this study (B3LYP),25,26 in combination with the 6-31+G* basis. The choice of the functional was based on previous studies.27,28 Different structures were considered, built, and fully optimized. To begin with, fragments of the molecule were built up and fully optimized. The structures thus obtained were further used to build up larger fragments up to the whole molecule. Frequency calculations were carried out at the same level of theory, to take into account the zero-point energy correction (ZPE) and also to generate the theoretical IR spectrum. 2. Chemicals. Ne and NO were obtained from L’Air liquide with purities of 99.9995% and 99.9%, respectively. Impurities due to the decomposition of NO were reduced by reaction with mercury. NO gas was further purified by using trap-to-trap vacuum distillations. The purity of samples was confirmed spectroscopically. 3,3,3′,3′-Tetrakis(trifluoromethyl)- 1,1′(3H,3′H)-spirobi[2,1-benzoxasilole]- 1 (CAS registry number 70091-69-9) was prepared as follows.11 In a 500 mL flask, TMEDA (4.24 g, 19.59 mmol, 0.20 equiv.) was added to a stirred solution of nBuLi (120.0 mL, 1.92 M hexane solution, 230.42 mmol, 2.25 equiv.). This mixture was stirred at 25 °C for 15−20 min until it became cloudy and cooled to 0 °C. 1,1,1,3,3,3-Hexafluoro-2-phenylpropan-2-ol (24.41 g, 100.00 mmol, 1.00 equiv.) was dissolved in 30 mL of THF and then added dropwise to the TMEDA/nBuLi mixture. After 30 min, the ice bath was removed and the mixture was stirred for 20 h. This mixture was then added over 45 min at

Figure 2. Scheme of the experimental setup used for the present study.

The samples thus formed were further annealed up to 13 K. To this end, the sample was heated step by step from 5 to 13 K, with increments of 1 K during 1 min and then cooled to 5 K. Thus, all of the spectra were registered at the same temperature, which allows us to compare these spectra and to follow the relative intensities of the absorption bands of the species in the matrix. The pressure in the cryostat cell is always in the 10−7 mbar range except during the gas injection, i.e., during the synthesis 3297



Article

Table 1. Theoretical and Experimental IR Characterization of the Reagent frequency in Ne matrix

frequency in solid KBr

theoretical value B3LYP/6-31+G*

1464.9 1449.8 1288.4 1251.8 1236.1, 1222.2

1461

1486

rocking of the benzenic H atoms

1291

1289

symmetric stretching C−C−C

1204 1205 1206 1207 1177 1171 1170 1169 1133 1114 975 899 830 764 757 751 703 700

wagging C−C−C (F3C−C−CF3) wagging C−C−C (F3C−C−CF3) scissoring of the benzenic hydrogen scissoring of the benzenic hydrogen stretching of the benzenic C−C twisting C−C−C (F3C−C−CF3) twisting C−C−C (F3C−C−CF3) stretching of the benzenic C−C symmetric stretching Si−O antisymmetric stretching C−O antisymmetric stretching C−C antisymmetric stretching Si−O symmetric stretching Si−O antisymmetric stretching Si−C twisting of the benzenic C−C twisting of the benzenic C−C C−F vibrations C−F vibrations

1203

1175.5, 1185 1168

1112.7, 1104.4 985.4, 979.2 924.7

714.3

1112 981 921 847

748 705

attribution

Figure 3. Spectrum of the spirosilane 1 (a) in Ne matrix and (b) in solid phase.

The experimental method for the infrared setup is detailed here. Infrared spectra of the samples were recorded in the transmission-reflection mode between 4000 and 500 cm−1 using a Bruker 120 HR fourier transform infrared (FTIR) spectrometer equipped with a KBr/Ge beam splitter and liquid N2-cooled narrow band HgCdTe photoconductor. A resolution of 0.5 cm−1 was used. Bare mirror backgrounds recorded from 4000 to 500 cm−1 prior to sample deposition were used as references in processing the sample spectra. Absorption spectra in the mid-infrared region were collected on samples through a KBr window mounted on a flange separating the interferometer vacuum (10−5 mbar) from the cryostat cell. The spectra were subsequently subjected to baseline correction to compensate for infrared light scattering and interference patterns.

of the sample. During the deposition, the pressure increases up to 10−5 mbar. To this end, two secondary pumps are used in the device: The chamber containing the cryostat is pumped away by a water-cooled inlet baffle pumping stack Edwards, Standart Diffstack 160/700P. The pumping speed of this pump is 700 L s−1 for N2. The gas ramp used to prepare the injected mixtures is connected to a water-cooled inlet baffle pumping stack Edwards, Standart Diffstack 63/150M. The pumping speed of this pump is 135 L s−1 for N2. In addition with these secondary pumps, two Edwards RV 8 two stage rotary vane pumps, A65401903, are used to achieve a primary vacuum. 3298



Article

Figure 4. Structures of the species considered for the theoretical investigation: model molecules: (a) C6H8O2Si five-membered heterocyclic ring, (b) C10H4O2F12Si, and (c) 1,1,1,3,3,3-hexafluoro-2-phenyl-2-propanol.

In all of the figures presenting IR spectra, the different traces were vertically shifted, but neither the scales nor the abscissa were changed. A sketch of the apparatus used for all of the experiments reported here is presented in Figure 1. b. Solid-Phase Experiments. In order to complement the characterization of the reagent, the Bruker 120 HR FTIR spectrometer was also used to register the spectrum of the sample in the solid phase between two KBr windows. These spectra were registered at atmospheric pressure in the transmission-reflection mode between 4000 and 500 cm−1. The resolution was set at 2.0 cm−1. Bare KBr windows backgrounds recorded from 4000 to 500 cm−1 prior to sample deposition were used as references in processing the sample spectra.

■

to obtain a C10H4O2F12Si molecule (Figure 4b). Additionally, a molecule formed from a substituted benzylic alcohol bearing two CF3 substituents, namely the 1,1,1,3,3,3-hexafluoro-2phenyl-2-propanol was also theoretically investigated (Figure 4c). For these three molecules, full optimizations were carried out in order to obtain the geometry of the most stable structure. These optimized geometries were then used to further build up the guess structures for the C18H8O2F12Si spirosilane. The optimized structure of this latter molecule is presented in Figure 4d. Based on the theoretical vibrational frequencies such obtained, it was possible to attribute the main bands experimentally observed. Since the error between the theoretical and the experimental frequencies is below 25 cm−1 for all the vibrational modes, it appeared that in order to attribute the frequencies experimentally observed: (i) the B3LYP/6-31+G* level of theory is accurate enough and (ii) the use of scaling factors is not necessary. These attributions are summed up in the Table 1. These attributions are in agreement with previous studies.16−20,30−32 4. Reactivity toward NO. Once the spectrum of the isolated 1 reagent was interpreted, the reactivity of this species toward NO was spectroscopically investigated under similar experimental conditions. Compared with the spectrum obtained without any NO radicals, new peaks were observed, especially in two regions: in the 1590−1300 and 850−600 cm−1 ranges. The former one is characteristic of the N−O bonds in vicinity of a Si atom whereas the later one is due to Si−N. Thus, the observation of new bands in these two regions is already a proof that a reaction between 1 and NO leads to the formation of a product containing one (or more) Si−N bonds. Nevertheless, in order to gain some more insights onto this reaction, the concentration effect was investigated. In such experiments, an increase of the amount of NO introduced may lead to the formation of 1-NO in higher proportions, and possibly 1-(NO)2. Concentration Effect. The influence of the NO amount on the products formed was investigated using two different approaches. First, the spectra obtained when a NO/Ne 0.5:100 mixture was going through the flask containing the derivative 1 at different times, were compared. Second, the spectra obtained when NO/Ne x:100 (x = 0.5, 1.0, or 2.0) mixtures were used during 30 min, were compared. As shown in Figure 5, some bands contain multiple peaks. This may be due to two different effects: (1) Since these spectra were registered in Ne matrix, splitting may be due to matrix site effects. Indeed, Ne atoms form a crystalline-like framework and the molecules of interest may be trapped in different types of sites in the matrix. Thus a same vibration mode, if affected by different matrix environments, can be observed on the IR spectrum as several peaks. (2) The

RESULTS AND DISCUSSION

1. IR Spectrum of the Reagent in Ne Matrix. When 1 is trapped in a Ne matrix without any NO radicals, characteristic peaks are observed in the 1400−600 cm−1 range. In order to assign the main peaks of this IR spectrum, geometry optimizations and frequency calculations were carried out. The theoretical IR spectrum can directly be compared with the experimental one, since the matrix isolation method is used to simulate the gas phase. The frequencies of main peaks theoretically calculated and experimentally observed, as well as their assignments in terms of vibrations (based on the theoretical calculations) are reported in Table 1. 2. IR Spectrum of the Reagent in Solid Phase. The global structure of the spectrum obtained in solid phase at 300K under atmospheric pressure is close to the one observed at low pressure and low temperature, as shown in Figure 3 and in Table 1. However, the bands are shifted, because the latter experiment was carried out in Ne matrix that reproduces the gas phase environment, whereas the former spectrum was registered from a solid sample. Furthermore, much wider bands are observed in solid phase. This is mainly due to two parameters: (1) the spectrum obtained in solid phase corresponds to a much more concentrated sample and (2) the Boltzman distributions strongly depend on the temperatures, and at 300K hot bands can be observed, as well as combination bands involving excited states. 3. Theoretical Calculations. The spectra obtained for the isolated spirosilane itself is characterized by numerous bands that cannot be easily interpreted without calculations. This is the reason why a theoretical investigation was carried out. Since the whole molecule is relatively large (117 vibration modes), three model molecules were considered prior to the study of the molecule itself. The optimized structures of these model molecules are presented in Figure 4. First of all, a C6H8O2Si molecule in which two five-membered heterocyclic ring involving C, H, O and Si atoms was considered (Figure 4a). From this molecule, four CF3 substituents were added in order 3299



Article

Figure 5. Influence of the NO amount on the peaks observed in the characteristic range of the N−O bonds disturbed by Si in the ranges 1360−560 (spectrum A) and 1495−1400 cm−1 (spectrum B). (a) NO/Ne 1:100 introduced alone; (b) 1 trapped in Ne matrix; 1 introduced with a NO/Ne mixture: (c) NO/Ne 2:100.

formation of Si-NO bonds leads to a splitting Δ of the orbitals caused by a strong-field ligand. For a sake of readability, only the main bands were considered in the following discussion. The bands are attributed to the following: α, the spirosilane; β and γ, the formation of 1-(NO) and 1-(NO)2, respectively.

The reactivity of Si atoms with NO was investigated both theoretically and experimentally in Ar matrix Zhou et al.22 To this end, they used laser-ablated silicon toms from a silicon target. Their results proved that the Si−NO monomer is characterized by a stretching bond around 1550 cm−1. More precisely, they experimentally observed the Si-NO compound 3300



Article

Table 2. Comparison of the IR Spectra Obtained in Ne Matrixes with Different Amounts of NOa C18H8O2F12Si in Ne matrix

C18H8O2F12Si with diluted NO

1695.3 1643.3 1591.1 1579.6 1505.8 1501.2 1474.6 1464.9 1453.1 1449.8 1323.6 1288.4 1251.8 1236.1 1222.2 1208.1 1175.5 1112.7 1104.4 985.4 979.2 924.7 848.5 791.1 772.4 745.5 714.3 709.8 637.5 595.3

1591.1 1579.6 1505.8

1579.6 1505.8

a

C18H8O2F12Si with concentrated NO 2:100

1464.9

1464.9

1449.8

1449.8

1288.4 1251.8 1236.1 1222.2

1288.4 1251.8 1236.1 1222.2

1175.5 1112.7 1104.4 985.4 979.2 924.7

1175.5 1112.7 1104.4 985.4 979.2 924.7 848.5 791.1 772.4

714.3

714.3

attribution 1-(NO) 1-(NO)2 1-(NO)

1-(NO)2 rocking of the benzenic H atoms 1-(NO)2

symmetric stretching C−C−C

1-(NO)2 stretching of the benzenic C−C antisymmetric stretching C−O antisymmetric stretching C−C antisymmetric stretching C−C antisymmetric stretching Si−O 1-(NO) 1-(NO) 1-(NO) 1-(NO)2 C−F vibrations 1-(NO)2 1-(NO)

Frequencies attributed to Si(NO) are in bold, whereas the ones attributed to Si(NO)2 are in italic.

at 1548.7 cm−1 in an argon matrix. Additionally, the theoretical spectrum they calculated at the B3LYP/6-311+G* level of theory shows a peak at 1589.1 cm−1. They also theoretically and experimentally investigated the formation of Si(NO)2, NSiO, Si-η2-NO and SiON. Their results are summed up in the Table 3.

Additionally, in the case of the spirosilane studied here, bands at 772.4, 791.1, and 848.5 cm−1 appear to be due to the interaction between Si and NO. Isotopic Effect. In order to further confirm the attributions, the reactivity between the spirosilane and 15 NO was investigated. In the IR spectra obtained under such experimental conditions, all the peaks attributed to a NObearing product should be shifted. Such shifts were indeed observed, and the IR spectra obtained are consistent with the formation of a Si−N bond (Table 4). Annealing. The above-proposed attributions may be confirmed by further experiments. Indeed, the controlled and gradual increase of the temperature leads to the diffusion of the species, thus allowing subsequent reactions between primary products. Thus, these kinds of experiments are useful to follow the reactivity, to propose reaction pathways and to draw up some conclusions about the relative stability of species. In the present case, the 1-(NO) product may further react with another NO molecule trapped in the Ne matrix to form 1(NO)2. This is the reason why, to gain some more insight into the nature, the stability and the reactivity of the product characterized by the peaks attributed to the formation of a (C9H4OF6)2Si(NO), the sample containing 1, NO radicals and primary products from the reaction between 1 and NO was further annealed step by step up to the temperature at which the Ne matrix evaporates. Thus, peaks specifically due to the reaction between the spirosilane and NO can be split into three categories depending on their evolution within the temper-

Table 3. Summary of the Theoretical and Experimental Results of Zhou et al.22 frequency (cm−1) attribution

experimental value in Ar matrix

theoretical value (B3LYP/6-311+G*)

SiNO Si(NO)2 NSiO Si-η2-NO SiON

1548.7 1492.0 1283.8 992.1 (not observed)

1589.1 1618.5 1307.2 1030.4 1151.7

Thus the peak at 1591.1 cm−1 observed under our experimental conditions when 1 is trapped in a Ne matrix with NO radicals, may be due to the formation of a Si-NO bond. The small shift between the frequency we obtained and the previously reported value may be due to the nature of the rare gas matrix as well as the presence of the C9H4OF6 substituents. 3301



Article

Table 4. Vibrational Frequencies Attributed to the NOBearing Products Formed after the Reactivity between the Spirosilane 1 and Either 14NO or 15NO 1 with concentrated 14NO 2:100

1 with concentrated 15NO 2:100

attribution

1695.3 1643.3 1591.1 1501.2 1453.1 1208.1 848.5 791.1 772.4 745.5 637.5 595.3

1665.3 1607.1 1564.9 1476.7 1448.6 1205.8 832.1 774.2 762.0 740.3 635.4 592.9

1-(NO) 1-(NO)2 1-(NO) 1-(NO)2 1-(NO)2 1-(NO)2 1-(NO) 1-(NO) 1-(NO) 1-(NO)2 1-(NO)2 1-(NO)

Figure 6. Possible structure for the product formed after the addition of a NO radical to the C18H8O2F12Si.

position. Such structures seem to be the most probable ones for the 1-NO intermediate, because of the formation of a hypervalent bond (three centers/four electrons).37,38 The structure of the more stable 1-(NO)2 products probably consists of an octahedral arrangement around the Si atom.

■

SUMMARY A sample was generated from the condensation at 5 K of the vapors obtained when a flux of Ne is going through a heated balloon containing 1. The IR spectrum thus obtained was characterized, and the signal observed was attributed thanks to DFT calculations. The reactivity of this compound toward NO was further investigated by replacing the Ne flux with a NO:Ne mixture. Thus, NO can react with 1. On the other hand, the Ne matrix and the low temperature prevent the formation of aggregates and limit consecutive reactions. Under these experimental conditions, the formation of a Si−NO bond was spectroscopically characterized, thus suggesting that the reactivity between C18H8O2F12Si, and NO leads to the formation of 1-NO. When this sample is annealed up to 11 K, NO radicals can diffuse in the Ne matrix and further react with 1-NO to form 1-(NO)2. Although these two species are stable, 1-NO acts as an intermediate in the formation of 1(NO)2. As a consequence, we expect that 1-(NO)2 will be the only product observed from the reaction between 1 and NO in condensed phase. Thus, the spirosilane investigated in this paper may be used as a molecular probe for the detection of NO radicals.

ature: (1) Some peaks have constant intensities between 5 and 9 K and then disappear. These peaks, observed at 595.3, 772.4, 791.1, 848.5, 1591.1, and 1695.3 cm−1 are due to the formation of an adduct between the spirosilane and NO radical: 1-NO. (2) Peaks at 1270.4, 1332.1, and 1527.9 cm−1 are not observed at 5 K, appear at 9 K and disappear at 11 K. These peaks are attributed to transient species. (3) Peaks at 637.5, 745.5, 1208.1, 1453.1, 1501.2, and 1643.3 cm−1 are observed only with a high amount of NO, and their intensity increase weakly during the annealing. These peaks are attributed to the formation of a dimer between the spirosilane and NO radicals: 1-(NO)2. Thus, 1-NO appears to be an intermediate species in the reaction leading to the formation of 1-(NO)2. 5. Chemical Interpretation. Many other complexes between NO and metals were isolated and characterized both experimentally and theoretically.33,34 Thus, numerous works were focused on the interactions between metallic centers and NO radicals. On the other hand, the reactivity observed in the case of the Martin’s spirosilane leads to the formation of a species in which the Si atom is hypervalent. As a consequence, it can act as an electrophilic center. Indeed, in this compound the Si atom is bonded to an oxygen atom and a carbon atom that are characterized by the following electronegativities: χ(Si) = 1.8 eV1/2, χ(C) = 2.5 eV1/2, χ(O) = 3.5 eV1/2, according to the Pauling scale. Additionally, oxygen atoms bounded to the silicon are electrondepleted, due to the presence of the CF3 substituents on the adjacent carbon atom. This may explain why a nucleophilic species such as NO ligand will react with C18H8O2F12Si to form 1-NO and also 1-(NO)2. Furthermore, the formation of a hypervalent species is favored by interactions involving a three centers and four electrons bond. Our experimental results clearly demonstrated the formation of the 1-NO and 1-(NO)2 products. On the other hand, the structure of such species is not yet clearly identified in the literature. Indeed, the formation of pentavalent siliconcontaining radicals has been reported a few times,5,8,35,36 but the structures for the same are still to be determined. As a consequence, in the case of the 1-NO intermediate observed in the present study, different geometries can be proposed, as suggested in Figure 6. The oxygen atoms of the ligands are in apical position, and the addition of the NO radical took place at an equatorial

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS L.F. thanks the Institut Universitaire de France. H.L. thanks the DGA for a Ph.D. grant. Technical assistance was generously offered through FR 2769. This work was supported in part by the SMART federation of the Université Pierre et Marie Curie, Paris VI, and the PCMI (Physique et Chimie des Milieux Interstellaires) grant.

■

REFERENCES

(1) Seinfeld, J. H.; Pandis, S. N. Atmospheric chemistry: from air pollution to climate change; John Wiley & Sons: New York, 1998. (2) Cheeseman, K. H.; Slater, T. F. An Introduction to Free Radical Biochemistry. Br. Med. Bull. 1993, 49, 481−493. (3) Stamler, J. S.; Singel, D. J.; Loscalzo, J. Biochemistry of Nitric Oxide and its Redox-Activated Forms. Science 1992, 258, 1898−1902. (4) Prokof’ev, A. I.; Prokof’eva, T. I.; Bubnov, N. N.; Solodovnikov, S. P.; Belostotskaya, I. S.; Ershov, V. V.; Kabachnik, M. I. Radical-

3302



Article

Anion Ion Pairs with a Six-Coordinated Silicon Atom and Nonequivalent Ligands. Russian Chemical Bulletin. 1979, 1726−1733. (5) Prokof’ev, A. I.; Prokof’eva, T. I.; Bubnov, N. N.; Solodovnikov, S. P.; Belostotskaya, I. S.; Ershov, V. V.; Kabachnik, M. I. Ion Pairs in Radical Anions with Six-Coordinated Silicon Atoms. Russ. Chem. Bull. 1978, 27, 1321−1327. (6) Prokof’ev, A. I.; Prokof’eva, T. I.; Bubnov, N. N.; Solodovnikov, S. P.; Belostotskaya, I. S.; Ershov, V. V.; Kabachnik, M. I. Mechanism of the Intramolecular Exchange of the Unpaired Electron and Bond In Free O-Substituted Phenoxyl Radicals of the Esters of Orthosilicic Acid. Russ. Chem. Bull. 1979, 28, 1820−1827. (7) Rakhimov, R. R.; Prokof’ev, A. I.; Lebedev, Ya. S. Molecular and Chemical Dynamics of a Biradical with Six-Coordinate Silicon Atom. Russ. Chem. Bull. 1993, 42, 1648−1652. (8) Alberti, A.; Pedulli, G. F. Addition Reactions Of Silyl Radicals To Unsaturated Compounds. Rev. Chem. Intermed. 1987, 8, 207−246. (9) Chekalov, A. K.; Prokof’ev, A. I.; Bubnov, I. I.; Solodovnikov, S. I.; Zhdanov, A. A.; Kabachnik, M. I. Silicon-Containing Radicals and Diradicals with 3,5-di-tert-butylpyrocatechol Ligands. Russ. Chem. Bull. 1981, 30, 2064−2071. (10) Perozzi, E. F.; Martin, J. C. Facile Syntheses of Isolable Organic Derivatives of Hypervalent Sulfur, Phosphorus, and Silicon. Introduction of a Stabilizing Bidentate Ligand via its Dilithio Derivative. J. Am. Chem. Soc. 1979, 101, 1591−1593. (11) Perozzi, E. F.; Michalak, R. S.; Figuli, G. D.; Stevenson, W. H., III; Dess, D. B.; Ross, M. R.; Martin, J. C. Directed Dilithiation of Hexafluorocumyl Alcohol-Formation of a Reagent for the Facile Introduction of a Stabilizing Bidentate Ligand in Compounds of Hypervalent Sulfur (10-S-4), Phosphorus (10-P-5), Silicon (lO-Si-5), and Iodine (10-I-3). J. Org. Chem. 1981, 46, 1049−1053. (12) Stevenson, W. H., III; Wilson, S.; Martin, J. C.; Farnham, W. B. Pseudorotational Mechanism for the Inversion of 10-Si-5 Siliconates: Ligand Structure and Reactivity. J. Am. Chem. Soc. 1985, 107, 6340− 6352. (13) Jacox, M. E. The Spectroscopy of Molecular Reaction Intermediates Trapped in the Solid Rare Gases. Chem. Soc. Rev. 2002, 31, 108−115. (14) Bondybey, V. E.; Smith, A. M.; Agreiter, J. New Developments in Matrix Isolation Spectroscopy. Chem. Rev. 1996, 96, 2113−2134. (15) Almond, M. J.; Wiltshire, K. S. Matrix Isolation. Annu. Rep. Prog. Chem., Sect. C 2001, 97, 3−60. (16) Withnall, R.; Andrews, L. Matrix reactions of silane and oxygen atoms. Infrared spectroscopic evidence for the silanol, silanone, and silanoic and silicic acid molecules. J. Phys. Chem. 1985, 89, 3261−3268. (17) Shirk, A. E.; Shirk, J. S. Oxygen Atom Reactions with SiHCl3 in Solid Argon: the Infrared Spectrum of SiCl3OH. J. Mol. Spectrosc. 1982, 92, 218−228. (18) Arrington, C. A.; West, R.; Michl, J. On the Proposed Thermal Interconversion of Matrix-Isolated Dimethylsilylene and 1-methylsilene: their Reactions with Oxygen Atom Donors. J. Am. Chem. Soc. 1983, 105, 6176−6177. (19) Withnall, R.; Andrews, L. Matrix Reactions of Methylsilanes and Oxygen Atoms. J. Phys. Chem. 1988, 92, 594−602. (20) Hastie, J. W. Reactions and Structures of Matrix-Isolated SiO Species at Low Temperatures from Infrared Spectra. Inorg. Chim. Acta 1969, 3, 601−606. (21) Withnall, R.; Andrews, L. Infrared spectroscopic evidence for silicon-oxygen double bonds: methyl- and dimethylsilanones in solid argon. J. Am. Chem. Soc. 1986, 108, 8118−8119. (22) Zhou, M.; Jiang, L.; Xu, Q. Reactions of Silicon Atoms with NO. Experimental and Theoretical Characterization of Molecules Containing Si, N, and O. J. Phys. Chem. A 2004, 108, 9521−9526. (23) Andrews, L.; McCluskey, M. Bending modes of SiO2 and GeO2 in solid argon. J. Mol. Spectrosc. 1992, 154, 223−225. (24) Frisch, M. J.; et al. Gaussian 03; Gaussian, Inc.: Pittsburgh, PA, 2002. (25) Becke, A. D. Density-Functional Exchange-Energy Approximation with Correct Asymptotic Behavior. Phys. Rev. A 1988, 38, 3098−3100.

(26) Lee, C.; Yang, W.; Parr, R. G. Development of the Colle-Salvetti Correlation-Energy Formula into a Functional of the Electron Density. Phys. Rev. B 1988, 37, 785−789. (27) Stevens, J. E.; Macomber, L. D.; Davis, L. W. IR Spectra and Vibrational Modes of the Hydrofluoroethers CF3OCH3, CF3OCF2H, and CF3OCF2CF2H and Corresponding Alkanes CF3CH3, CF3CF2H, and CF3CF2CF2H. Open Phys. Chem. J. 2010, 4, 17−27. (28) Ignatyev, I. S.; Montejo, M.; Partal Urena, F.; Sundius, T.; Lopez Gonzalez, J. J. Vibrational Spectra and Structure of Methoxysilanes and Products of their Hydrolysis. Vibr. Spectrosc. 2006, 40, 1−9. (29) Zins, E. L.; Joshi, P.; Krim, L. Reactivity between a COContaining Ice and Non-Energetic OH Radicals. Ap. J. 2011, 738, 175/1−175/7. (30) Forney, D.; Jacox, M. E.; Irikura, K. K. Matrix Isolation Study of the Interaction of Excited Neon Atoms with CF4. Infrared Spectra of CF3+ and CF3−. J. Phys. Chem. 1994, 101, 8290−8297. (31) Jacobs, J.; Willner, H.; Pawelke, G. Fourier Transform Infrared Spectra of the Complexes XCN-HF (X = F, CF3, SF5) Isolated in an Argon Matrix. J. Phys. Chem. 1992, 96, 5793−5796. (32) White, L. D.; Tripp, C. P. A Low-Frequency Infrared Study of the Reaction of Methoxymethylsilanes with Silica. J. Colloid Interface Sci. 2000, 224, 417−424. (33) Bauschlicher, C. W., Jr.; Bagus, P. S. NiNO. A Model for NO on a Metal Surface. J. Chem. Phys. 1984, 80, 944−947. (34) Andrews, L.; Citra, A. Infrared Spectra and Density Functional Theory Calculations on Transition Metal Nitrosyls. Vibrational Frequencies of Unsaturated Transition Metal Nitrosyls. Chem. Rev. 2002, 102, 885−912. (35) Corriu, R.; Guérin, C.; Henner, B.; Wang, Q. Pentavalent Hydridosilicates: Some Aspects of the Reactivity of Potassium Tetraethoxyhydridosilicate. J. Organomet. Chem. 1989, 365, C7−C10. (36) Prokof’ev, A. I.; Prokof’eva, T. I.; Belostotskaya, I. S.; Bubnov, N. N.; Solodovnikov, S. P.; Ershov, V. V.; Kabachnik, M. I. Tautomerism in Silicon-Containing Free Radicals. Tetrahedron 1979, 35, 2471−2482. (37) Curnow, O. J. A Simple Qualitative Molecular-Orbital/ValenceBond Description. J. Chem. Educ. 1998, 75, 910−915. (38) Denmark, S. E.; Beutner, G. L. Lewis Base Catalysis in Organic Synthesis. Angew. Chem., Int. Ed. 2008, 47, 1560−1638.

3303


Study of Interaction between NO Radicals and Martin's Spirosilane by

Recommend Documents