7312
J. Phys. Chem. B 2007, 111, 7312-7317
Self-Organization of Water in Lithium/Nitrobenzene System Greg Moakes,† Luke L. Daemen,‡ Leslie T. Gelbaum,† Johannes Leisen,† Vladimir Marecek,§ and Jiri Janata*,† School of Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, Georgia 30332, Manuel Lujan, Jr. Neutron Scattering Center, Los Alamos National Laboratory, Los Alamos, NM 87545, and J. HeyroVsky Institute of Physical Chemistry, Academy of Sciences of the Czech Republic, DolejskoVa 3, 182 23 Praha, Czech Republic ReceiVed: March 11, 2007; In Final Form: April 18, 2007
The effect of lithium ion on the ordering of water in water-saturated nitrobenzene has been probed by 2H NMR, diffusion ordered spectroscopy and neutron scattering. It was shown that increased water concentration in LiClO4/wet nitrobenzene results in the formation of a metastable solvatomer with mixed water and nitrobenzene character, Li+(W/NB). This species is shown to decay over hours to two solvatomers, one dominated by nitrobenzene Li+(NB) and the other dominated by water Li+(W). To confirm the assignment of these solvation states, diffusion ordered deuterium NMR spectroscopy has been used to elucidate the hydrodynamic radii of these solvatomers. Neutron scattering yields vibrational spectroscopy information that shows how addition of lithium to the nitrobenzene/water system results in relatively slow self-organization of the water environment (hours).
1. Introduction Study of the interface between two immiscible electrolyte solutions (ITIES) can impact many fields. For example, in biochemistry, an understanding of interfacial structure, kinetics, and thermodynamics is critical for biological membrane modeling.1 In electrochemistry, liquid-liquid interface studies have been used in sensor design1-3 and to understand processes occurring in ion-selective electrodes with liquid membranes.1,4 In the field of ITIES, the most typical interface is that of nitrobenzene/water. Previous studies have shown that having traversed the liquid-liquid interface, ions exist in a number of solvation environments (solvatomers) with various contributions from water and nitrobenzene.5,6 The extent of hydration of ions in organic solvents has been studied by spectroscopic techniques such as FTIR and NMR,7-10 yet the identities of solvatomers that an ion forms in nitrobenzene are still not fully understood. Proton NMR has been used to elucidate the hydration numbers of different anions in nitrobenzene11 and to explain the effect of water content on the hydration of ions in nitrobenzene.10 The extent of hydrogen bonding of water in organic solvents has been extensively studied by FTIR spectroscopy.12 Furthermore, the hydrodynamic radii of various ions have been estimated from consideration of crystallographic radii and hydration numbers derived from Karl Fischer titration.10 The solvation environments of the lithium ion in water-saturated nitrobenzene have been investigated previously by 7Li NMR and FTIR resulting in identification of multiple solvatomers of Li+ in the Li+/nitrobenzene/water system.5,6 These solvatomers are: Li+ solvated by a mixture of water and nitrobenzene, primarily nitrobenzene and primarily water and will henceforth be termed Li+NB/W, Li+NB, and Li+W, respectively. Here we * Corresponding author. Phone: 404 894 4828. E-mail: jiri.janata@ chemistry.gatech.edu. † Georgia Institute of Technology. ‡ Los Alamos National Laboratory. § Academy of Sciences of the Czech Republic.
report the use of 2H NMR and neutron scattering to gain further insight into the dynamics of the solvation environment of Li+ in the mixed Li+/nitrobenzene/water system. The use of deuterium as a reporter instead of lithium means that a comparison can be made between NB/W solutions with and without lithium species. Diffusion ordered spectroscopy (DOSY) was used to determine the apparent hydrodynamic radii of the lithium ion in the Li+/nitrobenzene/water system. This approach has not been possible with 7Li NMR since the concentration of LiBr in the water-saturated nitrobenzene is ∼1 × 10-5 M, which is too low to obtain DOSY spectra within the short experimental window. Further evidence for the ordering effect of the lithium ion on water dissolved in nitrobenzene is presented here through the use of neutron scattering, which yields vibrational information of the water protons. This has several advantages over the optical counterparts and infrared and Raman spectroscopies since neutron scattering has great sensitivity to hydrogen, no selection rules, penetrability through matter (including bulky pressure cells), and simplicity of modeling the neutron-nucleus interaction in the quantitative prediction of the vibrational spectrum. The large incoherent scattering cross section of hydrogen (80.3 barns) compared to the lower incoherent scattering cross section of deuterium (2.1 barns) makes it possible to use selective isotopic substitutions to highlight contributions from hydrogen. In our case, we used protonated water (H2O) and deuterated nitrobenzene (C6D5NO2) to enhance the water signal while decreasing the scattering from nitrobenzene. 2. Experimental Methods 2.1. Materials. Lithium perchlorate (g99%, Aldrich) and lithium bromide (g99%, Aldrich) were used as received. Nitrobenzene (99%, ACROS Organics) was redistilled under vacuum and dried before use by molecular sieves (4A, SigmaAldrich). Glass 5 mm and 10 mm NMR sample tubes were obtained from Wilmad-Lab glass. All water used was deionized
10.1021/jp071972x CCC: $37.00 © 2007 American Chemical Society Published on Web 06/07/2007
Self-Organization of Water to 18 MΩ using a US-Filter Modulab water system (US Filter, Warrendale, PA). D2O (99.9%, Cambridge Isotope Laboratories) was used as received. 2.2. Procedures. Saturated aqueous solutions of LiClO4 and LiBr were prepared by adding each salt to D2O until solid salt remained. For preparation of ‘‘wet” NB solutions of LiClO4 and LiBr, each salt was added to D2O until saturated. This saturated aqueous solution was then stirred with vacuum distilled nitrobenzene for 3 h and then allowed to separate for 1 h. When the clear NB phase was pipetted from the equilibrating mixture, the experimental clock was set to t ) 0. The lithium content of the “wet” NB solution has been determined by atomic absorption spectroscopy to be 9.52 × 10-4 M for lithium perchlorate and 1.7 × 10-5 for lithium bromide.5,6 The published concentration of water in water-saturated “wet” NB is between 170 and 190 mM.9 2.3. NMR and Neutron-Scattering Experiments. The 1D NMR spectra were recorded using a Bruker AMX-400 spectrometer (Bruker BioSpin, Rheinstetten, Germany) with a 10 mm broadband probe. A 2H NMR spectrum of saturated (6M) LiClO4 in D2O was recorded before each experiment and the corresponding signal calibrated to 0 ppm. For the kinetic data, peak areas were calculated by a line fitting function of the Mestre-C NMR software (Mestrelab Research SL, Santiago de Compostela University, Santiago, Spain). The line fit function uses the Levenberg-Marquardt nonlinear least-squares and Downhill Simplex (Nelder and Mead) algorithms for estimating peak parameters (position, intensity, line width and line shape function). Temperature of the NMR probe was controlled to an accuracy of (1 K using the spectrometers VT accessories. No field frequency lock was used during the data acquisition. 2HDOSY experiments were recorded using a Bruker DRX-500 spectrometer (Bruker BioSpin, Rheinstetten, Germany) with a 5 mm broadband probe with z gradients. For the neutron scattering experiments, the Filter Difference Spectrometer (FDS) at the Manuel Lujan, Jr. Neutron Scattering Center (Los Alamos National Laboratory) was used. It measures the vibrational spectrum in neutron energy loss mode between 40 and 4000 cm-1 with a white beam of neutrons. Determination of the neutron incident energy is by time-of-flight, whereas the final neutron energy is imposed by a set of beryllium filters between the sample and the detector banks. The instrument resolution varies from 4% to 7% across the energy range. This resolution can be further improved by the use of maximum entropy techniques during data analysis to 2-5%. Protonated water (H2O) and deuterated nitrobenzene (C6D5NO2) were used to enhance the water signal relative to nitrobenzene. Because the concentration of water is small in comparison to the bulk nitrobenzene, the scattering contribution from deuterated nitrobenzene remains significant. For this reason we collected a blank (pure nitrobenzene-d5), as well as a spectrum for nitrobenzene-d5 saturated with H2O and another nitrobenzene-d5 sample stirred for 18 h with H2O saturated with LiBr. The samples were placed in annular aluminum sample holders, 20 mm in diameter with an annular gap of 2 mm and a height of 100 mm. The aluminum wall thickness was 0.8 mm. All samples were quenched (from room temperature) in liquid nitrogen before being cooled rapidly to 12 K. When the temperature reached 12 K, data collection started. Cooling was necessary because the vibrational modes broaden considerably with temperature. Because of the relatively high viscosity of the solutions and the rapid quenching, the material is amorphouss as verified by low-temperature X-ray diffraction of the quenched material.
J. Phys. Chem. B, Vol. 111, No. 25, 2007 7313
Figure 1. 2H NMR spectrum of nitrobenzene saturated with D2O (W/ NB) and nitrobenzene saturated with LiClO4 in D2O (Li(W/NB)). Experiments performed at 298 K.
Figure 2. 2H NMR spectrum of nitrobenzene equilibrated with D2O (W/NB) at 298 and 290 K (A, B). Nitrobenzene equilibrated with saturated LiClO4 in D2O (Li/W/NB) at 298 and 290 K (C,D).
3. Results and Discussion 3.1. Effect of Lithium Perchlorate on 2H Chemical Shift. The effect of LiClO4 on the ordering of water in nitrobenzene was studied by a simple 1D 2H NMR experiment (Figure 1). In this particular experiment, a pure D2O sample was calibrated to 0 ppm then two spectra were recorded in succession: spectrum 1 (solid black line) is the 2H NMR spectrum for nitrobenzene saturated with D2O. Spectrum 2 (dashed line) is the 2H NMR spectrum for nitrobenzene equilibrated with 6 M LiClO4 in D2O. Both experiments were repeated 5 times with observed chemical shift differences of less than 0.005 ppm. This confirms that the downfield shift after addition of LiClO4 is a real result rather than a drift in the instrument. This shift of the 2H resonance toward 0 ppm suggests an increase in water ordering leading to a 2H resonance that resembles that of bulk D2O. 3.2. Effect of Cooling on 2H Spectrum. At room temperature, a solution of LiClO4 in water-saturated nitrobenzene gives a single, clearly resolved resonance in a 1D 7Li spectrum.5 This species is hypothesized to be lithium ion with a solvation sphere dominated with nitrobenzene. Upon cooling from 298 to 290 K, a second resonance appears in the 7Li spectrum that has been attributed to a higher energy metastable water aggregate, with a mixed solvation shell of water and nitrobenzene Li+W/NB. This solvatomer is metastable and slowly decays over period of tens of minutes. In the presence of glass it forms a stable species Li+W, which contains mostly water and which is nucleated at the hydrophilic glass surface. In the absence of glass surface (i.e., in a Teflon lined NMR tube) it converts to the nitrobenzene solvated species, Li+NB. Figure 2 shows the effect of this cooling process on 2H chemical shift for nitrobenzene stirred with D2O
7314 J. Phys. Chem. B, Vol. 111, No. 25, 2007
Moakes et al.
Figure 3. 7Li NMR spectra of nitrobenzene saturated with LiClO4 in H2O at (A) 298 and (B) 290 K.
in the presence of LiClO4. Spectra A and B are wet nitrobenzene at 298 and 290 K, respectively. Spectra C and D are LiClO4 in wet nitrobenzene at 298 and 290 K, respectively. The cooling of both solutions results in the subtle shift (0.03 ppm) of the 2H chemical shift downfield, toward 0 ppm. This suggests that in both solutions there is an aggregation of water upon cooling. It is interesting to note that the chemical shift difference upon addition of lithium ion (0.095 ppm, Figure 1) is large in comparison to that caused by cooling. This is in direct support of previous FTIR data which suggests the ordering of water promoted by Li+ is large in comparison to that caused by cooling. However, cooling does not result in the occurrence of multiple resonances as seen in the 7Li NMR spectrum (Figure 3). It is important to remember that this is the same experiment only using a different “reporter” species. The apparent discrepancy in this result can be explained by considering the abundance of these reporter species. When the cooling experiment is performed and followed by 7Li NMR (Figure 3), the solution evidently becomes supersaturated with water, and becomes microheterogeneous; i.e., the decrease in temperature causes clusters of water to form. These results indicate that lithium partitions into this phase resulting in two “solvatomers”, the solvatomer consisting of lithium solvated in majority by nitrobenzene (1.163 ppm), and the solvatomer which results from cooling and resembles lithium solvated by a mixed solvation shell (0.776 ppm). The water clusters that form after a modest temperature drop of 8 K are a small fraction of the overall water content of nitrobenzene. It is for this reason that in the case of 2H spectrum (Figure 2B) the second solvatomer is unobservable. Conversely with 7Li, the lithium species preferentially solvate with water thus ‘illuminating’ this second, metastable, water-rich solvatomer Li+W/NB. 3.3. Addition of Bulk Water. In our previous work,5 it was found that cooling of the LiClO4/nitrobenzene/water system resulted in formation of multiple solvatomers (Figure 3). In this case, the water content of the wet nitrobenzene is ∼200 mM, compared to a LiClO4 concentration of 9 × 10-4 M; it is therefore clear that water dissolved in nitrobenzene is in great excess of lithium species. In the recent 2H NMR studies, the formation of these multiple solvatomers is undetectable due to the strong resonance of the dissolved deuterium. Previous studies
Figure 4. 2H NMR spectra of LiClO4 (wet NB) with 5 µL of D2O added and shaken. t ) (A) 0, (B) 90, and (C) 180 min.
also determined that the same effect could be observed by addition of a small amount of water to the solution on the surface of a wet glass capilary. It was decided that the concentration of the ordered water species could be enhanced to the point of detection by such methods. To the initial LiClO4/wet nitrobenzene system an additional 5 µL of D2O adhering to the surface of a glass capillary was added. Figure 4A-C below shows the kinetics of solvatomer transformation at t ) 0 (insertion of wet glass capillary), t ) 90 min, and t ) 180 min, respectively. It is clear from Figure 4 spectrum C that there are three primary solvatomers. Pure D2O is assigned to 0 ppm. Nitrobenzene containing LiClO4/D2O gives a single resonance at ∼-2.7 ppm. It is therefore clear that the downfield peak at -0.316 ppm is a mixed solvation shell with majority water. Conversely, the peak at -2.710 ppm represents a species almost entirely solvated by nitrobenzene. The shoulder which grows at 0.040 ppm is consistent with 7Li NMR results and represents a large water aggregate nucleated at the glass capillary surface. The change in relative intensities of the three peaks suggest that the solvatomer, which is formed immediately on addition of 5 µL D2O by the wetted capillary is metastable and appears to exchange its mixed solvation sphere for one dominated by water (0.040 ppm) and one dominated by nitrobenzene (-2.704 ppm). The results of these experiments are summarized in Figures 5 and 6. Figure 5 represents the decay of the peak at ∼-0.3 ppm (Figure 4) corresponding to Li+NB/W. The two curves represent the decay of this metastable species without (A) and with (B) presence of LiClO4 in the D2O. In both cases, the logarithm of (peak area) versus time is linear indicating first-
Self-Organization of Water
J. Phys. Chem. B, Vol. 111, No. 25, 2007 7315 fusion coefficients were measured for the 2H resonance at ∼-2.7 ppm, corresponding to the Li+(NB) solvatomer and at ∼0.3 ppm, corresponding to the Li+(W/NB) solvatomer. The resulting hydrodynamic radii, rh, were calculated from the Stokes-Einstein equation:13
D ) kT/6πηrh
Figure 5. Decrease in concentration of the (NB/W) solvatomer without (A) and with (B) presence of LiClO4 (6M) in D2O during preparation of wet nitrobenzene.
Figure 6. Increase in concentration of the (NB) solvatomer without (A) and with (B) presence of LiClO4 (6M) in D2O during preparation of wet nitrobenzene. Increase of Li+(W) solvatomer (C).
TABLE 1: Diffusion Coefficients and Estimated Hydrodynamic Radii for Water Species in Various Nitrobenzene Environments solution nitrobenzene/D2O nitrobenzene/LiBr(D2O) nitrobenzene/LiClO4(D2O)
diffusion coefficient, D (m2 s-1) rh, (Å) 2.33 × 10-9 ( 5 × 10-11 2.47 × 10-9 ( 5 × 10-11 2.31 × 10-9 ( 5 × 10-11
1.05 0.992 1.06
order kinetics. The rate of decrease for the peak area is much greater in the presence of lithium ion than without it, suggesting that Li+ increases the rate of decomposition for this species. The increase in the area of the most upfield peak (∼-2.7 ppm, Figure 4) corresponding to Li+(NB) is shown in Figure 6. Plots A and B represent the experiment performed without and with presence of LiClO4 in the D2O. The increase in the area of the most downfield peak (0.040 ppm, Figure 4) corresponding to Li+(W) is shown in plot C. There is no observable growth in this peak without presence of lithium. The increase in peak areas shown in plots A B and C do not follow first-order kinetics. On decay of the metastable Li+(NB/W) species, there is clearly a subsequent transport of water from the capillary wall followed by incorporation of this water into the Li+(W) and Li+(NB) species. The two superimposed processes make kinetics difficult to interpret, nonetheless, the kinetics of the process is comparable to that induced by cooling. 3.4. DOSY Studies. Table 1 shows the effect of lithium species on the diffusion coefficient of D2O in nitrobenzene. Diffusion coefficients were obtained from pulse field gradient, diffusion ordered spectroscopy (DOSY) experiment. The dif-
where D is the diffusion coefficient for the species (m2 s-1), k is the Boltzmann constant (1.3807 × 10-23 m2 kg s-2 K-1), η is the viscosity of the medium (8.94 × 10-4 kg m-1 s-1 for nitrobenzene), and rh is the hydrodynamic radius of the species (m). Somewhat surprising is the finding that the ordering effect of lithium on water in nitrobenzene, does not result in a significant increase of the hydrodynamic radii for the solvatomer Li+NB- at ∼-2.7 ppm. The diameters of the hydrated species in nitrobenzene are all smaller than that found for bulk water, which has a self-diffusion coefficient of 2.13 × 10-9 m2 s-1 and rh of 1.5 Å. This is intuitive since the decreased hydrogen bonding between water molecules in a hydrophobic medium reduces hydrodynamic radius versus that of bulk water. However, the fact that lithium does not increase the hydrodynamic radius significantly is surprising. Changes in the free OH stretch of FTIR spectra prove that lithium decreases “free water” content of nitrobenzene by increasing hydrogen bonding. Therefore, there is a discrepancy between FTIR and DOSY results. FTIR evidence proves irrefutably that addition of lithium to the wet nitrobenzene (observed in the 2H experiments at ∼-2.7 ppm) results in an increase in hydrogen-bonding of the water dissolved in nitrobenzene.5 7Li NMR results of LiBr in wet nitrobenzene suggest that lithium can order water to the point that a second clear solvatomer is observed, Li+NB/W, which is proposed to be an aggregate of water molecules within the nitrobenzene. However, this does not mean that the solvatomer Li+NB (2H, -2.7 ppm) should experience an increase in size, observed as a decrease in mobility observed by diffusion spectroscopy (DOSY). As such, the failure of lithium to increase the hydrodynamic radii of the solvatomer at ∼-2.7 ppm in the 2H spectra, does not contradict the earlier 7Li experiments. It is possible that in the case of the Li+NB solvatomer, addition of lithium does not lead to “pockets” of isolated hydrated species in the organic phase, but moreover a network of H-bonded water homogenously dispersed throughout the nitrobenzene. We propose here that addition of lithium species to wet nitrobenzene increases hydrogen bonding but not, as first thought, resulting in aggregates of hydrated lithium but in a hydrogen-bonded network of water molecules. This both explains the FTIR results and the fact that hydrodynamic radii elucidated from DOSY are small, on the order of single water molecules. A simple calculation of the Loschmidt’s number for lithium (ions per cm3) reveals an interatomic separation of ∼100 Å for the lithium cations, making it feasible that Li+ orders water through hydrogen bonding of its multiple solvation shells. DOSY data was also collected for the 2H resonance corresponding to the solvatomer Li+(NB/W) (-0.316 ppm, Figure 4), which has a diffusion coefficient of 2.25 × 10-9 m2 s-1, which is closer to that of bulk water (2.13 × 10-9 m2 s-1). This makes sense, since we know that the peak at -0.316 ppm in the 2H spectra results from addition of excess water. We attempted to measure the diffusion coefficient for the “shoulder” peak at 0.040 but the species did not diffuse sufficiently, adding evidence to the hypothesis that this resonance represents hydrated lithium immobilized at the glass capilary. A second possible explanation for the discrepancy between FTIR and DOSY results could result from a misinterpretation
7316 J. Phys. Chem. B, Vol. 111, No. 25, 2007
Figure 7. Neutron vibrational spectrum of d5-nitrobenzene saturated with water (dashed line) and d5-nitrobenzene saturated with LiBr(aq) (solid line).
of the DOSY results. When diffusion spectroscopy is used to measure diffusion of water, it is usually not considered that DOSY measures the diffusion of the NMR active nucleus irrespective of whether it truly represents the entire species. In the context of our experiment, it is possible that the diffusion of 2H (instead of D2O) is being measured (hence the large value of D) as the deuterium ‘hops’ from one water molecule to another. This is largely ignored,14,15 and a useful experiment to confirm or deny this would be to compare diffusion studies on both O and H of water. While we have used 2H to probe diffusion, 18O could also be used to determine if the diffusion coefficients vary for the entire molecule. If the values of D varies, the use of DOSY for study of water diffusion would be hampered by this effect. We do not currently have a sufficiently sensitive instrument for study of 18O DOSY, but future studies are desirable. 3.5 Neutron-Scattering Studies. Figure 7 shows the neutron vibrational spectra for nitrobenzene-d5 saturated with H2O (dashed line) and LiBr(aq, saturated) (solid line). Previous studies have proven LiBr to be particularly effective at ordering water in nitrobenzene5 hence its choice for this experiment. We also collected a vibrational spectrum of pure nitrobenzene-d5 as a blank. The internal modes of nitrobenzene-d5 were clearly observed above 500 cm-1 in all the vibrational spectra and are of little interest per se. The high-frequency part of the spectrum was not affected by the presence of water and/or lithium bromide and was used to perform a (normalized for counting time and sample volume) blank subtraction. The low-frequency part of the (blank-subtracted) spectra for nitrobenzene-d5 + H2O and nitrobenzene-d5 + H2O + LiBr is shown in Figure 7. The band that extends from 400 to 450 cm-1 (“400 cm-1 band”) is one (or more) librational modes of water. For a water molecule in a general force field, there will be three librations, one for each axis of rotation, but these will be seen only if there is a restoring force present for each libration. There is the “rock”, which is motion in the plane of the water molecule; the “twist”, which is a rotation about the C2 diad, and the “wag”, which is a rotation about the H-H axis, (typically) with increasing frequency.16 In free water these librational modes are weak and appear as extremely broad, ill-defined bands in Raman and IR spectra, where they are at all detectable. In systems in which water appears as a ligand to a metal cation, these librational modes become much more pronounced.17 Notice that, unlike with optical spectroscopy, the librational modes are usually the strongest contribution to the vibrational spectrum of water. Just as interesting is the disappearance of intensity in the vibrational spectrum at 175 and 220 cm-1. These modes are usually attributed to hindered translations in ice and other forms of associated water molecules (clusters). According to Bertie and Whalley.18 these modes are to be assigned to maxima in the density of vibrational states of water clusters owing to longitudinal acoustic and transverse optic vibrations, respectively. Their disappearance in the vibrational spectrum upon
Moakes et al. addition of LiBr is very likely indicative of a drastic rearrangement of water molecules in the system. This is consistent with a transition from water clusters (“free water”) in wet nitrobenzene to a more ordered state in which the water molecules rearrange around the lithium cations upon addition of lithium bromide to the system. Compared to wet nitrobenzene, the spectrum of wet nitrobenzene saturated with lithium bromide shows intensity appearing as a strong shoulder at about 269 cm-1 on the high-frequency side of the 258 cm-1 peak (arrow in Figure 7). This mode is the Li-O stretching mode observed by Rudolph et al.19 in the Raman spectrum of dilute solutions of lithium chloride and lithium bromide in water. According to Rudolph et al., the frequency of this mode depends little on the nature of the anion and the concentration of the salt in pure water (LiBr, 256-260 cm-1; LiCl, 260 cm-1; LiClO4, 250 cm-1). A calculation of the Li-O stretching frequency of a [Li(OH2)4]+ ion with Gaussian0320 (MP2/6-311G) gives 266 cm-1 for the (unscaled) frequency of the Li-O stretch. Another mode associated with the presence of lithium bromide appears, albeit more weakly, at 191 cm-1. Its position agrees well with that observed by Rudolph et al.19 for the O-H. Br stretch (190-204 cm-1 for aqueous LiBr, depending on concentration and 176-191 cm-1 for LiCl, also a function of concentration). The peaks at 258, and 350 cm-1 are largely unaffected by the presence of LiBr in the system. However they are not present in the blank (pure nitrobenzene-d5) and are undoubtedly associated with the presence of water in the system. The region between 280 and 320 cm-1 is also characterized by a decrease in intensity upon addition of LiBr to the system. It is possible that these modes result from the weak association (hydrogen bonding with the nitro group) of water with nitrobenzene in wet nitrobenzene. This association is then modified upon clustering of water around the lithium cations when lithium bromide is added to the system. Additional computational modeling will be necessary to clarify this. 4. Conclusions The use of deuterium as a reporter provides further evidence that multiple solvatomers can be observed in the Li/nitrobenzene/water system studied. This technique has provided a method by which to compare water organization in nitrobenzene with and without the presence of lithium species. Kinetic studies have supported previous reports that perturbation of the system by increase of water concentration leads to three solvatomers that undergo a slow dynamic exchange over 180 min. The main feature of this kinetics appears to be the appearance and subsequent decomposition of a metastable solvatomer consisting of Li+ solvated by a mixture of water and nitrobenzene. As this species decomposes, the resulting “free” water appears to be incorporated in the species we have termed Li+NB and Li+(W). These are solvatomers comprising a majority of nitrobenzene and water, respectively. Diffusion ordered spectroscopy has elucidated diffusion coefficients and hydrodynamic radii for water dissolved in nitrobenzene. Addition of lithium to the system has a negligible effect on both. It was noted that the hydrodynamic radius of water species in nitrobenzene Li+(NB) is somewhat smaller than that of bulk water. It was observed that the solvatomer Li+(NB/W) has a similar hydrodynamic radius to that of bulk water. The major conclusion from the DOSY studies is that while addition of lithium to wet nitrobenzene results in increased hydrogen bonding, it does not promote clustering of water molecules, but instead induces a formation of interconnected network of water.
Self-Organization of Water Neutron vibration studies have added further evidence to the idea that addition of lithium to water-saturated nitrobenzene results in a change in the hydrogen-bonding character of water. Acknowledgment. Support of this work by the Georgia Research Alliance Grant GRA.CG06.D and by the Ministry of Education, Youth and Sports of the Czech Republic Grant LC06063 are greatly appreciated. References and Notes (1) Samec, Z. Pure Appl. Chem 2004, 76, 2147. (2) Kakiuchi, T.; Senda, M. Bull. Chem. Soc. Jpn. 1984, 57, 1801. (3) Marecek, V.; Samec, Z. Anal. Lett. 1981, 14, 1241. (4) Koryta, J. Electrochim. Acta 1988, 33, 189. (5) Moakes, G.; Gelbaum, L. T.; Leisen, J. E.; Janata, J.; Marecek, V. J. Electroanal. Chem. 2006, 593, 111. (6) Moakes, G.; Gelbaum, L. T.; Leisen, J. E.; Janata, J. Faraday Trans. 2005, 129, 81. (7) (7) Osakai, T.; Ogawa, H.; Ozeki, T.; Girault, H. H. J. Phys. Chem. B 2003, 107, 9829. (8) Osakai, T.; Ogata, A.; Ebina, K. J. Phys. Chem. B 1997, 101, 8341. (9) Osakai, T.; Tokura, A.; Ogawa, H.; Hotta, H.; Kawakami, M.; Akasaka, K. Anal. Sci. 2003, 19, 1375. (10) Osakai, T.; Ebina, K. J. Phys. Chem. B 1998, 102, 5691. (11) Osakai, T.; Hoshino, M.; Izumi, M.; Kawakami, M.; Akasaka, K. J. Phys. Chem. B 2000, 104, 12021. (12) Bonner, O. D.; Choi, Y. S. J. Phys. Chem. 1974, 78, 1723.
J. Phys. Chem. B, Vol. 111, No. 25, 2007 7317 (13) Nernst, W. Z. Phys. Chem. 1892, 9, 137. (14) Price, W. S.; Ide, H.; Arata, Y. Division J. Chem. Phys. 2000, 113, 3686. (15) Price, W. S.; Ide, H.; Arata, Y. J. Phys. Chem. A 2003, 107, 4784. (16) Eisenberg, D.; Kauzmann, W. The Structure and Properties of Water; Oxford University Press: Oxford, 1969; Sections 3.5 and 4.7. (17) Socrates, G. Infrared and Raman Characteristic Group Frequencies - Tables and Charts, 3rd ed.; John Wiley and Sons, Ltd.: Chichester, U.K., 2001. (18) Bertie, J. E.; Whalley, E. J. Chem. Phys. 1967, 46, 1271. (19) Rudolph, W.; Brooker, M. H.; Pye, C. C. J. Phys. Chem. 1995, 99, 3793. (20) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, Revision C.02; Gaussian, Inc.: Wallingford, CT, 2004.
