Article pubs.acs.org/JPCA
Concentration-Dependent Dynamics of Hydrogen Bonding between Acetonitrile and Methanol As Determined by 1D Vibrational Spectroscopy Brian G. Alberding and Benjamin J. Lear* Department of Chemistry, Pennsylvania State University, University Park, Pennsylvania 16801, United States S Supporting Information *
ABSTRACT: Solutions of acetonitrile (MeCN) in methanol (MeOH) at various concentrations have been investigated by variable temperature Raman spectroscopy. In the ν(CN) region of the spectrum, the variable temperature spectra at each concentration show two overlapping bands from hydrogen bound and free MeCN. These two species undergo dynamic exchange that gives rise to increasing coalescence of the two bands with increasing temperature. By simulation of the band shape, the rate of exchange was determined at each temperature. Arrhenius plots yielded values for the activation energy, Ea, and the natural log of the preexponential factor, ln[A/s−1], for the hydrogen bond formation/cleavage. Both of these dynamic parameters were found to depend on the relative amounts of MeCN and MeOH in the solutions. In particular, two different concentration regimes of dynamic hydrogen bonding were observed. First, at low MeCN concentration, the dynamics are largely independent of changes in MeCN concentration. Second, at higher MeCN concentration (above ∼0.2 MeCN mole fraction) the dynamics are strongly dependent on further increases of MeCN content. Over the range of MeCN mole fractions that we studied (0.03−0.5), the ln[A/s−1] changes from 32.5 ± 0.1 to 30.1 ± 0.2 and Ea changes from 3.73 ± 0.08 to 2.7 ± 0.1 kcal/mol. We suggest the observed changes in dynamics arise from changes in the local solvent microstructure that occur above a critical mole fraction of MeCN.
■
INTRODUCTION Hydrogen bonding between solvent and solute is an effective means by which to exert influence over the properties of many chemical systems. When the solution phase chemical reactions are considered, the interactions between solute and solvent can affect the overall reaction mechanism and rate.1,2 These interactions can also affect the efficacy of catalysts. For homogeneous catalysts, catalytic cycles often involve the coordination and release of solvent molecules, which occupy sites vacated by the substrate after catalysis and preserve the system for the next molecule.3 In the case of heterogeneous catalysts, the solvation of the reactants and products, near the surface of the catalyst contribute to the overall catalytic cycle.4 Moreover, it has been demonstrated that molecules joined by hydrogen bonding pairs can function as effective mediators of electron transfer and that this electron transfer pathway can be created or destroyed via control over the hydrogen bonding interactions with solvent.5 Finally, hydrogen bonding influences the structure of biologically important molecules6,7 and can also promote self-healing in the structure of polymeric materials by design,8 but these properties can be disrupted via addition of protic solvents. In all of these cases, it is clear that solvent/ solute hydrogen bonding interactions control not only the thermodynamic landscape but also the chemical dynamics by which solvent molecules explore this landscape. For this reason there is a need to understand hydrogen bonding at a fundamental level, using simple solution-based systems. © 2014 American Chemical Society
This need has motivated studies into simple solvent systems,9−15 and from these studies a common theme has emerged: mixtures of solvents and solvent−solute interactions result in local changes to the solvent environment, giving rise to heterogeneously structured microdomains within the system. This is true of even fully miscible solvent systems, and appears to hold for both conventional solvents,16,17 as well as ionic liquids.18 The degree of heterogeneity and the local structures in such mixtures are dependent on the relative concentrations of the components. From this, we can also expect that the dynamics of hydrogen bonding should be sensitive to the concentrations of the components. We are particularly interested in determining how such changes in local microstructure may affect the ground state dynamics of hydrogen bonding. Even though it is often considered a strong intermolecular interaction, the hydrogen bond can be broken with the input of energy on the order of a few kcal/mol or less. Due to this low binding energy, a system involving hydrogen bonding at equilibrium undergoes the dynamic breaking and forming of hydrogen bonds (dynamic exchange between hydrogen bound and free forms) and this process occurs on fast ( 0. This observation is consistent with MeOH being more basic, more strongly electron donating (donor number 23.5), than is MeCN (donor number 14.1).36 The conclusion that unbound MeCN (III) is more stable than H-bound MeCN (I), however, is in apparent conflict with the analysis of Hochstrasser, which reported a Keq > 1 for the reaction that takes III to I and therefore a ΔG < 0 for the reaction.15 However, in calculating Keq, Hochstrasser did not include the concentration of free MeOH, adhering to the 4366
dx.doi.org/10.1021/jp4110147 | J. Phys. Chem. A 2014, 118, 4363−4371
The Journal of Physical Chemistry A
Article
IR spectroscopy also reported this observation.37 A similar shift in ν(CN) for the hydrogen bound species to lower frequency with increased MeCN concentration may be present as well. The peak for hydrogen bound MeCN becomes less welldefined at high MeCN concentration and this may be due to a shift in λmax to lower frequency. However, because λmax is not defined at high MeCN concentration, it is much more difficult to report this shift with certainty. Despite the changes in the intensities and positions of the bands associated with free and hydrogen bound MeCN, there are behaviors in the temperature-dependent spectra that remain constant as the relative MeCN concentration is increased. Specifically, we always observe separate peaks at low temperature, which coalesce at higher temperature. Recalling that this coalescence is assigned to dynamics exchange of these species, we can then simulate these spectra to extract out kinetic parameters associated with this process.
confident that any interference with our simulated results arising from these low and high energy extra features is minimal. The primary reason for this is that the band shape due to exchange is most apparent between the exchanging bands, and this provides a marker that aids in isolating dynamic exchange broadening from effects that provide symmetric broadening of the individual bands. Such symmetric effects are accounted for by our starting assumed band shape, as described above. Furthermore, as it is the change in intensity between bands that holds most of our interest, small features on the wings of the bands should have minimal impact on our extracted rates. In total, we expect the features on the outside of the band will impact the simulations of our spectra to a minimal extent, and we can still expect to have confidence in our determined rates of MeCN exchange between species III and I. The results of the simulations yield an extracted rate on the picosecond time scale, which fits the requirement for observing dynamic effects in vibrational spectra. Furthermore, at each concentration, the determined rate was found to increase with increasing temperature. These results obtained from the simulations are summarized as Arrhenius plots in Figure 3.
■
SIMULATION RESULTS The spectra shown in Figure 2 were simulated according to the theory described in the Experimental Details, and the simulation results are represented as the solid lines. In general, the simulations match well with the observed band shape but there are some small differences that have been displayed in Figure S1 of the Supporting Information for each of the studied concentrations. The most notable difference occurs on the high frequency edge of ν(CN) for the H-bound species. Similar deviations on the high energy side of other dynamically exchanging systems have also been observed, though no explanation has been offered for this extra intensity.38 In addition, we also observe a small feature on the low energy side of the coalescing bands. Related studies of the H-bonding exchange in water were determined to contain contributions from vibrational excitation transfer which obscured the dynamics of interest. These complications were most notable in polarization-dependent IR pump−IR probe experiments where vibrational energy transfer between adjacent oscillators causes an additional depolarization process.12,39 It is possible that vibrational energy transfer occurring on the picosecond time scale between adjacent MeCN could cause an additional component to the observed Raman features or line broadening, especially at high MeCN concentrations. For instance in our Raman experiment it is also possible that, within an aggregate of multiple identical (or nearly identical) CN oscillators, a single excitation could involve multiple oscillators. Such coupling between the oscillators would lead to broadening, or even splitting, of the vibrational bands associated with MeCN. This behavior could also be responsible for the above “extra” features. Both of the above broadening mechanisms could explain the extraneous intensity that we observe at the high and low energy wings of the bands. Moreover, as both of these effects are expected to be concentration dependent, they could be invoked to explain the new behavior (vide inf ra) of this system at high concentrations. However, we note that the “extra” intensity at the wings of the bands decrease in intensity with increasing mole fraction of MeCN, which we would not expect if the above mechanisms were the dominant source of the “additional” features. For this reason, we do not believe that having multiple chromophores in close proximity is giving rise to complications that dominate the dynamic spectroscopic features of the MeOH/MeCN systems. No matter the origin of the “extra” features in our spectra, the fact remains that they are present. However, we remain
Figure 3. Arrhenius plots at each acetonitrile concentrations, [MeCN] (acetonitrile mole fraction, χMeCN) of the hydrogen bonding exchange rate obtained from the spectral fitting program at each temperature from the Raman data: (black) 0.75 M (0.031), (red) 1.5 M (0.062), (green) 4.7 M (0.20), (blue) 7.5 M (0.33), and (cyan) 11 M (0.50).
For each point in Figure 3, the error in the rate is derived from the minimum change in the rate value that causes an observable change in the simulated spectra. For example, at −24.8 °C in the 0.75 M concentration sample, the lifetime value that gave the best fit to the experimental spectra was 16.5 ps and this value needed to be changed in either direction by a minimum of ∼0.4 ps to produce a noticeable change in the similarity between the simulated and experimental spectra. In terms of the Arrhenius plots shown in Figure 3, a lifetime value of 16.5 ± 0.4 ps translates to ln[rate (s−1)] = 24.83 ± 0.02. This error magnitude was consistent across all of the other temperatures and concentrations. The Arrhenius plots of the rates determined in this way each display a linear trend with an R2 value greater than 0.98. When the Arrhenius plots were fit to a linear equation, Ea and ln[A/s−1] for the hydrogen bonding exchange were obtained as the slope and y-intercept, respectively, at each concentration. A plot of these values versus concentration is displayed in Figure 4. Both parameters follow the same general trend of displaying two distinct regions of concentration dependence. In the low concentration region (MeCN mole fraction ∼0−0.2), the values for both Ea and ln[A/s−1] are 4367
dx.doi.org/10.1021/jp4110147 | J. Phys. Chem. A 2014, 118, 4363−4371
The Journal of Physical Chemistry A
Article
likelihood of encountering a MeOH molecule decreases, and so will ln[A/s−1]. Both of the above analyses are based upon the recognition that approaches to the reaction barrier require MeCN to encounter a MeOH molecule, and this perspective can also be invoked to explain the existence of the two limiting regions of dependence on the concentration of MeCN. As the amount of MeCN is increased initially, the system should still consist of only individual MeCN molecules dispersed relatively evenly throughout MeOH solvent and there should be no change in the hydrogen bonding dynamics experienced by the individual MeCN solute molecules because the system still consists of individual MeCN within pockets of MeOH. However, after a certain concentration of MeCN is reached, it becomes likely that an individual MeCN molecule will encounter another MeCN molecule instead of a MeOH molecule. That is, a heterostructure emerges in which MeOH surrounds groups of MeCN. This also means that it is less likely that two MeOH molecules will form II. Effectively, the local concentration of MeOH surrounding individual MeCN molecules becomes reduced. In this concentration regime, it will take progressively longer amounts of time for MeCN molecules to explore their environments and approach the hydrogen bonding reaction barrier. ln[A/s−1], therefore, becomes smaller with increasing amounts of MeCN, but only after a certain MeCN concentration is reached. The results presented here suggest that this concentration limit lies above MeCN mole fraction of ∼0.2 (4.7 M MeCN in MeOH) and thereafter ln[A/s−1] begins to decrease at greater MeCN mole fraction. As we will now see, similar considerations can be used to explain the behavior of Ea as a function of MeCN mole fraction. Discussion of the Concentration Dependence of Ea. For the trend in Ea, we must consider hydrogen bonding between MeOH molecules within the bulk solvent, and it will be easiest to consider the hydrogen bonding at the two extreme limits of low and high MeCN concentration. Neat MeOH, which has been studied experimentally and theoretically, consists of extended hydrogen bonded structures either as adjacent linear chains or loops that contain two hydrogen bonds per MeOH molecule.40−42 From the perspective of an individual MeOH molecule, it participates in one hydrogen donating and one hydrogen accepting hydrogen bond. These hydrogen bonded MeOH structures have been calculated to have binding energies on the order of 10 kcal/mol per MeOH molecule.40 When a small amount of MeCN is added, the system still consists of these hydrogen bonded MeOH structures but they also form a solvent shell around the MeCN at equilibrium. The hydrogen bonding dynamics at low concentration involve MeCN molecules navigating the space within a fully self-hydrogen bonded MeOH solvent shell. For MeCN to form a hydrogen bond with MeOH, it must overcome the binding energy in one of the MeOH hydrogen bonds, and this contributes to the overall activation energy of the MeCN hydrogen bonding dynamics. Taking advantage of the cooperative transition state, II, MeCN can overcome the MeOH−HOMe binding energy and proceed to the hydrogen bound species, I. This idea can also be represented as a double well potential energy surface along the hydrogen bonding coordinate and is shown in the black curve in Figure 5. In dilute MeCN solution, there is an energy difference (ΔG > 0) between species III and I along the MeCN hydrogen bonding coordinate because the MeOH−HOMe hydrogen bond is stronger than the MeCN−HOMe hydrogen bond.
Figure 4. Summary of the activation energy, Ea (black, left axis), and the natural log of the pre-exponential factor, ln[A/s−1] (red, right axis), obtained at each concentration from the Arrhenius plots. Trend lines are shown as dotted lines.
largely independent of concentration. At higher concentrations, Ea and ln[A/s−1] were both found to depend strongly on concentration, with both parameters decreasing as the concentration of MeCN was increased. These trends are considered next in more detail. Discussion of Concentration Dependence of ln[A/s−1]. To understand the observed trend in ln[A/s−1], we must first remember that the pre-exponential factor is the rate at which the system approaches the barrier that separates free and hydrogen bound MeCN in MeOH solution. When the rate at which the system approaches the barrier from the hydrogen bound side (I → II) is considered, the rate should depend on MeCN concentration (or equivalently on the MeOH concentration). If II is accepted as the transition state in solutions with dilute MeCN content, then the rate at which this transition geometry is attained depends upon the frequency of encounters between the hydrogen bound MeCN−HOMe complex and another MeOH molecule. An alternative way to view this mechanism is the transfer of a hydrogen bond from the MeCN to another MeOH, which also depends on one MeOH finding the hydrogen bound complex. In both perspectives, with decreasing MeOH concentration (or increasing MeCN concentration), the likelihood of two MeOH molecules encountering each other should decrease and ln[A/s−1] should correspondingly decrease. Of course, the MeCN−HOMe hydrogen bond could also be broken in the absence of a second MeOH molecule, but in that situation, the Ea for the exchange process will change. The implications of this are explored in the next section. Conversely, approaching the barrier from the free MeCN side (III → II) requires that an MeCN finds a MeOH molecule; thus this reflects a representation of how quickly (or the rate at which) an MeCN molecule can explore its local environment and find an available MeOH hydrogen atom to form a hydrogen bond. In the limit of very small MeCN concentration, the solution will consist of individual MeCN molecules within a shell of MeOH molecules. It is reasonable to expect that a MeCN molecule in this situation does not explore a very large space before it encounters a MeOH hydrogen atom for hydrogen bonding. In other words, the system can proceed along the hydrogen bonding coordinate very quickly to exchange between the free and hydrogen bound forms. However, as the mole fraction of MeCN increases, the 4368
dx.doi.org/10.1021/jp4110147 | J. Phys. Chem. A 2014, 118, 4363−4371
The Journal of Physical Chemistry A
Article
In this case, the latter species, III, is lower in energy as a result of the stronger MeOH−HOMe hydrogen bond as compared to the MeCN−HOMe bond. In both cases the system has one state with a MeCN−HOMe hydrogen bond. If one normalizes the system with respect to the energy of the MeCN−HOMe species, then it becomes apparent that the effect of increasing MeCN concentration is to take an asymmetric PES and bring up the lower energy state until the minima of the PES wells are isoenergetic. These considerations are depicted in Figure 5, where the energy of potential well IV has been set equal to that for I. This figure clearly demonstrates that the shift toward isoenegetic dynamics of hydrogen bond exchange must decrease the barrier, which is consistent with our observations and the physical picture described above. One last concentration regime to consider is the intermediate case. In this concentration regime a transition point has been surpassed and just enough MeCN has been added such that the system cannot be viewed as isolated MeCN molecules within a MeOH shell. Because MeCN and MeOH are miscible liquids, in a solution above the transition MeCN concentration, adequate MeCN is present to begin to disrupt the MeOH hydrogen bonding network. This idea has been suggested before, and molecular dynamics simulations have shown that increasing MeCN content in solvent mixtures with MeOH reduces the internal hydrogen bonding in MeOH and makes the MeOH hydrogen atom more available to hydrogen bond with other solutes.43 At these intermediate concentrations, the overall system will at any one moment contain contributions from various local environments. In some local environments, extended MeOH hydrogen bonding structures will exist and present a high barrier for MeCN−HOMe dynamic exchange whereas in other local environments excess MeCN will be present and the barrier to dynamic exchange will be lower. Such a system can be thought of as a composite of the two limiting conditions described above. As the MeCN content increases, there will be a greater contribution from the excess MeCN limit and the Ea decreases until the system can be described by the limit of individual MeOH molecules within a solvent shell of MeCN. At the same time, ln[A/s−1] will also decrease due to the inability to quickly find the transition state. Thus, the dynamics of the system should be expected to possess at least three regimes: (i) isolated MeCN in MeOH, (ii) isolated MeOH in MeCN, and (iii) cases in which there is microdomains of various MeCN and MeOH concentration. In the limiting cases (i) and (ii), we would expect to observe no dependence of the dynamics on MeCN concentration. However, in the intermediate case, we do expect changes to the dynamics, with changes in MeCN concentration. Moreover, due to changes in local structure of the solvent, we expect changes to both ln[A/s−1] and Ea. On the basis of our observations, we suggest that this transition between (i) and (iii) lies near a mole fraction of MeCN equal to 0.2.
Figure 5. Ground state potential energy surfaces (PESs) associated with a single acetonitrile undergoing hydrogen bond breaking and formation with methanol. These are representative of the dilute (black) and concentrated (red) solutions of acetonitrile in methanol. For clarity in comparing the barriers, the energy of the states containing acetonitrile-methanol hydrogen bounded complexes have been set as equal.
At the high MeCN concentration limit, the system can be viewed as bulk MeCN surrounding individual MeOH molecules. In this concentration limit, the dynamic H-bonding reaction between MeCN and MeOH can be depicted by the reactant, transition state, and product structures depicted in IV, V, and VI, respectively (Chart 2). This has two major Chart 2. Reactant (IV), Transition State (V), and Product (VI) Structures Involved in Dynamic H-Bonding in an Acetonitrile/Methanol Solution with High Acetonitrile Concentration
ramifications. First, bulk MeCN only contains dipole−dipole interactions and these interactions are much weaker than the internal hydrogen bonding present in methanol. Therefore, the system should prefer to have the individual MeOH molecules hydrogen bound to MeCN because the MeCN−HOMe hydrogen bond is stronger than the weak dipole interactions in bulk MeCN and MeOH is unable locate other MeOH molecules as hydrogen bonding partners. Thus, we approximate that each MeOH is always hydrogen bound to at least one MeCN molecule. Second, the transition state should now possess two MeCN molecules with a single MeOH molecule as depicted in structure V. Attainment of the transition state no longer requires breaking a MeOH−HOMe hydrogen bond and results in a comparatively lower Ea for solutions with high MeCN concentrations. Another way to picture the dependence of Ea upon MeCN concentration is to consider the symmetry of the associated PESs. At high MeCN concentration, the breaking of an MeCN−HOMe bond results in the formation of another identical hydrogen bonded pair as depicted by structures IV and VI. The overall reaction at this limit is isoenergetic. On the other hand, at low MeCN concentration, the breaking of the MeCN−HOMe bond to yield the observed unbound ν(CN) results in the formation of a MeOH−HOMe hydrogen bond.
■
CONCLUSIONS Dynamic exchange of hydrogen bond formation/breaking in solutions of MeCN in MeOH of varying concentration has been investigated using Raman spectroscopy. The dynamics were found to occur on the picosecond time scale, and band shape simulation allowed Ea and ln[A/s−1] to be determined. At low MeCN concentration (below below mole fraction ∼0.2), Ea and ln[A/s−1] were independent of concentration changes and on average their values were determined to be Ea = 3.73 ± 0.08 kcal/mol and ln[A/s−1] = 32.5 ± 0.1. Above an MeCN mole 4369
dx.doi.org/10.1021/jp4110147 | J. Phys. Chem. A 2014, 118, 4363−4371
The Journal of Physical Chemistry A
Article
Notes
fraction of approximately 0.2, these values were found to decrease significantly to Ea = 2.7 ± 0.1 kcal/mol and ln[A/s−1] = 30.1 ± 0.2 in the sample of highest MeCN concentration that was studied (which corresponds to an MeCN mole fraction of 0.5). Investigation of samples with higher MeCN concentration were prohibited due to loss of signal for the hydrogen bound species. An analysis of the dependence in these parameters on MeCN concentration suggested that the dynamic exchange occurs by a different mechanism within different concentration regimes. At low concentration, the results suggest the exchange proceeds through a 2MeOH−NCMe cooperative transition state, II, as suggested by Hochstrasser.14,15 Conversely, with increased MeCN content after the saturation point is reached, the system evolves toward the high MeCN limit where the hydrogen bonding dynamics proceed through a 2MeCN−HOMe transition state, V, which does not involve competition with hydrogen bonding in the bulk MeOH solvent. This transition in mechanism leads to a decrease in both ln[A/s−1] and Ea. Notably, we do not observe a smooth transition between these two regimes with increasing MeCN concentration. Rather, we observe an initial insensitivity of ln[A/s−1] and Ea to changes in MeCN concentration, followed by a regime of strong dependence on MeCN concentration. We ascribe this to a transition in local solvent structure, in which we move from a system consisting largely of individual MeCN in MeOH to a system where there exists microdomains of MeCN and MeOH. Finally, this study highlights that changes to molar ratio can have dramatic effects on a system’s local environment/ structure, and that these changes can (in turn) impact the system’s molecular dynamics. It also demonstrates that linear spectroscopies, such as Raman spectroscopy, can be used to determine and evaluate these types of properties in simple systems. Furthermore, this idea could possibly be extended to more complex systems that depend on hydrogen bonding for their structural and eventual functional properties. For example, the performance of a catalyst could be evaluated under different solvent conditions on the basis of changes in the resulting molecular dynamics. Likewise, the structure−property relationship in biological or polymeric macromolecules could be influenced by local solvent interactions and this could be evaluated by a study of the molecular dynamics using linear spectroscopy. Overall, linear spectroscopy is a technique that contains information about not only the identity of certain molecules through the presence of certain peaks but also the dynamics of those molecules through the temperature-dependent band shapes. A much more widespread use of the band shape information onfers the opportunity to gain additional insights from linear spectroscopic techniques used in routine studies.
■
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We acknowledge that this work was funded by and carried out at The Pennsylvania State University.
■
ABBREVIATIONS MeCN, acetonitrile; MeOH, methanol; PESs, potential energy surfaces; Ea, activation energy; Ln[A/s−1], the natural log of the pre-exponential factor
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information contains IR and Raman spectra showing the differences between the experimental data and simulated spectra at each concentration, Figure S1a−d. This material is available free of charge via the Internet at http:// pubs.acs.org.
■
REFERENCES
(1) Reichardt, C. Solvents and Solvent Effects in Organic Chemistry; Wiley-VCH: Weinheim, 2003. (2) Cabot, R.; Hunter, C. A. Molecular Probes of Solvation Phenomena. Chem. Soc. Rev. 2012, 41, 3485−3492. (3) Uyanik, M.; Yasui, T.; Ishihara, K. Hydrogen Bonding and Alcohol Effects in Asymmetric Hypervalent Iodine Catalysis: Enantioselective Oxidative Dearomatization of Phenols. Angew. Chem., Int. Ed. 2013, 52, 9215−9218. (4) Mukherjee, S.; Vannice, M. A. Solvent Effects in Liquid-Phase Reactions: I. Activity and Selectivity during Citral Hydrogenation on Pt/SiO2 and Evaluation of Mass Transfer Effects. J. Catal. 2006, 243, 108−130. (5) Goeltz, J. C.; Kubiak, C. P. Mixed Valency across Hydrogen Bonds. J. Am. Chem. Soc. 2010, 132, 17390−17392. (6) Takahashi, O.; Kohno, Y.; Nishio, M. Relevance of Weak Hydrogen Bonds in the Conformation of Organic Compounds and Bioconjugates: Evidence from Recent Experimental Data and HighLevel Ab Initio MO Calculations. Chem. Rev. 2010, 110, 6049−6076. (7) Scheiner, S. Weak H-Bonds. Comparisons of CHO to NHO in Proteins and PHN to Direct PN Interactions. Phys. Chem. Chem. Phys. 2011, 13, 13860−13872. (8) Chen, Y.; Guan, Z. Self-Assembly of Core−shell Nanoparticles for Self-Healing Materials. Polym. Chem. 2013, 4, 4885−4889. (9) Steinel, T.; Asbury, J. B.; Zheng, J.; Fayer, M. D. Watching Hydrogen Bonds Break: A Transient Absorption Study of Water. J. Phys. Chem. A 2004, 108, 10957−10964. (10) Asbury, J. B.; Steinel, T.; Fayer, M. D. Hydrogen Bond Networks: Structure and Evolution after Hydrogen Bond Breaking. J. Phys. Chem. B 2004, 108, 6544−6554. (11) Steinel, T.; Asbury, J. B.; Corcelli, S.; Lawrence, C.; Skinner, J.; Fayer, M. Water Dynamics: Dependence on Local Structure Probed with Vibrational Echo Correlation Spectroscopy. Chem. Phys. Lett. 2004, 386, 295−300. (12) Park, S.; Fayer, M. D. Hydrogen Bond Dynamics in Aqueous NaBr Solutions. Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 16731− 16738. (13) Singh, P. K.; Kuroda, D. G.; Hochstrasser, R. M. An Ion’s Perspective on the Molecular Motions of Nanoconfined Water: A Two-Dimensional Infrared Spectroscopy Study. J. Phys. Chem. B 2013, 117, 9775−9784. (14) Kim, Y. S.; Hochstrasser, R. M. Comparison of Linear and 2D IR Spectra in the Presence of Fast Exchange. J. Phys. Chem. B 2006, 110, 8531−8534. (15) Kim, Y. S.; Hochstrasser, R. M. Chemical Exchange 2D IR of Hydrogen-Bond Making and Breaking. Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 11185−11190. (16) Venables, D. S.; Schmuttenmaer, C. A. Spectroscopy and Dynamics of Mixtures of Water with Acetone, Acetonitrile, and Methanol. J. Chem. Phys. 2000, 113, 11222−11236. (17) Farwaneh, S. S.; Yarwood, J.; Cabaço, I.; Besnard, M. Infrared Studies of Hydrogen-Bonding of Methanol in Binary Mixtures with Acetonitrile. J. Mol. Liq. 1993, 56, 317−332. (18) Scharf, N. T.; Stark, A.; Hoffmann, M. M. Ion Pairing and Dynamics of the Ionic Liquid 1-Hexyl-3-Methylimidazolium Bis-
AUTHOR INFORMATION
Corresponding Author
*B. J. Lear: tel, 814-867-4625; fax, 814-865-3258; e-mail,
[email protected]. 4370
dx.doi.org/10.1021/jp4110147 | J. Phys. Chem. A 2014, 118, 4363−4371
The Journal of Physical Chemistry A
Article
(irifluoromethylsulfonyl)amide ([C6mim][NTf2]) in the Low Dielectric Solvent Chloroform. J. Phys. Chem. B 2012, 116, 11488−11497. (19) Zheng, J.; Kwak, K.; Fayer, M. D. Ultrafast 2D IR Vibrational Echo Spectroscopy. Acc. Chem. Res. 2007, 40, 75−83. (20) Gutowsky, H. S.; Holm, C. H. Rate Processes and Nuclear Magnetic Resonance Spectra. II. Hindered Internal Rotation of Amides. J. Chem. Phys. 1956, 25, 1228. (21) Kattnig, D. R.; Mladenova, B.; Grampp, G.; Kaiser, C.; Heckmann, A.; Lambert, C. Electron Paramagnetic Resonance Spectroscopy of Bis(triarylamine) Paracyclophanes as Model Compounds for the Intermolecular Charge-Transfer in Solid State Materials for Optoelectronic Applications. J. Phys. Chem. C 2009, 113, 2983−2995. (22) Nelsen, S. F.; Ismagilov, R. F.; Powell, D. R. Charge-Localized P-Phenylenedihydrazine Radical Cations: ESR and Optical Studies of Intramolecular Electron Transfer Rates. J. Am. Chem. Soc. 1997, 119, 10213−10222. (23) Turner, J. J. Handbook of Vibrational Spectroscopy; Wiley: New York, 2002. (24) Grevels, F.-W.; Kerpen, K.; Klotzbücher, W. E.; McClung, R. E. D.; Russell, G.; Viotte, M.; Schaffner, K. The Very Low Barrier of CO Site Exchange in Tricarbonyl(η4-1,5-Cyclooctadiene)iron: Picosecond Kinetics in Solution Investigated by Line Shape Simulation of the ν(CO) IR Bands and Complementary Evidence from the Course of 13 CO Incorporation in a Low-Temperature Matrix. J. Am. Chem. Soc. 1998, 120, 10423−10433. (25) Turner, J. J.; Gordon, C. M.; Howdle, S. M. Infrared Spectral Features Due to Very Rapid Fluxional Motion: Changes in the Infrared Carbonyl Stretching Spectra of Tricarbonyl(.eta.4-Norbornadiene)iron with Temperature. J. Phys. Chem. 1995, 99, 17532−17538. (26) Giordano, A. N.; Morton, S. M.; Jensen, L.; Lear, B. J. Direct Test of the Equivalency of Dynamic IR and Dynamic Raman Spectroscopies As Techniques for Observing Ultrafast Molecular Dynamics. J. Phys. Chem. A 2013, 117, 2067−2074. (27) Zheng, J.; Kwak, K.; Xie, J.; Fayer, M. D. Ultrafast CarbonCarbon Single-Bond Rotational Isomerization in Room-Temperature Solution. Science 2006, 313, 1951−1955. (28) Levinger, N. E.; Davis, P. H.; Behera, P. K.; Myers, D. J.; Stromberg, C.; Fayer, M. D. Isomerization and Intermolecular Solute− solvent Interactions of Ethyl Isocyanate: Ultrafast Infrared Vibrational Echoes and Linear Vibrational Spectroscopy. J. Chem. Phys. 2003, 118, 1312. (29) Ito, T.; Hamaguchi, T.; Nagino, H.; Yamaguchi, T.; Washington, J.; Kubiak, C. P. Effects of Rapid Intramolecular Electron Transfer on Vibrational Spectra. Science 1997, 277, 660−663. (30) Lear, B. J.; Glover, S. D.; Salsman, J. C.; Londergan, C. H.; Kubiak, C. P. Solvent Dynamical Control of Ultrafast Ground State Electron Transfer: Implications for Class II−III Mixed Valency. J. Am. Chem. Soc. 2007, 129, 12772−12779. (31) Watanabe, D.; Hamaguchi, H. Ultrafast Protonation/Deprotonation Dynamics of N,N-Dimethylacetamide in Hydrochloric Acid As Studied by Raman Band Shape Analysis. J. Phys. Chem. C 2009, 113, 11662−11666. (32) Bratos, S.; Tarjus, G.; Viot, P. Isotropic Raman Study of Ultrarapid Proton - transfer Reactions in Aqueous Mixtures. J. Chem. Phys. 1986, 85, 803. (33) Zheng, J.; Kwac, K.; Asbury, J.; Chen, X.; Piletic, I. R.; Fayer, M. D. Ultrafast Dynamics of Solute-Solvent Complexation Observed at Thermal Equilibrium in Real Time. Science 2005, 309, 1338−1343. (34) RAPID. Https://github.com/jensengrouppsu/rapid. (35) Eaton, G.; Pena-Nuñez, A. S.; Symons, M. C. R.; Ferrario, M.; McDonald, I. R. Spectroscopic and Molecular Dynamics Studies of Solvation of Cyanomethane and Cyanide Ions. Faraday Discuss. Chem. Soc. 1988, 85, 237−253. (36) Meade, M.; Hickey, K.; McCarthy, Y.; Waghorne, W. E.; Symons, M. R.; Rastogi, P. P. Solvation of Solutes with SO or CO Groups in Themethanol−acetonitrile Solvent System: A Combined Calorimetricand Spectroscopic Study. J. Chem. Soc., Faraday Trans. 1997, 93, 563−568.
(37) Eaton, G.; Pena-Nuñez, A. S.; Symons, M. C. R. Solvation of Cyanoalkanes [CH3CN and (CH3)3CCN]. An Infrared and Nuclear Magnetic Resonance Study. J. Chem. Soc., Faraday Trans. 1 1988, 84, 2181−2193. (38) Salsman, J. C.; Ronco, S.; Londergan, C. H.; Kubiak, C. P. Tuning the Electronic Communication and Rates of Intramolecular Electron Transfer of Dimers of Trinuclear Ruthenium Clusters: Bridging and Ancillary Ligand Effects. Inorg. Chem. 2006, 45, 547− 554. (39) Gaffney, K. J.; Piletic, I. R.; Fayer, M. D. Orientational Relaxation and Vibrational Excitation Transfer in Methanol−carbon Tetrachloride Solutions. J. Chem. Phys. 2003, 118, 2270−2278. (40) Kazachenko, S.; Bulusu, S.; Thakkar, A. J. Methanol Clusters (CH3OH)n: Putative Global Minimum-Energy Structures from Model Potentials and Dispersion-Corrected Density Functional Theory. J. Chem. Phys. 2013, 138, 224303. (41) Buck, U.; Huisken, F. Infrared Spectroscopy of Size-Selected Water and Methanol Clusters. Chem. Rev. 2000, 100, 3863−3890. (42) Jorgensen, W. L. Quantum and Statistical Mechanical Studies of Liquids. 7. Structure and Properties of Liquid Methanol. J. Am. Chem. Soc. 1980, 102, 543−549. (43) Kwac, K.; Geva, E. Solvation Dynamics of Formylperylene Dissolved in Methanol−Acetonitrile Liquid Mixtures: A Molecular Dynamics Study. J. Phys. Chem. B 2013, 117, 9996−10006.
4371
dx.doi.org/10.1021/jp4110147 | J. Phys. Chem. A 2014, 118, 4363−4371
