ARTICLE pubs.acs.org/JPCA
Structure and Dynamics of Water Dangling OH Bonds in Hydrophobic Hydration Shells. Comparison of Simulation and Experiment Jill Tomlinson-Phillips, Joel Davis, and Dor Ben-Amotz* Department of Chemistry, Purdue University, West Lafayette, Indiana 47907, United States Daniel Spangberg, Ljupco Pejov,† and Kersti Hermansson*
Department of Materials Chemistry, Uppsala University, Uppsala, Sweden ABSTRACT: Molecular dynamics and electric field strength simulations are performed in order to quantify the structural, dynamic, and vibrational properties of non-H-bonded (dangling) OH groups in the hydration shell of neopentane, as well as in bulk water. The results are found to be in good agreement with the experimentally observed highfrequency (∼3660 cm-1) OH band arising from the hydration shell of neopentanol dissolved in HOD/D2O, obtained by analyzing variable concentration Raman spectra using multivariate curve resolution (Raman-MCR). The simulation results further indicate that hydration shell dangling OH groups preferentially point toward the central carbon atom of neopentane to a degree that increases with the lifetime of the dangling OH.
I. INTRODUCTION The structural and dynamic consequences of the interaction between water and dissolved molecules are subjects of both biological and geological relevance, as well as of fundamental interest. The present work is focused on the hydration of nonpolar molecules and, in particular, the formation of dangling water OH bonds in the hydration shell around alkane solutes. Recent experiments have uncovered evidence of such dangling OH bonds in the hydration shells around alkane groups of various sizes and shapes.1 Here, we report the results of molecular dynamics and electric field distribution simulations of neopentane (CH3)4C in liquid water, as well as new experimental measurements of neopentanol in HOD/D2O. These studies are the first to directly link the spectroscopic and structural features of dangling water OH groups in hydrophobic hydration shells (as opposed to macroscopic hydrophobic interfaces).2 The results reveal that the hydration shell of neopentane has a structure that differs from both a clathrate hydrate and a macroscopic oil-water interface. Neopentane was selected for the hydration simulations both because of its high symmetry and because it is a prototypical large hydrophobic solute.3 The experiments were performed using the similar neopentanol solute because neopentane is too insoluble. Isotopically dilute HOD/D2O water was used to suppress OH resonance coupling (coherent energy transfer) contributions to the observed spectra4 (which are not present in the simulations). Recent experimental evidence of the enhanced formation of dangling OH groups at molecular hydrophobic interfaces has been obtained by combining variable-concentration Raman spectral measurements with multivariate curve resolution (RamanMCR).5 Such measurements reveal a narrow high-frequency OH stretch feature arising from the hydration shells around nonpolar (alkane) chains in dilute aqueous solutions of various alcohols.1 r 2011 American Chemical Society
The frequency of the observed OH band (3661 ( 3 cm-1) is similar to that of dangling OH groups observed at macroscopic oil-water interfaces,6 and its width (∼50 cm-1) implies that the dangling OH groups must persist for at least 0.2 ps (and perhaps much longer if configurational, inhomogeneous, broadening contributes significantly to the observed bandwidth).1 Moreover, the intensity of the observed band implies that there is approximately 1 long-lived dangling OH in 10 neopentanol hydration shells.1 Here, we report new Raman-MCR measurements of neopentanol in HOD/D2O to demonstrate that the dangling OH feature is present in isotopically dilute water (in the absence of OH resonance coupling), as well as to quantitatively compare the frequency, width, and intensity of the observed OH band with molecular dynamics and electric field strength simulations of neopentane in H2O. Although the number of intact and broken H-bonds in liquid water remains a controversial issue,7-11 there is little doubt that liquid water contains a wide distribution of nearest-neighbor O 3 3 3 O distances and O-H 3 3 3 O angles. Thus, any attempt to quantify the structure and dynamics of water in terms of distinct broken and formed H-bond populations is necessarily approximate and depends on how one chooses to identify the two populations.12 Moreover, regardless of precisely how one chooses to define a H-bond, both experimental (2D-IR) and simulation results suggest that most H-bonds in bulk water are only broken fleetingly13 and may be viewed as unstable transition states between H-bonded free-energy minimum structures.14 Special Issue: Victoria Buch Memorial Received: November 29, 2010 Revised: February 15, 2011 Published: March 17, 2011 6177
dx.doi.org/10.1021/jp111346s | J. Phys. Chem. A 2011, 115, 6177–6183
The Journal of Physical Chemistry A
ARTICLE
Figure 1. (A) Simulation snapshot of neopentane in liquid water, showing a long-lived water dangling OH bond. (B) Radial distribution function of the water O atoms relative the central C atom of neopentane. The first hydration shell extends to approximately the first minimum in the latter g(r), near 6.5 Å.
The enhanced formation of dangling water OH bonds around small hydrophobic groups contrasts with the traditional “iceberg” (or clathrate hydrate) model of hydrophobic hydration, which implies that water molecules reorganize around hydrophobic solutes so as to minimize the number of broken water H-bonds.15,16 Both simulations17 and neutron scattering18 experiments imply that the structure of water in hydrophobic hydration shells more closely resembles bulk water than solid clathrate hydrates. However, recent experimental19-25 and theoretical26-28 results also indicate that water H-bond exchange and reorientational dynamics are slowed in the hydration shells of small nonpolar, polar, and ionic solutes, as well as at protein surfaces.22,23,25-28 On the other hand, macroscopic air-water29 and oil-water6 interfaces contain a significant number of water dangling OH bonds (at an air-water interface, about one-fifth of the water molecules contain a dangling OH group),30,31 while the associated water dynamics is remarkably similar to that of bulk water.32 Our results go beyond the latter studies by quantifying the influence of hydrophobic solutes on the structure, dynamics, and vibrational spectra of hydration shell dangling water OH groups. The remainder of this paper is organized as follows. Our experimental and simulation methods are described in section II. The resulting structural, dynamic, and spectroscopic properties of dangling OH bonds in the neopentane hydration shell, as well as comparison with dangling OH bonds in bulk water, are described in section III. The conclusions are summarized and discussed in section IV.
II. EXPERIMENTAL AND SIMULATION METHODS The experimental (Raman-MCR) detection of dangling water OH bonds in the hydration shell of neopentanol was performed using the same procedure and conditions as those previously employed to measure the hydration shell spectra of other linear and branched alcohols.1 However, the present studies utilized isotopically dilute water containing ∼10% HOD in D2O as the solvent (obtained by mixing 5% v/v of H2O in D2O) in order to suppress OH resonance coupling contributions to the hydration shell spectra. The input Raman spectra were obtained from six solutions of neopentanol in HOD/D2O, ranging in concentration from 0 to 5 wt %. Molecular dynamics (MD) simulations of pure water and an aqueous neopentane solution were performed using GROMACS
(v4.0.7).33-36 The simulation runs covered 1 ns with 1 fs time steps and were sampled every 25 fs in order to analyze the dangling and H-bonded OH populations. Each simulation system contained 500 TIP4P water molecules.37 The aqueous solution additionally contained one neopentane molecule, which was modeled using the OPLS-aa potential.38 Electrostatic interactions were treated with the particle mesh Ewald (PME) method, and Lennard-Jones interactions were treated with a cutoff of 0.9 nm. A velocity-rescale Berendsen thermostat and Berendsen barostat were used to maintain the system at 300 K and 1 atm. Two quite different H-bond definitions were applied in order to evaluate the sensitivity of the results to the manner in which H-bonds are defined. One definition, suggested by Luzar and Chandler,39 identifies H-bonds as having an O 3 3 3 O distance (R) less than 3.5 Å and a OH 3 3 3 O angle (θ) less than 30. A second definition, proposed by Wernet et al.,7 requires that the bound on R (Å) be a quadratic function of θ (degrees), R(θ) e -0.00044θ2 þ 3.3. Although the two definitions naturally influence the predicted concentrations of dangling H-bonds, quite similar results and general conclusions are obtained using both the Luzar and Wernet H-bond definitions (as described in section III). A dynamic analysis is performed to determine the persistence time of dangling OH bonds by counting the consecutive snapshots over which an OH group that is initially identified as nonH-bonded persists. More specifically, the time correlation function of dangling OH bonds is defined as follows CðtÞ ¼
ÆBðtÞBð0Þæ ¼ ÆBðtÞæ ÆBð0Þ2 æ
ð1Þ
where B(t) = 1 when a particular OH group is identified as dangling while B(t) = 0 when it is identified as a H-bond donor. The time average Æ...æ is performed either over all of the OH groups within the first hydration shell of neopentane (i.e., within a 6.5 Å radius of the center of mass of neopentane, near the first minimum in the water-neopentane radial distribution function, as shown in Figure 1B) or over all of the OH groups in a bulk water simulation (at the same temperature and pressure). The second equality in eq 1 follows from the fact that the time correlation function is performed on OH groups which are initially found to be dangling, and therefore, B(0) = 1. Thus, ÆB(t)æ represents the conditional probability that a OH that was 6178
dx.doi.org/10.1021/jp111346s |J. Phys. Chem. A 2011, 115, 6177–6183
The Journal of Physical Chemistry A
ARTICLE
Table 1. Classification of Water H-Bond Configurations and Abundances H-bond definition, Luzarb
H-bond definition, Wernetb
H-bonding
Number of
Number of
Pure
Hydration shell AB
Pure
Hydration shell AB
designationa
donors
acceptors
water
around neopentane
water
around neopentane
NDNA
0
0
0.05
0.05
0.20
0.23
NDSA
0
1
0.67
0.85
2.02
2.13
NDDA
0
2
0.81
0.96
1.79
1.92
NDTA
0
3þ
0.05
0.04
0.05
0.04
SDNA
1
0
0.41
0.47
SDSA
1
1
7.15
8.3
13.2
14.5
SDDA SDTA
1 1
2 3þ
12.2 1.06
13.4 0.90
16.3 0.73
16.7 0.56
DDNA
2
0
DDSA
2
1
20.3
20.8
22.3
22.8
DDDA
2
2
49.2
47.9
38.2
36.3
DDTA
2
3þ
6.53
4.98
2.75
1.99
TDNA
3þ
0
0.01
0.01
0.00
0.00
TDSA
3þ
1
0.19
0.15
0.01
0.00
TDDA TDTA
3þ 3þ
2 3þ
0.38 0.00
0.28 0.02
0.01 0.00
0.01 0.00
0.94
0.89
1.07
1.48
1.17
1.50
a
The number of donor (D) and acceptor (A) H-bonds per water molecule are designated as none (N), single (S), double (D), and triple (T); thus, for example, SDDA represents a single-donor-double-acceptor water molecule. b The values in columns 4-7 are expressed as percentages of the total number of water molecules, either in bulk water or in the neopentane hydration shell (within 6.5 Å of neopentane), for both the Luzar et al.32 and Wernet et al.4 H-bond definitions. The standard deviation of the smallest reported digit of the percentage values is typically less than (1 (and never greater than (5).
initially dangling remains dangling at time t (without ever having formed a H-bond during the intervening time interval). The above correlation function is similar to that previously used to determine the time correlation function of formed H-bonds (by identifying B(t) = 1 when a OH is identified as a H-bond donor and B(t) = 0 when it is not).39 Our results indicate that the latter two correlation functions decay on quite different time scales and reveal that dangling OH groups are far more strongly perturbed by hydrophobic solutes than are formed H-bonds (as further described in section III). OH vibrational frequency predictions were obtained from our simulations using the electric field Stark shift approximation.40-45 In particular, we used the following linear correlation, which we obtained by fitting the electric field strength distribution obtained from our TIP4P pure water simulation, to the experimental OH stretch Raman band of HOD/D2O (see Appendix for further details). V -1 -8 ω ðcm Þ ¼ 3782 - ð1:99 10 ÞE ð2Þ m The orientations of the OH groups relative to the center of mass of neopentane were analyzed by computing J ¼ B r OC 3 B r OH
ð3Þ
whereBr OC is the vector between the water oxygen and the central carbon atom of the neopentane, and Br OH is the vector between the oxygen and hydrogen of the OH group. If J is negative, the OH group is classified as pointing away from the neopentane molecule, while if it is positive, then it is classified as pointing toward neopentane. Note that this orientational metric assures that there would be an equal probability of pointing toward or
away from neopentane if the orientational distribution of OH groups were perfectly isotropic.
III. RESULTS AND DISCUSSION OH Structure. Our simulations reveal the enhanced formation of dangling water OH bonds in the hydration shell of neopentane, as illustrated in Figure 1A, which shows a representative simulation snapshot containing one such dangling OH bond. In order to determine the fraction of water molecules with a particular H-bonding configuration, we counted the number of H-bonds that each water molecule donates and accepts. The results obtained both in bulk water and within the hydration shell of neopentane, using either the Luzar39 or the Wernet7 H-bond definitions, are given in Table 1. The notation in the left-hand column indicates whether each water molecule is a non- (N), single- (S), double- (D), or triple- (T) H-bond donor (D) and acceptor (A). For example, a tetrahedrally H-bonded water molecule, which is both double-donating and double-accepting, is labeled DDDA. Water molecules that contain dangling OH groups are all those with either single- or nondonating designations (SD and ND) but with any number of acceptors (NA, SA, DA, or TA). The triple donor and acceptor probabilities in Table 1 include the rare occasions in which water molecules are found in configurations that are classified as either donating or accepting more than three H-bonds (i.e., when more than three O 3 3 3 O distances and OH 3 3 3 O angles fall within the range classified as a H-bond). In our MD simulations, over 90% of the water molecules are found to have SDSA, SDDA, DDSA, or DDDA H-bond configurations, both in bulk water and in the neopentane hydration shell. 6179
dx.doi.org/10.1021/jp111346s |J. Phys. Chem. A 2011, 115, 6177–6183
The Journal of Physical Chemistry A
Figure 2. (A) Time correlation functions of the H-bonded water molecules (upper two curves) and non-H-bonded water molecules (lower two curves). The lower two curves demonstrate that dangling OH groups persist significantly longer in the neopentane hydration shell (solid curve) than they do in bulk water (dashed curve). The H-bonded populations (upper two curves) have an approximately single exponential time correlation function (with a time constant of about 0.6 ps) and are only slightly longer-lived in the neopentane hydration shell. (B) Excess number of dangling OH groups found within 6.5 Å of neopentane, relative to the number that would have been there if all of the hydration shell water molecules had the same structure as bulk water.
The results in Table 1 indicate that the fraction of waters with at least one dangling OH group is 2-3% greater around neopentane than that in bulk water. Although the Luzar and Wernet H-bond definitions produce different absolute values for the various fractions, the two definitions produce qualitatively (and semiquantitatively) similar results. The following more detailed analyses of the influence of neopentane on the dynamics and vibrational spectra of dangling (and H-bonded) OH groups pertain to results obtained using the Luzar definition. Our simulation results indicate that the probability of finding a OH group that is dangling is only slightly greater in the neopentane hydration shell than it is in bulk water. More specifically, the ratio of probabilities of dangling to H-bonded OH groups is about 0.12 in bulk water and 0.135 for water molecules within 6.5 Å of the center of mass of neopentane. This ratio is equivalent to the equilibrium constant, Keq = [Dangling]/[H-bonded], associated with converting a H-bonded OH to a dangling OH. Thus, the above population ratios (Keq values) imply that in bulk water a dangling OH group has a free energy on the order of ΔG ≈ 5.3 kJ/mol higher than that for a H-bonded OH , while ΔG ≈ 5.0 kJ/mol in the neopentane hydration shell. The latter estimates are not very sensitive to the precise way in which a H-bond is defined as the Wernet H-bond definition implies that the above free energies are only about 1 kJ/mol smaller than those obtained using the Luzar H-bond definition. OH Dynamics. Figure 2 shows results pertaining to the dynamics of dangling (and H-bonded) OH groups, both in the neopentane hydration shell (solid curves) and in bulk water (dotted curves). The time correlation functions shown in Figure 2A clearly reveal that the H-bonded OH groups are significantly longer lived than the dangling OH groups both in bulk water and around neopentane. Moreover, the dangling OH groups around neopentane are clearly longer-lived than those in bulk water. More specifically, the probability that a dangling OH group will persist for at least 0.5 ps is nearly 5 times larger in the neopentane hydration shell than that in bulk water. The latter
ARTICLE
Figure 3. The OH stretching vibrational spectrum calculated using the electric field Stark shift correlation (eq 2) for configurations from the MD simulation of the aqueous neopentane solution. (A) The total OH band (solid curve) is decomposed into sub-bands arising from water molecules with SDSA, SDDA, DDSA, and DDDA H-bonding designations. (B) The spectral contributions arising from the H-bonded and non-H-bonded (dangling) donor OH groups on the SDDA water molecules.
ratio increases to well over a factor of 10 for dangling OH groups that persist for at least 1 ps. Figure 2B shows the total number of dangling OH groups in the hydration shell of neopentane, in excess of the number which would have been there if the hydration shell had the same structure as bulk water. These results indicate that there is, on average, about one excess dangling OH group in the neopentane hydration shell. In our simulations, about half of the latter dangling OH groups are found to live longer than 0.5 ps, while only about 10% live at least 1 ps. The latter number of dangling OH groups is very similar to that obtained from the experimental area of the dangling OH band in the hydration shell of neopentanol (assuming that the Raman cross section of OH groups is the same for both the dangling and H-bonded OH groups).1 Thus, the calculated intensity of the longlived dangling OH band (obtained from our electric field simulations) has the same order of magnitude as that obtained experimentally. The simulations were performed on neopentane, while the experiments were performed using the more highly soluble neopentanol. Because the exposed hydrophobic surface area of the latter molecule is smaller than that of neopentane, the number of dangling OH groups obtained in the neopentane solution simulations are expected to somewhat overestimate the values pertaining to neopentanol solutions. Thus, the simulation results are in resonable agreement with the experimental conclusion that about one in ten of the neopentanol hydration shells contain one water dangling OH group which lives longer than 0.2 ps.1 OH Vibrational Frequency. Figure 3 shows OH vibrational frequency calculations obtained using electric field Stark-shift simulations (eq 2). Figure 3A shows how the total OH band shape compares with the band shapes arising from water molecules with different H-bonding configurations. Figure 3B shows how the OH band of the SDDA subpopulation may be further decomposed to reveal contributions arising from each of the two OH groups on each 500A water molecule, one of which is designated as a H-bond donor and the other as dangling. The latter results clearly 6180
dx.doi.org/10.1021/jp111346s |J. Phys. Chem. A 2011, 115, 6177–6183
The Journal of Physical Chemistry A
Figure 4. The calculated (A) and experimental (B) dangling OH bands in the hydration shell around neopentane (A) and neopentanol (B) are compared with the corresponding bulk water OH spectra (dotted curves). (A) The calculated spectrum of dangling OH groups which live longer than 0.4 ps (solid curve) is decomposed into contributions arising from dangling OH groups pointing either toward or away from the center of mass of neopentane. (B) Comparison of the experimental dangling OH band arising from the hydration shell of neopentanol with the OH band of the solvent (which consists of 10% HOD in D2O). The truncated portion of the experimental spectrum, to the left of the Raman-MCR dangling OH band, contains large features arising from H-bonded OH groups (as well as the CH stretch of neopentanol).
confirm that dangling OH groups have higher OH stretch frequencies than H-bonded OH groups and are consistent with the well-known correlation between OH frequency and H-bond strength. To the best of our knowledge, these results are the first to directly connect specific dangling water OH structures with the corresponding high-frequency OH stretch band in hydrophobic hydration shells. Figure 4 compares theoretical and experimental results pertaining to long-lived dangling OH populations in the solute hydration shell. The solid curve in Figure 4A shows the calculated OH vibrational band for all the dangling OH groups around neopentane that live longer than 0.4 ps (and quite similar spectra were obtained for other cutoff times between 0.2 and 1 ps). The peak maximum is found at ∼3660 cm-1. Figure 4B shows experimental (Raman-MCR) results obtained from solutions of neopentanol in HOD/D2O. The mean frequencies of the theoretical and experimental dangling OH bands are in remarkably good agreement with each other as both peaks are centered at approximately 3660 cm-1. Moreover, the fact that the experimental dangling OH frequency obtained from the hydration shell of neopentanol in HOD/D2O is virtually identical to that previously reported for neopentanol in H2O indicates that the hydration shell dangling OH species are not significantly perturbed by OH resonance coupling in H2O. This again makes sense as the dangling OH groups have frequencies that are nonresonant with the neighboring H-bonded OH groups. Our calculations further indicate that the hydration shell dangling OH frequency is slightly higher for the longer-lived
ARTICLE
dangling OH groups. More specifically, the dangling OH frequency increases by ∼10 cm-1 when the lower bound on the lifetime increases from 0.1 to 0.4 ps. The width of the experimental dangling OH band shown in Figure 4B is clearly somewhat narrower in width than the corresponding theoretical dangling OH band shown in Figure 4A. The ∼50 cm-1 width of the experimental band places a lower bound of ∼0.2 ps on the lifetime of the associated dangling OH species.1 The width of the theoretical dangling OH band is obtained without including the effects of either exchange narrowing (spectral diffusion), lifetime broadening, or dephasing. In other words, at best, the theoretical band shape represents the predicted static distribution of OH frequencies in the absence of any dynamic effects. Thus, the difference in width between the experimental and theoretical dangling OH bands may well reflect limitations of the Stark-shift simulation strategy. The simulation results shown in Figure 4A reveal further information regarding the orientational distribution of the dangling OH bonds ariund neopentane. More specially, in Figure 4A, the total OH band (solid curve) for all dangling OH bonds in the neopentane hydration shell that have a lifetime longer than 0.4 ps has been decomposed into contributions from OH groups which are pointing either toward or away from the central carbon of neopentane. These results reveal that the OH groups have a greater preference for pointing toward neopentane than pointing away from it. Moreover, we find (not shown here) that the very short lived hydration shell dangling OH species do not point toward neopentane. For example, only 50.1% of the population of hydration shell dangling OH groups that persist for less than 0.1 ps point toward neopentane. This percentage increases to 67.9% for the hydration shell dangling OH population that persists for longer than 0.1 ps. Such time-dependent changes in the dangling OH angular distribution make sense, because the shortest lived dangling OH groups are expected to include fleeting structures caught in the act of hopping between two H-bonded states, with a broad angular distribution.
IV. SUMMARY Molecular dynamics and Stark-shift simulations have been performed in order to link the structural and spectroscopic properties of dangling OH groups in the hydration shell around a hydrophobic solute (neopentane). Although the predicted absolute number of dangling OH groups is contingent on the criteria used to define a H-bond, our results and conclusions are found to be substantially the same when using two quite different H-bond definitions. Most significantly, both the concentrations and lifetimes of the hydration shell dangling OH species identified in our simulation are found to be remarkably consistent with the observed intensity and frequency of the experimental (RamanMCR) hydration shell dangling OH vibrational bands. More specially, our new experimental Raman-MCR spectra obtained from solutions of neopentanol in HOD/D2O reveal a dangling OH band which is virtually identical to that previously observed for neopentanol in H2O.1 The latter agreement indicates that dangling OH groups are not significantly influenced by OH resonance coupling. The frequency of the dangling OH band obtained from electric field Stark-shift simulations (∼3660 cm-1) is in essentially prefect agreement with that observed experimentally. The width of the experimental dangling OH band (∼50 cm-1) requires that it have a lifetime of at least 0.2 ps.1 Our simulation 6181
dx.doi.org/10.1021/jp111346s |J. Phys. Chem. A 2011, 115, 6177–6183
The Journal of Physical Chemistry A
ARTICLE
Present Addresses †
Institute of Chemistry, Faculty of Science, “Sts. Cyril and Methodius University”, Skopje, Republic of Macedonia.
’ ACKNOWLEDGMENT This work was supported by the NSF (CHE-0847928) and the Swedish Research Council (VR). Our fond memories of Victoria Buch, and her keen interest in interfacial water, will continue to inspire us. We would also like to thank the Swedish Infrastructure for Computing (SNIC) and the Uppsala Multidisciplinary Center for Advanced Computational Science (UPPMAX) for computing resources under Project s01509-14. J.T.-P. acknowledges a travel grant from Uppsala University. Figure 5. The experimental OH stretch Raman band of 10% HOD in D2O (points) is compared with the band obtained from the electric field strength distribution in TIP4P water. The relationship between the Raman shift frequency (cm-1) and the electric field strength, ω (cm-1) = 3782 - 1.99 10-8E (V/m), is obtained from a least-squares fit to the experimental OH band.
results confirm that dangling OH groups around neopentane are significantly longer lived than those in bulk water. More specifically, our simulation results indicate that each neopentane hydration shells contains approximately one excess dangling OH group (relative to bulk water), and about a tenth of those dangling OH groups are predicted to live for at least 1 ps. The latter dangling OH probabilities are of the same order of magnitude as the experimentally derived estimate that about 1 in 10 hydration shells of neopentanol contain a long-lived dangling OH group in the hydration shell of neopentanol.1 Moreover, our simulation results indicate that the latter long-lived dangling OH groups are highly likely to point toward the neopentane. Finally, our simulation results imply that the free-energy changes associated with converting a H-bonded OH group to a dangling OH group is on the order of 5 kJ/mol in the neopentane hydration shell (which is only about 0.3 kJ/mol smaller than the corresponding free-energy change in bulk water).
’ APPENDIX Vibrational Stark Shift Calibration. The correlation between the electric field strengths and the experimental vibrational frequency was fitted to a linear function (eq 2). The experimental OH stretch Raman band for 10% HOD in D2O was obtained as described in section II. A histogram of the probability of the electric field strength at the H position projected along the OH bond vector for pure TIP4P water was calculated. The experimental spectrum, P(ω), and the simulated electric field strength, R(E), were correlated with the function ω(E) = a - bE by minimizing eq 4 with respect to a and b as well as the dummy variable m, which was introduced to scale the theoretical electric field strength histogram Z ð4Þ r ¼ ðPðωðEÞÞ - mRðEÞÞ2 dE
where the integral is evaluated numerically. The obtained coefficients a and b were subsequently used to calculate all OH spectra, both of pure water and of aqueous neopentane. Figure 5 shows a comparison of the resulting predicted (curve) and experimental (points) OH stretch bands of HOD/D2O.
’ REFERENCES (1) Perera, P. N.; Fega, K. R.; Lawrence, C.; Sundstrom, E. J.; Tomlinson-Phillips, J.; Ben-Amotz, D. Observation of water dangling OH bonds around dissolved nonpolar groups. Proc. Natl. Acad. Sci. U.S. A. 2009, 106, 12230–12234. (2) Stirnemann, G.; Rossky, P. J.; Hynes, J. T.; Laage, D. Water reorientation, hydrogen-bond dynamics and 2D-IR spectroscopy next to an extended hydrophobic surface. Faraday Discuss. 2010, 146, 263–281. (3) Huang, X.; Margulis, C. J.; Berne, B. J. Do molecules as small as neopentane induce a hydrophobic response similar to that of large hydrophobic surfaces?. J. Phys. Chem. B 2003, 107 (42), 11742–11748. (4) Yang, M.; Skinner, J. L. Signatures of coherent vibrational energy transfer in IR and Raman line shapes for liquid water. Phys. Chem. Chem. Phys. 2010, 12 (4), 982–991. (5) Perera, P.; Wyche, M.; Loethen, Y.; Ben-Amotz, D. Soluteinduced perturbations of solvent-shell molecules observed using multivariate Raman curve resolution. J. Am. Chem. Soc. 2008, 130 (14), 4576–4579. (6) Moore, F. G.; Richmond, G. L. Integration or Segregation: How Do Molecules Behave at Oil/Water Interfaces?. Acc. Chem. Res. 2008, 41 (6), 739–748. (7) Wernet, P.; Nordlund, D.; Bergmann, U.; Cavalleri, M.; Odelius, M.; Ogasawara, H.; Naslund, L. A.; Hirsch, T. K.; Ojamae, L.; Glatzel, P.; Pettersson, L. G. M.; Nilsson, A. The structure of the first coordination shell in liquid water. Science 2004, 304 (5673), 995–999. (8) Soper, A. K. An asymmetric model for water structure. J. Phys.: Condens. Matter 2005, 17 (45), S3273–S3282. (9) Smith, J. D.; Cappa, C. D.; Messer, B. M.; Drisdell, W. S.; Cohen, R. C.; Saykally, R. J. Probing the local structure of liquid water by X-ray absorption spectroscopy. J. Phys. Chem. B 2006, 110 (40), 20038–20045. (10) Leetmaa, M.; Ljungberg, M.; Ogasawara, H.; Odelius, M.; Naslund, L. A.; Nilsson, A.; Pettersson, L. G. M. Are recent water models obtained by fitting diffraction data consistent with infrared/Raman and x-ray absorption spectra?. J. Chem. Phys. 2006, 125, 244510. (11) Tokushima, T.; Harada, Y.; Horikawa, Y.; Takahashi, O.; Senba, Y.; Ohashi, H.; Pettersson, L. G. M.; Nilsson, A.; Shin, S. High resolution X-ray emission spectroscopy of water and its assignment based on two structural motifs. J. Electron Spectrosc. Relat. Phenom. 2010, 177 (2-3), 192–205. (12) Kumar, R.; Schmidt, J. R.; Skinner, J. L. Hydrogen bonding definitions and dynamics in liquid water. J. Chem. Phys. 2007, 126 (20), 204107. (13) Eaves, J. D.; Loparo, J. J.; Fecko, C. J.; Roberts, S. T.; Tokmakoff, A.; Geissler, P. L. Hydrogen bonds in liquid water are broken only fleetingly. Proc. Natl. Acad. Sci. U.S.A. 2005, 102 (37), 13019–13022. (14) Laage, D.; Hynes, J. T. A molecular jump mechanism of water reorientation. Science 2006, 311 (5762), 832–835. (15) Frank, H. S.; Evans, J. W. Free volume and entropy in condensed systems 3. Entropy in binary liquid mixtures; partial molar 6182
dx.doi.org/10.1021/jp111346s |J. Phys. Chem. A 2011, 115, 6177–6183
The Journal of Physical Chemistry A entropy in dilute solutions; structure and thermodynamics of aqueous electrolytes. J. Chem. Phys. 1945, 13, 507. (16) Glew, D. N. Aqueous Solubility and Gas-Hydrates— Methane-Water System. J. Phys. Chem. 1962, 66 (4), 605–609. (17) Meng, E. C.; Kollman, P. A. Molecular dynamics studies of the properties of water around simple organic solutes. J. Phys. Chem. 1996, 100 (27), 11460–11470. (18) Buchanan, P.; Aldiwan, N.; Soper, A. K.; Creek, J. L.; Koh, C. A. Decreased structure on dissolving methane in water. Chem. Phys. Lett. 2005, 415 (1-3), 89–93. (19) Rezus, Y. L. A.; Bakker, H. J. Observation of immobilized water molecules around hydrophobic groups. Phys. Rev. Lett. 2007, 99 (14), 148301. (20) Bakulin, A. A.; Liang, C.; Jansen, T. L.; Wiersma, D. A.; Bakker, H. J.; Pshenichnikov, M. S. Hydrophobic Solvation: A 2D IR Spectroscopic Inquest. Acc. Chem. Res. 2009, 42 (9), 1229–1238. (21) Petersen, C.; Tielrooij, K. J.; Bakker, H. J. Strong temperature dependence of water reorientation in hydrophobic hydration shells. J. Chem. Phys. 2009, 130 (21), 214511. (22) Qvist, J.; Halle, B. Thermal signature of hydrophobic hydration dynamics. J. Am. Chem. Soc. 2008, 130 (31), 10345–10353. (23) Qvist, J.; Persson, E.; Mattea, C.; Halle, B. Time scales of water dynamics at biological interfaces: peptides, proteins and cells. Faraday Discuss. 2009, 141, 131–144. (24) Wang, Z. H.; Pang, Y.; Dlott, D. D. Long-lived interfacial vibrations of water. J. Phys. Chem. B 2006, 110 (41), 20115–20117. (25) Ebbinghaus, S.; Kim, S. J.; Heyden, M.; Yu, X.; Heugen, U.; Gruebele, M.; Leitner, D. M.; Havenith, M. An extended dynamical hydration shell around proteins. Proc. Natl. Acad. Sci. U.S.A. 2007, 104 (52), 20749–20752. (26) Laage, D.; Stirnemann, G.; Hynes, J. T. Why Water Reorientation Slows without Iceberg Formation around Hydrophobic Solutes. J. Phys. Chem. B 2009, 113 (8), 2428–2435. (27) Stirnemann, G.; Hynes, J. T.; Laage, D. Water Hydrogen Bond Dynamics in Aqueous Solutions of Amphiphiles. J. Phys. Chem. B 2010, 114 (8), 3052–3059. (28) Matyushov, D. V. Terahertz response of dipolar impurities in polar liquids: On anomalous dielectric absorption of protein solutions. Phys. Rev. E 2010, 81, 021914/1–021914/11. (29) Du, Q.; Freysz, E.; Shen, Y. R. Surface Vibrational Spectroscopic Studies of Hydrogen-Bonding and Hydrophobicity. Science 1994, 264 (5160), 826–828. (30) Wilson, K. R.; Cavalleri, M.; Rude, B. S.; Schaller, R. D.; Nilsson, A.; Pettersson, L. G. M.; Goldman, N.; Catalano, T.; Bozek, J. D.; Saykally, R. J. Characterization of hydrogen bond acceptor molecules at the water surface using near-edge X-ray absorption finestructure spectroscopy and density functional theory. J. Phys.: Condens. Matter 2002, 14 (8), L221–L226. (31) Du, Q.; Superfine, R.; Freysz, E.; Shen, Y. R. Vibrational Spectroscopy of Water at the Vapor Water Interface. Phys. Rev. Lett. 1993, 70 (15), 2313–2316. (32) McGuire, J. A.; Shen, Y. R. Ultrafast vibrational dynamics at water interfaces. Science 2006, 313 (5795), 1945–1948. (33) Berendsen, H. J. C.; Vanderspoel, D.; Vandrunen, R. Gromacs — A Message-Passing Parallel Molecular-Dynamics Implementation. Comput. Phys. Commun. 1995, 91 (1-3), 43–56. (34) Lindahl, E.; Hess, B.; van der Spoel, D. GROMACS 3.0: A package for molecular simulation and trajectory analysis. J. Mol. Model. 2001, 7 (8), 306–317. (35) Van der Spoel, D.; Lindahl, E.; Hess, B.; Groenhof, G.; Mark, A. E.; Berendsen, H. J. C. GROMACS: Fast, flexible, and free. J. Comput. Chem. 2005, 26 (16), 1701–1718. (36) Hess, B.; Kutzner, C.; van der Spoel, D.; Lindahl, E. GROMACS 4: Algorithms for highly efficient, load-balanced, and scalable molecular simulation. J. Chem. Theory Comput. 2008, 4 (3), 435–447.
ARTICLE
(37) Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D.; Impey, R. W.; Klein, M. L. Comparison of Simple Potential Functions for Simulating Liquid Water. J. Chem. Phys. 1983, 79 (2), 926–935. (38) Jorgensen, W. L.; Maxwell, D. S.; TiradoRives, J. Development and testing of the OPLS all-atom force field on conformational energetics and properties of organic liquids. J. Am. Chem. Soc. 1996, 118 (45), 11225–11236. (39) Luzar, A.; Chandler, D. Hydrogen-bond kinetics in liquid water. Nature 1996, 379 (6560), 55–57. (40) Buckingham, A. D. Solvent Effects in Infra-Red Spectroscopy. Proc. R. Soc. London, Ser. A 1958, 248 (1253), 169–182. (41) Hermansson, K.; Lindgren, J.; Probst, M. M. Nonadditivity of OH Frequency-Shifts in Ion Water-Systems. Chem. Phys. Lett. 1995, 233 (4), 371–375. (42) Sadlej, J.; Buch, V.; Kazimirski, J. K.; Buck, U. Theoretical study of structure and spectra of cage clusters (H2O)n, n = 7-10. J. Phys. Chem. A 1999, 103 (25), 4933–4947. (43) Fecko, C. J.; Eaves, J. D.; Loparo, J. J.; Tokmakoff, A.; Geissler, P. L. Ultrafast hydrogen-bond dynamics in the infrared spectroscopy of water. Science 2003, 301 (5640), 1698–1702. (44) Corcelli, S. A.; Skinner, J. L. Infrared and Raman line shapes of dilute HOD in liquid H2O and D2O from 10 to 90 C. J. Phys. Chem. A 2005, 109 (28), 6154–6165. (45) Smith, J. D.; Saykally, R. J.; Geissler, P. L. The effects of dissolved halide anions on hydrogen bonding in liquid water. J. Am. Chem. Soc. 2007, 129 (45), 13847–13856.
6183
dx.doi.org/10.1021/jp111346s |J. Phys. Chem. A 2011, 115, 6177–6183
