Jump to ContentJump to Main Navigation
Show Summary Details
More options …

Zeitschrift für Physikalische Chemie

International journal of research in physical chemistry and chemical physics

Editor-in-Chief: Rademann, Klaus

12 Issues per year


IMPACT FACTOR 2017: 1.144
5-year IMPACT FACTOR: 1.144

CiteScore 2017: 1.08

SCImago Journal Rank (SJR) 2017: 0.495
Source Normalized Impact per Paper (SNIP) 2017: 0.495

Online
ISSN
2196-7156
See all formats and pricing
More options …
Volume 232, Issue 5-6

Issues

Intermolecular Interactions of Pyridine in Liquid Phase and Aqueous Solution Studied by Soft X-ray Absorption Spectroscopy

Masanari Nagasaka
  • Corresponding author
  • Institute for Molecular Science and SOKENDAI (Graduate University for Advanced Studies), Myodaiji, Okazaki 444-8585, Japan
  • Email
  • Other articles by this author:
  • De Gruyter OnlineGoogle Scholar
/ Hayato Yuzawa / Nobuhiro Kosugi
  • Corresponding author
  • Institute for Molecular Science and SOKENDAI (Graduate University for Advanced Studies), Myodaiji, Okazaki 444-8585, Japan
  • Email
  • Other articles by this author:
  • De Gruyter OnlineGoogle Scholar
Published Online: 2018-01-30 | DOI: https://doi.org/10.1515/zpch-2017-1054

Abstract

Intermolecular interactions of pyridine in liquid and in aqueous solution are studied by using soft X-ray absorption spectroscopy (XAS) at the C, N, and O K-edges. XAS of liquid pyridine shows that the N 1s→π* peak is blue shifted and the C 1s→π* peak of the meta and para sites is red shifted, respectively, as compared with XAS of pyridine gas. These shifts in liquid are smaller than those in clusters, indicating that the intermolecular interaction of liquid pyridine is weaker than that of pyridine cluster, as supported by the combination of quantum chemical calculations of the core excitation and molecular dynamics simulations of the liquid structure. On the other hand, XAS spectra of aqueous pyridine solutions (C5H5N)x(H2O)1−x measured at different molar fractions show that in the pyridine rich region, x>0.7, the C and N 1s→π* peak energies are not so different from pure liquid pyridine (x=1.0). In this region, antiparallel displaced structures of pyridine molecules are dominant as in pure pyridine liquid. In the O K-edge XAS, the pre-edge peaks sensitive to the hydrogen bond (HB) network of water molecules show the red shift of −0.15 eV from that of bulk water, indicating that small water clusters with no large-scale HB network are formed in the gap space of structured pyridine molecules. In the water rich region, 0.7>x, the N 1s→π* peaks and the O 1s pre-edge peaks are blue shifted, and the C 1s→π* peaks of the meta and para sites are red-shifted by increasing molar fraction of water. The HB network of bulk water is dominant, but quantum chemical calculations indicate that small pyridine clusters with the HB interaction between the H atom in water and the N atom in pyridine are still existent even in very dilute pyridine solutions.

Keywords: aqueous pyridine solution; inner-shell calculation; molecular dynamics simulation; soft X-ray absorption spectroscopy

Dedicated to: Eckart Rühl on the occasion of his 60th birthday.

1 Introduction

Pyridine is liquid at room temperature due to aromatic π–π interactions of pyridine rings. The structures of pyridine dimer and trimer are extensively studied to understand the intermolecular interaction [1], [2], [3], [4], [5], [6], [7], [8]. The structure of liquid pyridine is investigated by molecular dynamics (MD) [9], [10], [11], [12] and Monte Carlo simulations [13]. These studies indicate that a pyridine dimer shows an antiparallel displaced structure, in which the N atoms are on opposite sites due to the dipole interaction. Pyridine in solid phase has a similar antiparallel stacking structure of pyridine rings [14]. Soft X-ray absorption spectroscopy (XAS) applied to small pyridine clusters with the size of≤20 molecules per cluster at the C and N K-edges has shown that the 1s→π* peak shifts from gas to clusters are dependent on several conformations and core-hole sites [15]. The theoretical investigation suggests that pyridine clusters have a similar antiparallel displaced structure. It is interesting to compare XAS of pyridine liquid with XAS of pyridine clusters to reveal detailed intermolecular interactions in liquid phase.

Pyridine is soluble in water at any concentration, and the formation of hydrogen bond (HB) between the N atom of pyridine and the H atom of water is expected. The HB structure between pyridine and water is mainly studied by vibrational spectroscopy: infrared [16] and Raman [17], [18], [19], [20], [21], [22], [23] spectroscopies. Schlücker et al. [19] measured Raman spectroscopy of aqueous pyridine solutions at different concentrations, and discussed the structures with help of density functional theory (DFT) calculations of several structural models. Local structures and HB structures of aqueous pyridine solutions are investigated by neutron diffraction [24], nuclear magnetic resonance [25], [26], and also theoretical methods [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39].

Recently, some groups have proposed that pyridine is not fully mixed with water, but an inhomogeneous pyridine-water mixture is formed in aqueous solution. In the region of a small molar fraction of pyridine, the formation of pyridine clusters in aqueous solutions is discussed in small angle neutron scattering [40], [41], [42] and Rayleigh scattering [43], [44], [45]. On the other hand, in the region of a large molar fraction of pyridine, the formation of small water clusters is discussed in the Monte Carlo simulation [46].

As described above, local structures of aqueous pyridine solutions are extensively studied by experimental and theoretical methods. However, the structural change of aqueous pyridine solution at different concentrations has not been fully understood yet, because it is difficult to break down intermolecular interactions into pyridine and water experimentally. XAS is an element specific method to reveal local structures of liquid, and the intermolecular interaction of pyridine can be revealed by analyzing the C and N 1s→π* transitions of pyridine molecules [47], [48], [49], [50]. Furthermore, in the C K-edge XAS, the ortho C 1s→π* peak is separately observed from that of meta and para sites of a pyridine ring. The intermolecular interaction of water can be revealed by analyzing the O K-edge XAS of aqueous pyridine solutions, where the structural change of HB network is discussed from the energy shift of the pre-edge peak [51], [52], [53], [54], [55]. The local structure of 3-methylpyridine (3MP) in aqueous solutions is also discussed based on O K-edge XAS [56], [57]. The pre-edge peak is lowered in energy with increasing the molar fraction of 3MP, and is not shifted above the molar fraction of 0.8. The HB interaction of 3MP with water is larger than that of pyridine [58].

Recently, we have developed a liquid flow cell for XAS in transmission mode, and investigated local structures of several aqueous solutions [55]. In the present study, we investigate intermolecular interactions of pyridine in liquid phase and aqueous solution (C5H5N)x(H2O)1−x at different concentrations by XAS in C, N, and O K-edges. The intermolecular interactions of pyridine and water are analyzed using the peak shifts observed in C and N K-edge XAS and in O K-edge XAS, respectively. The peak shifts are evaluated by quantum chemical calculations of inner-shell excitations for several structural models obtained by MD simulations.

2 Methods

2.1 XAS spectroscopy

XAS spectra have been measured on the soft X-ray undulator beamline BL3U at UVSOR-III Synchrotron [59]. Details of our transmission-type liquid flow cell are described elsewhere [55], [60], [61]. In the liquid flow cell, a liquid layer is sandwiched between two 100 nm thick Si3N4 membranes with the windows size of 2×2 mm2. SiC membranes with the thickness of 100 nm are used for N K-edge XAS. Teflon spacers with the thickness of 100 μm are set between the supporting plates of the Si3N4 or SiC membranes. Liquid samples are exchangeable in situ by using a tubing pump. The temperature of liquid samples is controllable and is set to be 25°C in the present work. The liquid cell is installed in a helium chamber, which is separated by a Si3N4 membrane with the window size of 200×200 μm2 (30×30 μm2 is chosen for some other experiments) from the soft X-ray beamline in an ultrahigh vacuum condition. The thickness of a liquid layer is controllable from 20 nm to 2000 nm by adjusting the helium pressure of a helium chamber from 0.12 MPa to 0.1013 MPa [55], [60]. The energy resolutions at the C, N, and O K-edges are set to 0.14 eV, 0.2 eV, and 0.4 eV in the present work, respectively. XAS spectra are based on the Lambert-Beer law, ln(I0/I), where I0 and I are the transmission signals of blank and liquid, respectively, measured by using a photodiode detector in a helium chamber. The photon energies at the C, N, and O K-edges are calibrated by using the first peak (287.96 eV) of methanol gas [62], the first peak of N 1s→π* band (400.84 eV) of N2 gas [63], and the O 1s→π* peak top (530.80 eV) of O2 gas [64], respectively. C and N K-edge XAS spectra of pyridine gas are measured by mixing pyridine vapor into helium in a helium chamber.

2.2 MD simulation

MD simulations are performed by using GROMACS 5.1.2 [65], [66]. The potential of pyridine molecule is described by OPLSAA [67], [68], and that of water molecule is TIP5P [69]. Temperature is controlled by the Nośe-Hoover thermostat method [70], [71]. Pressure is adjusted by the Parrinello–Rahman method [72]. The simulation is performed at a time step of 1 fs with a periodic boundary condition and the partial-mesh Ewald method [73]. In order to obtain equilibrium structures, randomly distributed structures are optimized by the simulations, which run during 100 ps at 100 K in the NVT condition, 100 ps at 200 K and 1 atm in the NPT condition, and 2 ns at 25°C and 1 atm in the NPT condition.

In liquid pyridine, the unit cell consists of 500 pyridine molecules. The equilibrium structures are obtained by sampling the structures every 1 ps at 25°C and 1 atm during a simulation time of 2 ns. In the dilute aqueous pyridine solution (x=0.001), the unit cell consists of 10 pyridine and 9990 water molecules. The equilibrium structures are obtained by sampling the structures every 1 ps during a simulation time of 10 ns.

2.3 Inner-shell calculation

In order to discuss the peak shifts of XAS spectra, we have carried out inner-shell calculations. Geometries of pyridine and water molecules are fully optimized at MP2 level [74] with the basis function of aug-cc-pVDZ [75] by using Gaussian 09 [76]. Model structures of pyridine in liquid phase and aqueous solution are configured by using intermolecular distances and angles determined from radial distribution functions (RDF) obtained by MD simulations.

Inner-shell calculations on the C and N 1s→π* transitions of pyridine molecules have been carried out by using the GSCF3 code [77], [78]. The ground and core excited states are obtained within the Hartree–Fock method, namely, ∆SCF (self-consistent field). The core hole is localized on a specified C or N atoms. The zero point vibrational energy and electron correlation involving van der Waals interactions are ignored in the present calculations. Primitive basis functions are taken from the contracted Gaussian-type functions of Huzinaga et al. [79], (73/7) for C, N, and O and (6) for H. They are augmented with polarization functions ζd=0.617 and 0.864 for C and N, respectively. The contraction schemes are (3111121/3112/1*) for C and N, (3111121/3112) for O, and (42) for H atoms. Diffuse functions are not explicitly included because Rydberg states are not considered in the present calculations. The calculated spectra are convoluted by Gaussian profiles with the width of 0.1 eV.

3 Liquid pyridine

N and C K-edge XAS

Figure 1 shows N K-edge XAS spectra of pyridine gas and liquid at 25°C. Figure 2 shows C K-edge XAS spectra of pyridine gas and liquid at 25°C. Similarly to previous C K-edge spectra of pyridine gas [47], [48], [49], [50], the C 1s→π* transition shows two peaks: The C1 peak corresponds to the meta and para sites and the C2 peak to the ortho site of a pyridine ring. Table 1 shows peak shifts at different sites (C1, C2, and N) from pyridine gas to liquid and cluster structures. The energy shifts of small pyridine clusters are taken from the previous XAS study [15]. The C1 peak of liquid pyridine shows the red shift of −0.077 eV from gas. This shift is smaller than that of cluster (−0.09 eV). The C2 peak of liquid pyridine is slightly blue shifted from gas (0.018 eV), whereas the C2 peak of clusters is not shifted from gas within accuracy. The N peak of liquid pyridine is blue shifted to 0.089 eV, and it is slightly smaller than that of clusters (0.095 eV). Thus, the energy shifts of liquid pyridine are slightly smaller than those of cluster peaks.

N K-edge XAS of pyridine in gas and liquid phases. The inset shows a core excited N atom in pyridine molecule.
Fig. 1:

N K-edge XAS of pyridine in gas and liquid phases. The inset shows a core excited N atom in pyridine molecule.

C K-edge XAS of pyridine in gas and liquid phases. The inset shows core excited C atoms in pyridine molecule. The C1 peak is derived from meta and para sites, and the C2 peak is ortho sites.
Fig. 2:

C K-edge XAS of pyridine in gas and liquid phases. The inset shows core excited C atoms in pyridine molecule. The C1 peak is derived from meta and para sites, and the C2 peak is ortho sites.

Tab. 1:

Energy shifts of XAS peaks at different sites (C1, C2, and N) of pyridine molecules in eV from gas to liquid and cluster structures.

3.2 Inner-shell calculations

In order to obtain local structures of pyridine in liquid and in cluster, we have performed quantum chemical inner-shell calculations. The DFT study reported that the structure of a pyridine dimer is an antiparallel displaced structure [2], as shown in the inset of Figure 3a. This structure is most stable due to the dipole interaction of pyridine rings, and can be used for the model of pyridine clusters. The distance d between center-of-mass of pyridine rings (Pycom) in the perpendicular direction is 3.32 Å and is displaced by 1.24 Å along the molecular planes. The tilted angle ϕ from the normal vector of the pyridine ring is 20.5°. On the other hand, we have obtained RDF Pycom–Pycom from the MD simulation of liquid pyridine at 25°C, and found that the first peak position is 3.72 Å. Assuming that a pyridine molecule is displaced by 1.24 Å along the molecular planes, the distance d becomes 3.51 Å, and the tilted angle ϕ is 19.47°. The intermolecular distance d of liquid pyridine is larger than that of a cluster.

Calculated spectral shifts of pyridine from gas to liquid and cluster structures at different sites. (a) Calculated N K-edge spectra relative to the peak of pyridine gas. The inset shows a schematic of liquid and cluster structures which is determined from the distance d and the tilted angle ϕ from normal vector of the pyridine ring plane. (b) Calculated C K-edge spectra relative to the peak of ortho site in pyridine gas.
Fig. 3:

Calculated spectral shifts of pyridine from gas to liquid and cluster structures at different sites. (a) Calculated N K-edge spectra relative to the peak of pyridine gas. The inset shows a schematic of liquid and cluster structures which is determined from the distance d and the tilted angle ϕ from normal vector of the pyridine ring plane. (b) Calculated C K-edge spectra relative to the peak of ortho site in pyridine gas.

Figure 3 shows the calculated inner-shell spectra of gas, liquid, and cluster, whose structures are obtained by MD simulations. As shown in the inset of Figure 3a, the center molecule in the pyridine trimer is used for the inner-shell calculations in the C and N K-edges. Figure 3a shows the calculated N K-edge spectra, where the peak energy is relative to the N 1s→π* peak of pyridine gas. The peak energy of calculated C K-edge spectra shown in Figure 3b is relative to the C 1s→π* peak of the ortho site of pyridine gas. Table 2 shows the calculated energy shifts of different sites of pyridine in liquid and in cluster from gas. The calculated energy shift of the N atom becomes higher in the order of gas, liquid, and cluster. These energy shifts are consistent with the energy shifts of the N atom obtained by XAS experiments. The calculated energy shift of the ortho site in liquid is 0.017 eV, and is consistent with that obtained by the experiment (0.018 eV). The calculated energy shifts of the meta and para sites become lower in the order of gas, liquid, and cluster, and is good agreement with those of the C1 peaks, though the present calculations predict that the energy shift of the para site is larger than that of the meta site. As described above, the calculated energy shifts of liquid and cluster structures are consistent with the experimental shifts. The smaller energy shifts in liquid pyridine than in cluster pyridine arise from the longer intermolecular distance between pyridine molecules in liquid than in cluster.

Tab. 2:

Calculated energy shifts of peaks from gas to liquid and cluster structures at the different sites of pyridine molecules in eV taken from Figure 3.

4 Aqueous pyridine solutions

N and C K-edge XAS

Figure 4 shows N K-edge XAS spectra of aqueous pyridine solutions (C5H5N)x(H2O)1−x at different concentrations at 25°C. The N peaks are evidently blue shifted by decreasing molar fraction of pyridine (increasing molar fraction of water). Figure 5 shows C K-edge XAS spectra of aqueous pyridine solutions at different concentrations. By increasing molar fraction of water, the C1 peak of the meta and para site shows red shift; on the other hand, the C2 peak of the ortho site is blue shifted.

N K-edge XAS of aqueous pyridine solutions at different concentrations. The N peaks are blue shifted to the direction of an arrow by decreasing molar fraction of pyridine. The inset shows a core excited N atom in a pyridine molecule.
Fig. 4:

N K-edge XAS of aqueous pyridine solutions at different concentrations. The N peaks are blue shifted to the direction of an arrow by decreasing molar fraction of pyridine. The inset shows a core excited N atom in a pyridine molecule.

C K-edge XAS of aqueous pyridine solutions at different concentrations. The C1 and C2 peaks are red and blue shifted to the direction of arrows by decreasing molar fraction of pyridine. The inset shows core excited C atoms in a pyridine molecule.
Fig. 5:

C K-edge XAS of aqueous pyridine solutions at different concentrations. The C1 and C2 peaks are red and blue shifted to the direction of arrows by decreasing molar fraction of pyridine. The inset shows core excited C atoms in a pyridine molecule.

Figure 6 shows the energy shifts of the C1, C2, and N peaks at different concentrations in aqueous pyridine solutions (C5H5N)x(H2O)1−x from pure pyridine liquid (x=1.0). The energy shifts of the C1, C2, and N peaks show two concentration regions with the borders of x=0.7. In the region of x>0.7, the peak energies are almost constant. In the region of 0.7>x, on the other hand, the C1, C2, and N peak energies are changed linearly by increasing molar fraction of water. The C1 peak is red shifted, and the energy shift from pure pyridine (x=1.0) is −0.072 eV at x=0.05. The C2 peak is slightly blue shifted, and the energy shift is 0.026 eV at x=0.05. The N peak is blue shifted by increasing molar fraction of water, and reaches to 0.110 eV at x=0.05. The rather large energy shift of the N peak is reasonable, considering that the energy shift arises from HB between the N atom of pyridine and the H atom of water in aqueous pyridine solution.

Energy shifts of the C1, C2, and N peaks in aqueous pyridine solutions (C5H5N)x(H2O)1−x from pure pyridine liquid (x=1.0) obtained by XAS in N and C K-edges shown in Figures 4 and 5, respectively. The inset shows different core excited sites in a pyridine molecule.
Fig. 6:

Energy shifts of the C1, C2, and N peaks in aqueous pyridine solutions (C5H5N)x(H2O)1−x from pure pyridine liquid (x=1.0) obtained by XAS in N and C K-edges shown in Figures 4 and 5, respectively. The inset shows different core excited sites in a pyridine molecule.

4.2 Inner-shell calculations

In order to obtain hydration structures of pyridine, we have performed the MD simulation of a dilute aqueous pyridine solution at x=0.001. Figure 7 shows RDF of oxygen atoms (Ow) in water molecules around pyridine molecules. Pyridine molecules are completely hydrated in the simulated dilute solution. Figure 7a shows RDF of the N atom in pyridine with Ow at the direction from Pycom to the N atom. The first peak of Ow is 2.89 Å from the N atom. Tilted angle of water molecules is 45° along the direction from Pycom to the N atom. This result means the formation of HB between the N atom in pyridine and the H atom in water, as shown in the inset of Figure 7. Figure 7b shows RDF of Pycom with Ow along the normal vector of pyridine ring plane. From the first peak of RDF, water molecules exist at the distance of 3.47 Å in the upper and lower directions of pyridine rings. The configuration of water molecule is shown in the inset of Figure 7, in which the tilted angle of water is 59° from the normal vector of the pyridine ring plane. The first peak width of RDF Pycom–Ow is broader than that of N–Ow. The intermolecular interactions of pyridine with water at the upper and lower directions are weaker than the HB interaction of the N atom with water.

RDF of pyridine-water mixtures in the dilute solution. (a) RDF of the N atom in pyridine with the oxygen atom (Ow) in water along the direction from Pycom to the N atom. (b) RDF of Pycom with Ow along the normal vector of pyridine ring planes.
Fig. 7:

RDF of pyridine-water mixtures in the dilute solution. (a) RDF of the N atom in pyridine with the oxygen atom (Ow) in water along the direction from Pycom to the N atom. (b) RDF of Pycom with Ow along the normal vector of pyridine ring planes.

We have performed inner-shell calculations of aqueous pyridine solutions at different structural models obtained by MD simulations. Figure 8a and b show calculated spectra of three different models (HB1, HB2, and HB3) at the N and C K-edges, respectively. Table 3 shows the calculated energy shifts of different sites of pyridine at three HB models from liquid pyridine. The HB1 model is a pyridine-water dimer, in which HB between the N atom of pyridine and water is formed. In calculated N K-edge XAS of the HB1 model, the energy shift of the N peak is 0.076 eV from liquid pyridine. This energy shift is slightly smaller than the experimental value (0.110 eV at x=0.05), but the blue shift is consistent. In the calculated C K-edge spectra of the HB1 model, on the other hand, the energy shift of the ortho site from liquid pyridine is low (−0.036 eV), and is inconsistent with that of the C2 peak (0.026 eV at x=0.05). The energy shifts of the meta and para sites are 0.029 eV and 0.015 eV, respectively, and are largely different from the C1 peak (−0.072 eV at x=0.05). The peak shifts estimated in the HB1 model are in disagreement with the experimental ones.

Calculated spectral shifts of different models (HB1, HB2, and HB3) of pyridine-water mixtures from liquid pyridine at different sites. The inset shows schematics of different HB model structures. (a) Calculated N K-edge spectra relative to the peak of liquid pyridine. (b) Calculated C K-edge spectra relative to the ortho peak of liquid pyridine.
Fig. 8:

Calculated spectral shifts of different models (HB1, HB2, and HB3) of pyridine-water mixtures from liquid pyridine at different sites. The inset shows schematics of different HB model structures. (a) Calculated N K-edge spectra relative to the peak of liquid pyridine. (b) Calculated C K-edge spectra relative to the ortho peak of liquid pyridine.

Tab. 3:

Calculated energy shifts of peaks from liquid to different HB models (HB1, HB2, and HB3) of pyridine-water mixtures at different sites of pyridine in eV taken from Figure 8.

We have calculated the HB2 model. As shown in the inset of Figure 8, two additional water molecules are weakly bound in the upper and lower directions of the pyridine ring plane in comparison with HB1. This model is taken from MD simulations. The energy shift of the HB2 model in the calculated N K-edge spectra (−0.005 eV) is inconsistent with the experiment. In the calculated C K-edge spectra, the ortho and para sites show red shift (−0.115 eV) and blue shift (0.023 eV), and are also inconsistent with the experimental shifts of the C2 and C1 peaks, respectively. The peak shifts estimated in the HB2 model are still in disagreement with the experimental ones.

It is reported that pyridine is not completely mixed with water but pyridine clusters are formed in aqueous solutions [40], [41], [42], [45]. As shown in the inset of Figure 8, the HB3 model has an antiparallel displaced structure (d=3.51 Å, ϕ=19.47°) obtained by MD simulations of liquid pyridine, and a water molecule forms HB with the N atom in the center pyridine molecule. In the calculated N K-edge spectrum of the center pyridine molecule, the N peak shows a blue shift of 0.153 eV from that of liquid pyridine, consistent with the experiment (0.110 eV). In the calculated C K-edge spectrum, the energy shift of the ortho site is negligibly small (−0.006 eV), though the experimental C2 peak shows small but blue shift (0.026 eV). On the other hand, the energy shifts of the meta and para site are −0.013 eV and −0.070 eV, respectively, consistent with the experimental C1 peak shift (−0.072 eV). As described above, the energy shifts of the C1 and N peaks in the XAS spectra of aqueous pyridine solutions are consistently explained by the inner-shell calculations of the HB3 model. The C2 peak shift is inconsistent and the ortho site would have interaction with some other water molecules.

4.3 O K-edge XAS

In order to investigate structural changes of HB between water molecules under interacting with pyridine molecules, we have measured O K-edge XAS of aqueous pyridine solutions (C5H5N)x(H2O)1−x at different concentrations as shown in Figure 9. The pre-edge peak (535 eV) corresponds to a transition from O 1s to 4a1 unoccupied orbital. Because the 4a1 orbital is mainly distributed at O atoms of water molecules, the energy shift of the pre-edge peak reflects HB between water. In the O K-edge XAS spectra of pure liquid water, the pre-edge peak shows a red shift by increasing temperature [51], [52], [53], [54], [55]. This is due to the elongation of the intermolecular distance between water molecules [80]. The intensity of the post-edge maximum at 540 eV is also decreased when the HB network is partially broken and the multiple scattering in the HB network is weakened [80].

O K-edge XAS of aqueous pyridine solutions at different concentrations. The pre-edge peak is red shifted to the direction of an arrow by increasing molar fraction of pyridine. The inset is expansion of the pre-edge region.
Fig. 9:

O K-edge XAS of aqueous pyridine solutions at different concentrations. The pre-edge peak is red shifted to the direction of an arrow by increasing molar fraction of pyridine. The inset is expansion of the pre-edge region.

Figure 10 shows the energy shift of the pre-edge peak as a function of molar fraction. In 0.7>x, the pre-edge peak is linearly shifted to the lower energy by increasing molar fraction of pyridine. In x>0.7, the pre-edge peak shift is almost constant. The energy shift of the pre-edge peak at x=0.9 is about −0.15 eV from bulk water (x=0.0). Considering the energy shift of bulk water from 25°C to 80°C is about −0.07 eV [80], the lower energy shift of the pre-edge peak at x=0.9 is rather large. At the high molar fraction of pyridine, tetrahedral HB between water molecules is elongated and partly broken by the hydrophobic interactions of pyridine molecules. The decrease of the post-edge intensity at the higher molar fraction of pyridine also indicates the breaking of the HB network between water molecules. It is consistent with the previous Monte Carlo simulation that small water clusters are formed at the high concentration of pyridine [46]. The energy shift of the pre-edge peak in aqueous pyridine solutions is close to that of aqueous 3MP solutions [56], in which the pre-edge peak is lowered by increasing molar fraction of 3MP and becomes constant at x>0.8. The peak energy shift in aqueous 3MP solutions (−0.4 eV) is much larger than that in aqueous pyridine solutions, indicating that there is no HB network in water and a strong HB interaction with 3MP.

Energy shifts of the pre-edge peaks in the O K-edge XAS spectra of aqueous pyridine solutions (C5H5N)x(H2O)1−x from pure liquid water (x=0.0). The inset shows typical HB of bulk water.
Fig. 10:

Energy shifts of the pre-edge peaks in the O K-edge XAS spectra of aqueous pyridine solutions (C5H5N)x(H2O)1−x from pure liquid water (x=0.0). The inset shows typical HB of bulk water.

In order to investigate the HB network of water under interacting with pyridine, we have performed the MD simulation of the dilute pyridine solution at x=0.001. Figure 11 shows RDF Ow–Ow around water in bulk phase and around the N atom of pyridine, where the first peaks are found at 2.73 Å and 2.74 Å, respectively; on the other hand, the second peaks are 4.48 Å and 4.61 Å, respectively. The second peak around N atom of pyridine becomes larger by the elongation of HB network of water by the hydrophobic hydration of pyridine. The MD simulations are consistent with the formation of small water clusters in the higher pyridine concentration, that is influenced by the hydrophobic hydration of pyridine.

RDF Ow–Ow around water in bulk phase and around the N atom of pyridine shown in Figure 7a in aqueous pyridine solutions.
Fig. 11:

RDF Ow–Ow around water in bulk phase and around the N atom of pyridine shown in Figure 7a in aqueous pyridine solutions.

4.4 Intermolecular interactions at different concentrations

As above discussed, in the pyridine rich region, x>0.7, the C and N 1s→π* peak energies are no so different from those of pure pyridine (x=1.0), indicating that pyridine molecules keep antiparallel displaced structures with no evident change of the intermolecular interaction around pyridine. The pre-edge peaks in the O K-edge XAS spectra are almost constant in this region as well and the energy shift from pure water (x=0.0) is about −0.15 eV (red shift), indicating that the HB network in water is very much fragmented and the HB in small water clusters is elongated (or partly broken) by the hydrophobic interaction of pyridine. These results are consistent with the previous Monte Carlo simulations [46].

In the water rich region, 0.7>x, the C, N, and O K-edge peaks of aqueous pyridine solutions are linearly shifted by increasing molar fraction of water. It is reasonable that the pre-edge peak of the O K-edge XAS spectra is shifted to that of bulk water. The theoretical analysis of the π* peak shifts in the C and N K-edge XAS spectra has indicated that the HB between the N atom of pyridine and the H atom of water is gradually established but small pyridine clusters with antiparallel displaced structures are still existent in the HB network of bulk water. These results are consistent with the previous studies [40], [41], [42], [45].

5 Summary

In the present work, we have investigated intermolecular interactions of pyridine in liquid phase and aqueous solution by XAS at the C, N, and O K-edges; the intermolecular interaction of pyridine by C and N K-edge XAS and the intermolecular interaction of water by O K-edge XAS.

From gas to pure liquid pyridine, the N 1s→π* peak is blue shifted, and the C 1s→π* peak of the meta and para sites (C1) is red shifted. These behaviors are the same as in small pyridine clusters [15], but the peak shifts for pyridine liquid are smaller than those of pyridine clusters. The difference in the peak shift arises from the difference in the intermolecular distance of the antiparallel displaced structures as reproduced by quantum chemical calculations of these core excited states by using the geometries determined by MD simulations. The intermolecular distance between center-of-mass of pyridine rings in the perpendicular direction is 3.51 Å in liquid and 3.32 Å in clusters.

We have investigated local structures of pyridine aqueous solutions (C5H5N)x(H2O)1−x at different concentrations. In the pyridine rich region, x>0.7, the energy of C and N peaks are no so different from those of pure pyridine (x=1.0), indicating that pyridine molecules keep antiparallel displaced structures with no evident change of the intermolecular interaction around pyridine. The pre-edge peaks in the O K-edge XAS spectra are almost constant in this region as well and show the red shift of −0.15 eV from that of pure water (x=0.0). It indicates that the HB network in water is very much fragmented and the HB in small water clusters is elongated (or partly broken) by the hydrophobic interaction of pyridine. These results are consistent with the previous Monte Carlo simulations [46].

In the water rich region, 0.7>x, the C, N, and O K-edge peaks of aqueous pyridine solutions are linearly shifted by increasing molar fraction of water. It is reasonable that the pre-edge peak in the O K-edge XAS spectra is shifted to that of bulk water. From the energy shifts of the π* peaks in the C and N K-edge XAS and inner-shell calculations, we have found that small pyridine clusters exist in the HB network of water and is formed HB between N atom of pyridine and the H atom of water. These results are consistent with the previous studies [40], [41], [42], [45].

Acknowledgements

M.N. and N.K. are grateful to Prof. Eckart Rühl for his continuous encouragement of the extension from cluster work to liquid work. The authors are also grateful to Mr. Toshio Horigome for his contribution to the development of liquid flow cells and to the staff members of UVSOR-III Synchrotron for their kind supports. This work is supported by JSPS Grants-in-Aid for Scientific Research (JSPS KAKENHI Grant Nos. 20350014, 23245007, 23685006, 26248010, and 17H03013), Grant for Basic Science Research Projects from the Sumitomo Foundation, and Morino Foundation for Molecular Science. The computations were performed using Research Center for Computational Science, Okazaki, Japan.

References

  • 1.

    B. K. Mishra, N. Sathyamurthy, J. Phys. Chem. A 109 (2005) 6. PubMedGoogle Scholar

  • 2.

    M. Piacenza, S. Grimme, ChemPhysChem 6 (2005) 1554. CrossrefPubMedGoogle Scholar

  • 3.

    C. M. Baker, G. H. Grant, J. Chem. Theory Comput. 3 (2007) 530. CrossrefPubMedGoogle Scholar

  • 4.

    E. G. Hohenstein, C. D. Sherrill, J. Phys. Chem. A 113 (2009) 878. PubMedCrossrefGoogle Scholar

  • 5.

    O. A. Zhikol, O. V. Shishkin, Int. J. Quant. Chem. 112 (2012) 3008. CrossrefGoogle Scholar

  • 6.

    D. B. Ninković, J. M. Andrić, S. D. Zarić, ChemPhysChem 14 (2013) 237. PubMedCrossrefGoogle Scholar

  • 7.

    J. D. Zhang, Y. Gao, W. Yao, S. J. Li, F. M. Tao, Comput. Theor. Chem. 1049 (2014) 82. CrossrefGoogle Scholar

  • 8.

    A. J. Misquitta, A. J. Stone, J. Chem. Theory Comput. 12 (2016) 4184. PubMedCrossrefGoogle Scholar

  • 9.

    Z. Gamba, M. L. Klein, Chem. Phys. 130 (1989) 15. CrossrefGoogle Scholar

  • 10.

    K. Sagarik, E. Spohr, Chem. Phys. 199 (1995) 73. CrossrefGoogle Scholar

  • 11.

    I. Bakó, T. Radnai, G. Pálinkás, Z. Naturforsch. A 51 (1996) 859. Google Scholar

  • 12.

    Z. Trumpakaj, B. Linde, J. Mol. Struct. 1085 (2015) 268. CrossrefGoogle Scholar

  • 13.

    W. L. Jorgensen, N. A. McDonald, Theochem. J. Mol. Struct. 424 (1998) 145. CrossrefGoogle Scholar

  • 14.

    G. V. Janjić, D. B. Ninković, S. D. Zarić, Acta Crystallogr. Sect. B 69 (2013) 389. CrossrefGoogle Scholar

  • 15.

    I. L. Bradeanu, N. Kosugi, R. Flesch, E. Rühl, J. Phys. Chem. A 112 (2008) 9192. PubMedCrossrefGoogle Scholar

  • 16.

    H. Takahashi, K. Mamola, E. K. Plyler, J. Mol. Spectrosc. 21 (1966) 217. CrossrefGoogle Scholar

  • 17.

    K. Fukushima, M. Yamaji, Bull. Chem. Soc. Jpn. 55 (1982) 2779. CrossrefGoogle Scholar

  • 18.

    E. Zoidis, J. Yarwood, Y. Danten, M. Besnard, Mol. Phys. 85 (1995) 373. CrossrefGoogle Scholar

  • 19.

    S. Schlücker, R. K. Singh, B. P. Asthana, J. Popp, W. Kiefer, J. Phys. Chem. A 105 (2001) 9983. CrossrefGoogle Scholar

  • 20.

    S. Schlücker, M. Heid, R. K. Singh, B. P. Asthana, J. Popp, W. Kiefer, Z. Phys. Chem. 216 (2002) 267. Google Scholar

  • 21.

    E. R. Berg, S. A. Freeman, D. D. Green, D. J. Ulness, J. Phys. Chem. A 110 (2006) 13434. CrossrefPubMedGoogle Scholar

  • 22.

    A. G. Kalampounias, G. Tsilomelekis, S. Boghosian, J. Mol. Liq. 198 (2014) 299. CrossrefGoogle Scholar

  • 23.

    A. G. Kalampounias, G. Tsilomelekis, S. Boghosian, Spectrochim. Acta Mol. Biomol. Spectrosc. 135 (2015) 31. CrossrefGoogle Scholar

  • 24.

    Y. Kameda, Y. Amo, T. Usuki, J. Mol. Liq. 164 (2011) 29. CrossrefGoogle Scholar

  • 25.

    W. M. Litchman, J. Magn. Reson. 14 (1974) 286. Google Scholar

  • 26.

    J. P. Kintzinger, Mol. Phys. 30 (1975) 673. CrossrefGoogle Scholar

  • 27.

    W. Adam, A. Grimison, R. Hoffmann, C. Z. de Ortiz, J. Am. Chem. Soc. 90 (1968) 1509. CrossrefGoogle Scholar

  • 28.

    M. Genest, J. Comput. Chem. 8 (1987) 67. CrossrefGoogle Scholar

  • 29.

    B. Parbhoo, O. B. Nagy, J. Mol. Struct. 177 (1988) 393. CrossrefGoogle Scholar

  • 30.

    J. Smets, W. McCarthy, G. Maes, L. Adamowicz, J. Mol. Struct. 476 (1999) 27. CrossrefGoogle Scholar

  • 31.

    A. Dkhissi, L. Adamowicz, G. Maes, J. Phys. Chem. A 104 (2000) 2112. CrossrefGoogle Scholar

  • 32.

    I. Pápai, G. Jancsó, J. Phys. Chem. A 104 (2000) 2132. CrossrefGoogle Scholar

  • 33.

    Z.-L. Cai, J. R. Reimers, J. Phys. Chem. A 106 (2002) 8769. CrossrefGoogle Scholar

  • 34.

    T. Malaspina, K. Coutinho, S. Canuto, J. Chem. Phys. 117 (2002) 1692. CrossrefGoogle Scholar

  • 35.

    E. E. Fileti, K. Coutinho, T. Malaspina, S. Canuto, Phys. Rev. E 67 (2003) 061504. CrossrefGoogle Scholar

  • 36.

    K. V. Berezin, V. V. Nechaev, S. N. Zotov, J. Struct. Chem. 45 (2004) 388. CrossrefGoogle Scholar

  • 37.

    A. Y. Li, H. B. Ji, L. J. Cao, J. Chem. Phys. 131 (2009) 164305. CrossrefPubMedGoogle Scholar

  • 38.

    L. J. A. Cao, A. Y. Li, H. B. Ji, L. Xu, Y. Zhang, Theochem-J. Mol. Struct. 959 (2010) 80. CrossrefGoogle Scholar

  • 39.

    M. Macchiagodena, G. Mancini, M. Pagliai, G. Del Frate, V. Barone, Chem. Phys. Lett. 677 (2017) 120. CrossrefGoogle Scholar

  • 40.

    L. Almásy, G. Jancsó, J. Mol. Liq. 113 (2004) 61. CrossrefGoogle Scholar

  • 41.

    I. Bakó, G. Pálinkás, J. C. Dore, H. Fischer, P. Jóvári, Chem. Phys. Lett. 388 (2004) 468. CrossrefGoogle Scholar

  • 42.

    W. Marczak, B. Czech, L. Almásy, D. Lairez, Phys. Chem. Chem. Phys. 13 (2011) 6260. PubMedCrossrefGoogle Scholar

  • 43.

    W. Schindler, H. A. Posch, Chem. Phys. 43 (1979) 9. CrossrefGoogle Scholar

  • 44.

    C. H. Wang, S. L. Whittenburg, P.-A. Lund, D. H. Christensen, J. Chem. Phys. 72 (1980) 4228. CrossrefGoogle Scholar

  • 45.

    N. Ito, T. Kato, J. Phys. Chem. 88 (1984) 801. CrossrefGoogle Scholar

  • 46.

    A. L. L. Sinoti, J. R. D. S. Politi, L. C. G. Freitas, J. Braz. Chem. Soc. 7 (1996) 133. CrossrefGoogle Scholar

  • 47.

    J. A. Horsley, J. Stöhr, A. P. Hitchcock, D. C. Newbury, A. L. Johnson, F. Sette, J. Chem. Phys. 83 (1985) 6099. CrossrefGoogle Scholar

  • 48.

    C. Hannay, D. Duflot, J.-P. Flament, M.-J. Hubin-Franskin, J. Chem. Phys. 110 (1999) 5600. CrossrefGoogle Scholar

  • 49.

    C. Kolczewski, R. Püttner, O. Plashkevych, H. Ǻgren, V. Staemmler, M. Martins, G. Snell, A. S. Schlachter, M. Sant‘Anna, G. Kaindl, L. G. M. Pettersson, J. Chem. Phys. 115 (2001) 6426. CrossrefGoogle Scholar

  • 50.

    A. Baiardi, M. Mendolicchio, V. Barone, G. Fronzoni, G. A. C. Jimenez, M. Stener, C. Grazioli, M. de Simone, M. Coreno, J. Chem. Phys. 143 (2015) 204102. CrossrefPubMedGoogle Scholar

  • 51.

    J. D. Smith, C. D. Cappa, K. R. Wilson, B. M. Messer, R. C. Cohen, R. J. Saykally, Science 306 (2004) 851. PubMedCrossrefGoogle Scholar

  • 52.

    P. Wernet, D. Nordlund, U. Bergmann, M. Cavalleri, M. Odelius, H. Ogasawara, L. Å. Näslund, T. K. Hirsch, L. Ojamäe, P. Glatzel, L. G. M. Pettersson, A. Nilsson, Science 304 (2004) 995. PubMedCrossrefGoogle Scholar

  • 53.

    T. Pylkkänen, A. Sakko, M. Hakala, K. Hämäläinen, G. Monaco, S. Huotari, J. Phys. Chem. B 115 (2011) 14544. CrossrefPubMedGoogle Scholar

  • 54.

    J. Meibohm, S. Schreck, P. Wernet, Rev. Sci. Instrum. 85 (2014) 103102. CrossrefPubMedGoogle Scholar

  • 55.

    M. Nagasaka, H. Yuzawa, N. Kosugi, J. Electron Spectrosc. Relat. Phenom. 200 (2015) 293. CrossrefGoogle Scholar

  • 56.

    H. Arai, Y. Horikawa, K. Sadakane, T. Tokushima, Y. Harada, Y. Senba, H. Ohashi, Y. Takata, S. Shin, Phys. Chem. Chem. Phys. 14 (2012) 1576. CrossrefPubMedGoogle Scholar

  • 57.

    T. Tokushima, Y. Horikawa, O. Takahashi, H. Arai, K. Sadakane, Y. Harada, Y. Takata, S. Shin, Phys. Chem. Chem. Phys. 16 (2014) 10753. PubMedCrossrefGoogle Scholar

  • 58.

    M. Darvas, P. Jedlovszky, G. Jancsó, J. Phys. Chem. B 113 (2009) 7615. PubMedCrossrefGoogle Scholar

  • 59.

    T. Hatsui, E. Shigemasa, N. Kosugi, AIP Conf. Proc. 705 (2004) 921. CrossrefGoogle Scholar

  • 60.

    M. Nagasaka, T. Hatsui, T. Horigome, Y. Hamamura, N. Kosugi, J. Electron Spectrosc. Relat. Phenom. 177 (2010) 130. CrossrefGoogle Scholar

  • 61.

    M. Nagasaka, K. Mochizuki, V. Leloup, N. Kosugi, J. Phys. Chem. B 118 (2014) 4388. PubMedCrossrefGoogle Scholar

  • 62.

    Y. Tamenori, K. Okada, O. Takahashi, S. Arakawa, K. Tabayashi, A. Hiraya, T. Gejo, K. Honma, J. Chem. Phys. 128 (2008) 124321. CrossrefPubMedGoogle Scholar

  • 63.

    R. Flesch, A. A. Pavlychev, J. J. Neville, J. Blumberg, M. Kuhlmann, W. Tappe, F. Senf, O. Schwarzkopf, A. P. Hitchcock, E. Rühl, Phys. Rev. Lett. 86 (2001) 3767. CrossrefPubMedGoogle Scholar

  • 64.

    M. Coreno, M. de Simone, K. C. Prince, R. Richter, M. Vondráček, L. Avaldi, R. Camilloni, Chem. Phys. Lett. 306 (1999) 269. CrossrefGoogle Scholar

  • 65.

    B. Hess, C. Kutzner, D. van der Spoel, E. Lindahl, J. Chem. Theory Comput. 4 (2008) 435. PubMedCrossrefGoogle Scholar

  • 66.

    M. J. Abraham, T. Murtola, R. Schulz, S. Páll, J. C. Smith, B. Hess, E. Lindahl, SoftwareX 1-2 (2015) 19. CrossrefGoogle Scholar

  • 67.

    W. L. Jorgensen, J. Tirado-Rives, Proc. Natl. Acad. Sci. USA 102 (2005) 6665. CrossrefGoogle Scholar

  • 68.

    C. Caleman, P. J. van Maaren, M. Y. Hong, J. S. Hub, L. T. Costa, D. van der Spoel, J. Chem. Theory Comput. 8 (2012) 61. CrossrefPubMedGoogle Scholar

  • 69.

    M. W. Mahoney, W. L. Jorgensen, J. Chem. Phys. 112 (2000) 8910. CrossrefGoogle Scholar

  • 70.

    S. Nosé, J. Chem. Phys. 81 (1984) 511. CrossrefGoogle Scholar

  • 71.

    W. G. Hoover, Phys. Rev. A 31 (1985) 1695. CrossrefGoogle Scholar

  • 72.

    M. Parrinello, A. Rahman, J. Appl. Phys. 52 (1981) 7182. CrossrefGoogle Scholar

  • 73.

    T. Darden, D. York, L. Pedersen, J. Chem. Phys. 98 (1993) 10089. CrossrefGoogle Scholar

  • 74.

    M. Head-Gordon, J. A. Pople, M. J. Frisch, Chem. Phys. Lett. 153 (1988) 503. CrossrefGoogle Scholar

  • 75.

    T. H. Dunning, Jr., J. Chem. Phys. 90 (1989) 1007. CrossrefGoogle Scholar

  • 76.

    M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta, F. Ogliaro, M. J. Bearpark, J. Heyd, E. N. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. P. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, N. J. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian 09, Gaussian, Inc., Wallingford, CT, USA (2009). Google Scholar

  • 77.

    N. Kosugi, H. Kuroda, Chem. Phys. Lett. 74 (1980) 490. CrossrefGoogle Scholar

  • 78.

    N. Kosugi, Theor. Chim. Acta 72 (1987) 149. CrossrefGoogle Scholar

  • 79.

    S. Huzinaga, J. Andzelm, M. Klobukowski, E. Eadzio-Andzelm, Y. Sakai, H. Tatewaki, Gausssian Basis Sets for Molecular Calculations, Elsevier, Amsterdam (1984). Google Scholar

  • 80.

    M. Nagasaka, H. Yuzawa, N. Kosugi, J. Phys. Chem. B 121 (2017) 10957. CrossrefPubMedGoogle Scholar

About the article

Received: 2017-10-16

Accepted: 2017-12-28

Published Online: 2018-01-30

Published in Print: 2018-05-24


Citation Information: Zeitschrift für Physikalische Chemie, Volume 232, Issue 5-6, Pages 705–722, ISSN (Online) 2196-7156, ISSN (Print) 0942-9352, DOI: https://doi.org/10.1515/zpch-2017-1054.

Export Citation

©2018, Masanari Nagasaka and Nobuhiro Kosugi et al., published by De Gruyter, Berlin/Boston. This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. BY-NC-ND 4.0

Citing Articles

Here you can find all Crossref-listed publications in which this article is cited. If you would like to receive automatic email messages as soon as this article is cited in other publications, simply activate the “Citation Alert” on the top of this page.

[1]
Masanari Nagasaka, Hayato Yuzawa, Kenji Mochizuki, Eckart Rühl, and Nobuhiro Kosugi
The Journal of Physical Chemistry Letters, 2018, Page 5827

Comments (0)

Please log in or register to comment.
Log in