- Email: [email protected]
water interfaces
Journal of Molecular Structure 1113 (2016) 70e78
Contents lists available at ScienceDirect
Journal of Molecular Structure journal homepage: http://www.elsevier.com/locate/molstruc
Vibrational spectra and molecular dynamics of hydrogen peroxide molecules at quartz/water interfaces Ye-qing Lv a, b, Shi-li Zheng a, Shao-na Wang a, Wen-yi Yan a, Yi Zhang a, Hao Du a, * a
National Engineering Laboratory for Hydrometallurgical Cleaner Production Technology, Key Laboratory of Green Process and Engineering, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, PR China b University of Chinese Academy of Sciences, Beijing 100049, PR China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 4 November 2015 Received in revised form 11 December 2015 Accepted 29 January 2016 Available online 1 February 2016
The influence of H2O2 on the water vibration at quartz interface was examined using sum-frequency generation (SFG) spectroscopy, and the effect of H2O2 concentration has been systematically studied. Further, the number density and radical distribution of water molecules, H2O2 molecules, and quartz surface silanol groups were calculated using molecular dynamics (MD) simulation to provide molecular level interpretation for the SFG spectra. It is concluded from this study that the hydrogen peroxide molecules prefers to donate H-bonds to the in-plane silanol groups rather than accepting H-bonds from out-of-plane silanol groups, as evidenced by the strengthening of the peak located at 3400 cm1 assigned to “liquid-like” hydrogen-bonding network. The SFG results have been supported by the MD calculation results, which demonstrate that the relative intensity of the peak located at 3400 cm1 to that of located at 3200 cm1 increases monotonously with the increase in the number of hydrogen peroxide in the first hydration shell of silanol. © 2016 Elsevier B.V. All rights reserved.
Keywords: Hydrogen peroxide Quartz Vibrational spectra SFG MD
1. Introduction Hydrogen peroxide, H2O2, is one of the environmentally friendly (‘green’) oxidants whose degradation products are oxygen and water, and has been widely used in many industrial fields, such as water treatment [1], ozone decomposition [2], heterogeneous catalytic reactions [3,4], and skin disinfection [5]. However, H2O2 is not stable due to its tendency to disproportionate exothermally into molecular oxygen and water under the influence of catalytic impurities (e.g., metal ions), elevated temperatures, pH levels, solar radiation, and airborne particles. In this regard, continuous efforts have been made to design new compounds for safe storage and handling of H2O2. Some of the more successful compounds include solid peroxohydrates, which are used mainly as bleaching agents in washing powders [6], and ureaehydrogen peroxide (1:1) adduct, which is used in hair bleaching, skin disinfection, teeth whitening, and as an efficient oxidant in organic synthesis [7,8]. Recently, a new method, solegel technique, featuring the incorporation of H2O2 into a silica xerogel matrix was successfully
* Corresponding author. E-mail address: [email protected] (H. Du). http://dx.doi.org/10.1016/j.molstruc.2016.01.085 0022-2860/© 2016 Elsevier B.V. All rights reserved.
developed for the safe handling and storage of H2O2 [9,10]. A large number of researches have been reported in terms of the stability of hydrogen peroxide in silica gel, focusing on the release and reaction activity of hydrogen peroxide [10e12]. However, limited studies have been reported with respect to the interaction between hydrogen peroxide and silica gel from the molecular level, which is of great importance for in-depth understanding of the properties of hydrogen peroxide incorporated in silica xerogel. Spectroscopic techniques including IR spectrum and Raman, as well as molecular simulations have been utilized to examine the structure of hydrogen peroxide on the surface of silica gel [9]. It is generally agreed that the fundamental structure of silica consists of three dimensional tetrahedral SiO4 units arranged in a variety of siloxane structures, both linear and cyclic, with hydrophilic hydroxyls on the surface [13]. Any hydrogen-bonding sensitive molecule close to the silica surface would be anchored to its surface by hydrogen bonds, and the hydroxyl functions, SieOH, are believed to be responsible for the adsorption properties of silica [14,15]. Owing to its favorable geometry, hydrogen peroxide may form a relatively strong hydrogen bond with the oxygen atom in the siloxane bridge, SieOeSi. Further, the presence of H2O molecules, strongly linked to H2O2, facilitate the generation of cluster geometry by forming a less strained ring, thus reinforcing the hydrogen bonding network, as
Y.-q. Lv et al. / Journal of Molecular Structure 1113 (2016) 70e78
well as the binding of hydrogen peroxide to silica surface [9]. Nevertheless, previous studies generally focus on the interpretation of interfacial structure based on indirect spectra examination due to the limitation of the experimental methods, and direct observation of H2O2-silica interaction still remains a challenge, which could be resolved by performing surface sensitive spectroscopic examination. In this regard, sum-frequency generation (SFG) spectroscopy would provide valuable information for further understanding of the influence of H2O2 on silica interfacial phenomena. SFG spectroscopy is based on a second-order nonlinear optical process that is forbidden in media with inversion symmetry, thus, it is highly interface-sensitive [16]. It has been recognized as an ideal tool for the investigation of liquid/solid interfacial phenomena [17]. Examination of hydrophilic quartz surface has provided rich information with respect to surface water structures and waterequartz interactions. It has been well established that the SFG peak located at 3200 cm1 corresponds to the well-ordered, tetrahedrally coordinated, symmetric OeH stretching mode, defined as the ice-like band because of the internal OeH and intermolecular H-bond OeH$$$O network are close to those encountered in ice. And the assignment of the 3400 cm1 band is generally attributed to more randomly oriented (asymmetric) OeH stretching vibrations (liquid-like hydrogen-bonding network) [18]. Interpretation of SFG spectra of complex system is quite challenging due to the difficulty to resolve the contribution from different molecular vibrations observed in the spectra. Molecular simulation in this regard could serve as a complimentary method for the determination of interfacial water geometries and contributions of modeled molecules to SFG spectra. The silicaewater interface had been examined using density functional theory based molecular dynamics simulations [19]. It has been pointed that the silanol groups on the water/quartz interface can be separated into two families, naming out-of-plane and in-plane. The out-of-plane silanols donate a strong and a short H-bond to a nearby water molecule, which is responsible for intensification of the 3200 cm1 vibrational band observed in SFG experiments. The in-plane silanols, however, accept one H-bond from a nearby water molecule. These water molecules are not strongly bound to the surface, and they are responsible for the liquid-like OeH band in the SFG spectrum, localizing in around the 3400 cm1 [19]. Interactions between hydrogen peroxide and water molecules have been extensively studied using computational methods, focusing on the structural of H2O2-(H2O)n complexes [20,21]. However, fundamental researches on the silica-H2O2 interactions are limited. In view of the forgoing, the purpose of this study is to systematically study the influence of H2O2 on the water vibration at quartz interface using SFG spectroscopy and molecular dynamics (MD) simulation, providing fundamental information for better understanding of the H2O2/SiO2/H2O surface interactions. 2. Methods 2.1. SFG spectroscopy A picosecond Nd:YAG laser (PL2251, Ekspla) was used to pump an optical parametric generation/optical parametric amplification/ difference frequency generation (OPG/OPA/DFG) system to generate tunable infrared radiation in the 2.3e10 mm range. The second harmonic output of the YAG (yttrium-aluminum-garnet) laser (532 nm) was the source of the visible beam. The two beams are about 4.0 and 6.0 mJ, respectively, and they were introduced to co-focus on the quartz sample surface to generate SFG signal (Fig. 1). The produced SFG beam was reflected from the quartz surface and filtered through irises and a monochromator,
71
Fig. 1. Experimental setup for SFG studies of the structure of hydrogen peroxide solution at a quartz prism surface.
eventually being detected by a photomultiplier tube. The SFG signals were normalized to the intensities of the visible and the infrared beam. The prism was fixed on top of the cell using screws as illustrated in Fig. 1, and the amount of liquid in the cell was about 2 mL. All measurements were carried out at constant temperature (22 ± 1 C). The incident angle of the visible light was about 70 , which is close to the critical angle of total reflection (qc) of quartz/ water (72 ) so that a strong surface field can be generated. The incident angle of the infrared light was set at 50 , which is far from qc in order to avoid a large change in the Fresnel factor when the IR frequency is scanned in the absorption range of the OH stretching vibration of water [22]. The output/input polarization combination for the spectra is s output, s visible input, and p infrared input, respectively. This polarization combination (ssp) is sensitive only to the vibrational mode normal to the surface of the isotropic media. The following equations express the SFG intensity (ISFG)
2 ð2Þ ISFG fcNR þ A0 expði4Þ=ðu u0 þ iG0 Þ ð2Þ
where u is the infrared frequency, cNR is the non-resonant contribution to the surface non-linear susceptibility, and u0, A0, 4, and G0 are the resonant frequency, transition amplitude, phase difference between the resonant and non-resonant terms and homogeneous width, respectively [16,23e25]. Before each experiment, the hemicylindrical fused quartz prism was cleaned in “piranha” solution (3:1 v/v concentrated H2SO4/30% H2O2; Caution: extremely oxidative and should be handled carefully) for 30 min and then rinsed thoroughly with deionized water. It was then dried at 110 C for 5 min and irradiated in a UV/ozone cleaner for 30 min. At the beginning of each experiment, the SFG spectrum of pure water/quartz interface was measured to confirm that the quartz surface was contamination free. Following that, the SFG spectra of the interface in contact with the H2O2 solutions (concentrations of 0.5 M, 1.0 M, 1.5 M, and 2.0 M) were measured sequentially from the lowest concentration to the highest concentration. For each solution with a particular concentration, the cell and quartz prism were rinsed thoroughly with the solution before spectroscopic measurements to guarantee that the bulk solution in the cell had the desired electrolyte concentration. All SFG spectra were normalized to the SFG signal reflected from a quartz crystal in order to reduce the effects of laser fluctuation, change of beam overlap, and IR absorption (~10%) of the fused quartz window. A nonlinear least-squares routine was used to fit the SFG spectra [22]. The quality of the fit was evaluated by r2 values (>0.98), and the best fits are shown as solid lines in the SFG spectra presented in the following sections.
72
Y.-q. Lv et al. / Journal of Molecular Structure 1113 (2016) 70e78
2.2. Molecules dynamic simulation A MD simulation program DL_POLY_214 [26] was used to study the structure of water and hydrogen peroxide molecules at the quartz surface under four different hydrogen peroxide concentrations (0.5 M, 1.0 M, 1.5 M, and 2.0 M). For these systems, a simple cubic cell containing water and hydrogen peroxide molecules and the quartz mineral was defined with periodic boundary conditions. The simple point charge (SPC/E) water model, incorporated with the CLAYFF force field [27] for quartz was the basis for the simulation. The intermolecular potential used in the simulations was given as a combination of the LennardeJones and the Coulomb electrostatic interactions, and can be expressed as:
D h . E 6 i 12 þ qi qj rij Upair ¼ S S 4ε sij rij sij rij i
j
where ε is the energy parameter, s is the size parameter, q is the charge, and r is the distance between species i and j. LorentzeBerthelot mixing rules were applied to calculate the potential parameters of pairs.
pffiffiffiffiffiffiffi εij ¼ εi εj sij ¼ si þ sj 2 The intermolecular potential parameters are listed in Table 1 [28e30]. The initial configuration of the quartz mineral was constructed using lattice parameters provided by American Mineralogist Crystal Structure Database. Initially, the quartz crystal was simulated in a NPT (constant number, pressure, and temperature) assemble with the pressure fixed at 0.1 Mpa and the temperature fixed at 300 K for 100 ps. After adding SPC/E water and hydrogen peroxide molecule into the system, simulation was performed in a NVT (constant number, volume, and temperature) assemble using Hoover's thermostat. The Leap-frog method with a time step of 1 fs was used to integrate the particle motion. The Ewald sum has been used to account for the electrostatic interactions [27,31,32]. A final simulation time of 1 ns (106 steps each of 1 fs) including a 500 ps equilibration period was performed. The final results were analyzed based on the production of the 500 ps simulation after the equilibration period. 3. Results and discussion 3.1. SFG The interfacial water structure at a hydrophilic quartz surface was first examined and the result is shown in Fig. 2. From Fig. 2, two broad bands of comparable intensities centered at ~3200 and ~3400 cm1 are observed. The positions of the two bands and their relative intensities are in good agreement with the results reported by Shen et al. [33e35]. The 3200 cm1 band corresponds to the
Table 1 Potential parameters for quartz/water/hydrogen peroxide interaction. Species
Charge (e)
ε (kcal/mol)
s (Å)
Ref.
Silicon Bridging oxygen Hydroxyl hydrogen Surface oxygen Water hydrogen Water oxygen Hydrogen peroxide hydrogen Hydrogen peroxide oxygen
2.10 1.05 0.425 0.95 0.41 0.82 0.4976 0.4976
1.84 106 0.1554 0 0.1554 0 0.1554 0 0.1554
3.706 3.166 0 3.166 0 3.166 0 3.166
[28] [28] [28] [28] [29] [29] [30] [30]
Fig. 2. SFG spectra of water/quartz interfaces, the solid line represents the best fit.
well-ordered, tetrahedrally coordinated, symmetric OeH stretching mode (ice-like hydrogen-bonding network), which is characteristic of perfect hexagonal arrangement found in the SFG spectrum of an ice/air interface [36]. The assignment of the 3400 cm1 band is generally attributed to more randomly oriented (asymmetric) OeH stretching vibrations (liquid-like hydrogenbonding network) [37]. This is supported by the fact that as the temperature rises, the hexagonal ice structure distorts, significantly increasing the intensity of this band [38]. Further analysis of Fig. 2 shows that there's a little peak at around 2920 cm1, which is in line with the observation reported by Bjornstrom et al. using Raman spectroscopy [39]. It has been suggested that water can exist in silica gel pores either as the socalled “bulk like” water (water droplets in the pores) or as a thin film adsorbed on the silica surface. Water molecules penetrated in the pores undergo transformation from bulk water to a thin film strongly adhering to the hydrophilic surface of the pores, creating red shift of the SiOeH stretching vibrations towards the 2950e2850 cm1 region [39]. The influence of hydrogen peroxide concentration on the SFG spectra has been studied and the results are shown in Fig. 3. From Fig. 3, it is clear that in general, the addition of H2O2 does not change the SFG spectra significantly, as evidenced by the presence of the featuring water vibrational peaks centered at ~3200 and ~3400 cm1. However, in comparison with pure water/quartz system, the intensity of the peak located at 2800e3000 cm1 increases obviously, which is due to the strong hydrogen bonding of quartz surface silanol groups to the nearby water molecules, correlating with the elongation of the SiOeH bond [39], as well as the contribution from H2O2 vibrations, in good agreement with the infrared spectrums of hydrogen peroxide reported in the literature [40e42]. Further analysis of Fig. 3 shows that the intensity of the peak located at 3400 cm1 (I2) relative to that of the peak located at 3200 cm1 (I1) becomes more substantial as the H2O2 concentration increases. In 0.5 M H2O2 solution, I1 is larger than I2, and as the concentration of hydrogen peroxide solution increases, I2 gradually increases and finally exceeds I1. The change of the relative peak intensity (I2/I1) has been summarized in Table 2. From Table 2, it can be seen that the relative peak intensity increases monotonically from 0.90 to 1.11 as the concentration of hydrogen peroxide solution increases from 0.5 to 2.0 M. This result could be understood from possible surface-induced ordering or disordering of interfacial hydrogen peroxide molecules. When the hydrogen peroxide is added, structures of water H-bond donors and acceptors in the interfacial layer will be affected, which can be seen from the SFG
Y.-q. Lv et al. / Journal of Molecular Structure 1113 (2016) 70e78
73
(a)
(b)
(c)
(d) Fig. 4. Oxygen (dotted line) and hydrogen (solid line) number density profiles of silanol groups along z-axis. The color representation of the snapshot from simulation is as follows: Si, yellow; O, red; and H, white. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article).
Fig. 3. SFG spectra of (a) 0.5 M H2O2, (b) 1.0 M H2O2, (c) 1.5 M H2O2, and (d) 2.0 M H2O2 on the quartz surface, the solid lines represent the best fits.
Table 2 The influence of H2O2 concentration on the relative peak intensity (I2/I1) at quartz surface. CH2O2 (M, mol/L)
0.5
1.0
1.5
2.0
I1 ( 103) I2 ( 103) I2/I1
2.60 2.34 0.90
2.42 2.44 1.01
2.44 2.61 1.07
2.17 2.40 1.11
spectrum shown in Fig. 3. There is an increase in the liquid-like hydrogen-bond as evidenced by the dominance of the 3400 cm1 peak. However, it is interesting to notice that the 3200 cm1 peak intensity is almost unchanged, suggesting that the amount of icelike hydrogen-bond appears less affected by the addition of hydrogen peroxide. The reasons for the change of SFG spectra with respect to the H2O2 solution concentration will be discussed in detail in the following sections. 3.2. MD simulation In order to further understand the interactions between H2O2 and quartz as well as the influence of H2O2 on interfacial hydrogen bonding structures, MD simulations have been performed. The water/quartz system was first examined, and the number density profiles of silanol groups are shown in Fig. 4. From Fig. 4, it can be seen that the surface oxygen atoms mainly locate at 10.88 Å, and the hydrogen atoms, however, have two main positions, 9.63 and 10.87 Å, suggesting that the silanol groups on the surface can be separated into two families, out-of-plane and in-plane. The configuration of surface silanol groups can be seen from
the snapshot of simulation shown in Fig. 4. It is clear that out-ofplane silanol groups extend into the water phase, with hydrogen atoms locating at 9.63 Å, while the in-plane silanol groups locate in the plane of the quartz surface with the hydrogen atoms locating at 10.87 Å [14,43e45]. The number density profiles of liquid water are shown in Fig. 5. The number density distribution is, to a good approximation, symmetric with respect to the center of the simulation cell, with a slight asymmetry to be attributed to the limited duration of the simulation run and to the limited sample size, and the results are in good agreement with the literature [19,46]. From Fig. 5, it is concluded that water is strongly adsorbed at the quartz surface, as evidenced by the presence of sharp water peaks at the vicinity of quartz surface. In order to examine the interaction between the interfacial water molecule with the quartz surface silanol groups, angle a, which donates the angle of the water dipole moment with respect to the quartz surface normal (schematically shown Fig. 6) has been analyzed, with the results summarized in Fig. 7. It is obvious from Fig. 7 that the majority of water molecules in the first hydration shell of quartz surface have two main orientations, referring to angle a with value of 53 and 130 , respectively. These two species of molecules with different dipole orientations are considered to be generated from the formation of H-bond with out-of-plane silanols (53 ) and in-plane silanols (130 ). In order to study the surface H-bonds in detail, RDFs between the water hydrogen (Hw) and surface silanol oxygen (Os), gHw-Os(r), as well as water oxygen (Ow) and silanol hydrogen (Hs), gOw-Hs(r), were investigated and the results are shown in Fig. 8, From Fig. 8, it can be seen that there are two obvious peaks in both figures, due to the H-bonds with the out-of-plane and in-plane silanols. In gHwOs(r), the position of the first peak is 1.78 Å, which stems from the water molecule donating one hydrogen atom to the in-plane silanols. The length of Hw/Os H-bond is similar to that of the Hw/Ow H-bond, which is 1.80 Å [47]. It should be noted that the H-bond
74
Y.-q. Lv et al. / Journal of Molecular Structure 1113 (2016) 70e78
Fig. 7. Water dipole moment (angle a) relative number density distribution along the surface normal in the first hydration shell of quartz surface. The solid lines represent the best fits.
Fig. 5. Oxygen (solid line) and hydrogen (dotted line) number density profiles of liquid water along z-axis in the water/quartz system. The color representation of the snapshot from simulation is as follows: Si, yellow; O, red; and H, white. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article).
distance correlates with the corresponding vibrational frequencies (according to Badger's rule). A similar correlation has been previously observed for hydroxyls in molecules and for aluminosilicate cluster calculations [48,49]. This H-bond will contribute to the 3400 cm1 vibrational band observed in SFG experiments, which had also been concluded using density functional theory based molecular dynamics simulations [19]. The distance of Ow$$$Hs is 1.62 Å, and this H-bond originates from the out-of-plane silanol groups, donating a strong and a short H-bond to a nearby water molecule. The Ow$$$Hs length, however, is similar (still shorter) to the short Ow$$$Hw of 1.74e1.75 Å for ice, according to neutron diffraction measurement [50]. As mentioned above, the H-bond distance correlates with the corresponding frequencies, therefore, this local strong H-bond network is
Fig. 6. Schematic illustration describing the orientation of the water dipole moment (a).
responsible for the 3200 cm1 vibrational band observed in SFG experiments. Based on the above analysis, the structure of water molecule at the quartz surface can be obtained and the schematic illustration of the dipole moment orientation of water molecules in the first hydration shell of quartz surface is shown in Fig. 9. On the basis of their respective covalent Ow$$$Hw and H-bond lengths, two types of interfacial water molecules H-bonded to the quartz surface can be clearly identified, which are liquid-like water and ice-like water. It can be seen that liquid-like water molecules will donate a H-bond to the in-plane surface silanols and have Hw$$$Os H-bonds of 1.78 Å and dipole moment orientation of 130 with respect to the quartz surface normal. The ice-like water molecules in contrast will accept a Hbond from the out-of-plane surface silanols and have a very short Ow$$$Hs H-bond of 1.62 Å with dipole moment orientation of 53 . The effect of H2O2 has further been studied, and the number density profiles of hydrogen peroxide molecules are shown in Fig. 10. The origin of coordinate is the position of silanol oxygen atoms. It is observed from Fig. 10 that the hydrogen peroxide molecule once approach to the quartz surface will be strongly
Fig. 8. Radical distribution functions between the surface silanol (Os-oxygen atom; Hshydrogen atom) and the water solvent (Ow-oxygen atom; Hw-hydrogen atom) in the water/quartz system. The above functions are (a) HweOs and (b) OweHs, respectively.
Y.-q. Lv et al. / Journal of Molecular Structure 1113 (2016) 70e78
75
Fig. 9. Schematic illustrations of the dipole moment orientation of water molecules in the first hydration shell of quartz surface. The color representation is as follows: Si, yellow; O, red; and H, white. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article).
Fig. 10. (a) Oxygen and (b) hydrogen number density profiles of hydrogen peroxide away from the quartz surface in the hydrogen peroxide solution/quartz system.
adsorbed, evidenced by the pronounced peaks close to the quartz surface. Further, it can be seen that as the distance away from the surface reaches 6 Å, the number density of hydrogen peroxide molecules is nearly unchanged. It is interesting to find out from Fig. 10 that as the concentration of hydrogen peroxide increases, the height of the peak close to the quartz surface increases accordingly, indicating the number of hydrogen peroxide adsorbed at the surface increases. Comparing with the number density profile of water molecules (Fig. 5), it is not difficult to find out that the hydrogen peak of hydrogen peroxide close to the quartz surface is more pronounced. Meanwhile, there is only one sharp peak in oxygen number density profile originated from hydrogen peroxide (Fig. 10(a)), suggesting that the two oxygen atoms in hydrogen peroxide molecules have the same distance away from the quartz surface. In other words, the hydrogen peroxide molecules once being adsorbed, prefer to orient themselves parallel to the surface. In order to understand how the hydrogen peroxide molecules interact with the silanols, the RDFs between the surface silanol and the water and hydrogen peroxide molecules are presented in Fig. 11. It can be seen that the RDFs in Fig. 11(a) and (b) are similar to that
have been shown in Fig. 8, suggesting that the interactions between the surface silanol and the water molecules are much less affected by the presence of hydrogen peroxide due to its limited quantity and low surface adsorption. To further confirm the effect of hydrogen peroxide to the interfacial water structures, the water dipole moment (angle a) relative number density distributions along the quartz surface normal in the first hydration shell of quartz surface under four different hydrogen peroxide concentrations (ranging from 0.5 to 2.0 M), are calculated and shown in Fig. 12. From Fig. 12, it is obvious that for all the systems studied, the presence of hydrogen peroxide molecules have negligible influence on the interfacial water molecule dipole moment orientations, which is due to the limited adsorption of hydrogen peroxide molecules relative to the water molecules. For the RDFs between the surface silanol and the hydrogen peroxide molecules, it is observed from Fig. 11(c) and (d) that the positions of the first peak are 1.83 and 1.67 Å, respectively. The first peak in Fig. 11(c) is due to the hydrogen peroxide molecules donating H-bonds to the in-plane surface silanols and the first peak in Fig. 11(d) is due to the hydrogen peroxide molecules accepting H-bonds from the out-of plane silanols. The H-bonds formed due to
76
Y.-q. Lv et al. / Journal of Molecular Structure 1113 (2016) 70e78
Fig. 11. Radical distribution functions between the surface silanol and the hydrogen peroxide (OHP-oxygen atom; HHP-hydrogen atom)and water molecules. (a) OseHw, (b) HseOw, (c) Os-HHP, and (d) Hs-OHP.
the interaction between the surface hydrogen peroxide molecules and quartz surface silanols are longer than the H-bonds between water molecules and silanols, which are 1.78 and 1.62 Å, respectively. Marilia and co-workers [51] have pointed that the OeH bonds in H2O2 will be substantially lengthened in solution (by about 0.02 Å), and the OeH bond will be weakened consequently, which is in good agreement with our results. Further, it is clear that the first peak in Fig. 11(c) is higher than the second one, while opposite result has been observed in Fig. 11(d), suggesting that the hydrogen peroxide molecules adsorbed at the quartz surface are more likely to donate H-bonds to the in-plane surface silanol groups. Previous quantum mechanics and molecular mechanics (QM/MM) study has concluded that H2O2 is a better proton donor than H2O, however a weaker proton acceptor [51]. Therefore, it is conclude from the above analysis that the hydrogen peroxide molecules prefer to donate H-bonds to the in-plane silanol groups rather than accept H-bonds from out-of-plane silanol groups; thus, contribute to the 3400 cm1 peak. Based on the above analysis, the structure of hydrogen peroxide at quartz surface can be obtained, and the schematic illustration of the interactions between hydrogen peroxide/water molecules with the surface silanols is shown in Fig. 13. It is found out that the interactions between the surface silanols and the water molecules are much less affected by the presence of hydrogen peroxide due to limited adsorption. And the adsorbed hydrogen peroxide molecules prefer to orient themselves parallel to the surface and donate their hydrogen atoms to the in-plane silanols, contributing to the 3400 cm1 peak, in agreement
with the results obtained from density functional quantum simulation [9]. Another characteristic in Fig. 11(c) and (d) is that although the position of peaks remains unchanged, the height varies with the concentration of hydrogen peroxide, caused by the increase in hydrogen peroxide molecules adsorbed at the quartz surface as evidenced in Fig. 10. The coordination numbers of different atoms around silanol groups in 0.5 M, 1.0 M, 1.5 M, and 2.0 M hydrogen peroxide solutions are calculated, and the results are summarized in Table 3. From Table 3, it can be seen that the number of water molecules in the first hydration shell of silanol groups remains nearly unchanged, and decreases only slightly with the increase in hydrogen peroxide concentration, in good agreement with the results observed in Figs. 11 and 12. Further analysis of the number of hydrogen peroxide molecules in the first hydration shell of surface silanol groups can find that the coordination number increases monotonously as the hydrogen peroxide concentration increases. It is known that hydrogen peroxide molecule, donating its hydrogen atoms to the in-plane silanols, contributes to the intensification of the 3400 cm1 peak in the SFG spectra. The correlation between the coordination number of hydrogen peroxide in the first hydration shell of silanol groups with relative peak intensity (I2/I1) from SFG experiments was analyzed and the results are presented in Fig. 14. It is obvious I2/I1 increases monotonously with the molecular numbers of hydrogen peroxide in the first hydration shell of silanol groups, which donates hydrogen atoms to in-plane silanols and contributes to the 3400 cm1 intensity in the SFG spectra.
Y.-q. Lv et al. / Journal of Molecular Structure 1113 (2016) 70e78
77
Table 3 The coordination number of different atoms around silanol groups in 0.5 M, 1.0 M, 1.5 M, and 2.0 M hydrogen peroxide solutions. C (mol/L)
n (the number of molecules in the first hydration shell) OseHHP
0.5 1.0 1.5 2.0
Fig. 12. Water dipole moment (angle a) relative number density distribution along the surface normal in the first hydration shell of quartz surface under four different hydrogen peroxide concentrations, ranging from 0.5 to 2.0 M. The solid lines represent the best fits.
4. Conclusion In this study, the molecular structure of hydrogen peroxide molecules at the silica surface has been examined using SFG
4.35 5.46 9.88 1.09
OseHw 3
10 103 103 102
2.60 2.60 2.61 2.55
HseOHP 1
10 101 101 101
1.27 1.98 2.60 2.88
HseOw 3
10 103 103 103
2.55 2.55 2.54 2.54
101 101 101 101
Fig. 14. The change of relative intensity (I2/I1) with the coordination number of hydrogen peroxide in the first hydration shell of silanol groups.
spectroscopy, and it is observed that as the concentration of hydrogen peroxide solution increases, there is an increase in the intensity of 3400 cm1 peak; however, the intensity of 3200 cm1 peak remains nearly unchanged. Molecular dynamics simulation study further reveals that the surface silanol groups can be separated into two families, out-ofplane and in-plane. The out-of-plane silanols donate a strongly
Fig. 13. Schematic illustration of the interactions between hydrogen peroxide and water molecules and surface silanols. The color representation is as follows: Si, yellow; O, red; and H, white. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article).
78
Y.-q. Lv et al. / Journal of Molecular Structure 1113 (2016) 70e78
hydrogen-bond to the nearby water molecule with the dipole moment orientation of 53 with respect to the surface normal, attributing to the 3200 cm1 peak intensity in the SFG spectra. The in-plane silanols, in contrast, accept a weakly hydrogen-bond from the nearby water molecule and the dipole moment orientation of water molecule centered at 130 , attributing to the 3400 cm1 peak intensity in the SFG spectra. Because H2O2 is a better proton donor than H2O but a weaker proton acceptor, and consequently, the H-bonds between the hydrogen peroxide molecules and silanol groups is longer than that of the water/silanol H-bonds. As the hydrogen peroxide is added to the water/quartz system, the hydrogen peroxide molecules prefer to parallel to the surface and donate H-bonds to the in-plane silanol groups rather than accept H-bonds from out-of-plane silanol groups, and thus, contribute to the water vibrational peak located at 3400 cm1. Acknowledgments We gratefully acknowledge the financial support from the Major State Basic Research Development Program of China (973 program) under grant No. 2013CB632601 and 2013CB632605, National Natural Science Foundation of China under grant No. 51274178 and 51404227. References [1] R.A. Patil, S.B. Kausley, P.L. Balkunde, et al., J. Water Health 11 (2013) 443e456. [2] Am Water Works Res F, Langlais B, Reckhow D A, et al., CRC Press. 1991. € m, J. Colloid Interface Sci. 130 (1989) [3] P. Tengvall, H. Elwing, I. Lundstro 405e413. [4] I.I. Salame, T.J. Bandosz, J. Colloid Interface Sci. 210 (1999) 367e374. [5] J.C. Kim, K.U. Lee, W.C. Shin, et al., Colloids Surfaces B Biointerfaces 36 (2004) 161e166. [6] A. McKillop, W.R. Sanderson, Tetrahedron 51 (1995) 6145e6166. [7] H.E. Strassler, W. Scherer, J.R. Calamia, N. Y. State Dent. J. 58 (1992) 30e35. [8] R.S. Varma, K.P. Naicker, Org. Lett. 1 (1999) 189e192. _ [9] J. Zeglinski, G.P. Piotrowski, R. Pie˛ kos, J. Mol. Struct. 794 (2006) 83e91. _ ski, A. Cabaj, M. Strankowski, et al., Colloids Surfaces B Biointerfaces [10] J. Zegli n 54 (2007) 165e172. [11] S. Bednarz, B. Rys, D. Bogdał, Molecules 17 (2012) 8068e8078. [12] F. Sudur, B. Pleskowicz, N. Orbey, Ind. Eng. Chem. Res. 54 (2015) 1930e1940.
[13] G.M. Rignanese, J.C. Charlier, X. Gonze, Phys. Chem. Chem. Phys. 6 (2004) 1920e1925. [14] A.V. Bandura, J.D. Kubicki, J.O. Sofo, J. Phys. Chem. C 115 (2011) 5756e5766. [15] J. Rubio, J.A. Kitchener, J. Colloid Interface Sci. 57 (1976) 132e142. [16] Y.R. Shen, Nature 337 (1989) 519e525. [17] Q. Du, R. Superfine, E. Freysz, et al., Phys. Rev. Lett. 70 (1993) 2313. [18] V. Ostroverkhov, G.A. Waychunas, Y.R. Shen, Chem. Phys. Lett. 386 (2004) 144e148. [19] M. Sulpizi, M.P. Gaigeot, M. Sprik, J. Chem. Theory Comput. 8 (2012) 1037e1047. [20] A.D. Kulkarni, R.K. Pathak, L.J. Bartolotti, J. Chem. Phys. 124 (2006), 1e214309. [21] A.D. Kulkarni, R.K. Pathak, L.J. Bartolotti, J. Phys. Chem. A 109 (2005) 4583e4590. [22] S. Ye, S. Nihonyanagi, K. Uosaki, Phys. Chem. Chem. Phys. 3 (2001) 3463e3469. [23] F. Vidal, A. Tadjeddine, Rep. Prog. Phys. 68 (2005) 1095. [24] G.L. Richmond, Chem. Rev. 102 (2002) 2693e2724. [25] S. Gopalakrishnan, D. Liu, H.C. Allen, et al., Chem. Rev. 106 (2006) 1155e1175. [26] W. Smith, T.R. Forester, J. Mol. Graph. 14 (1996) 136e141. [27] R.J. Kirkpatrick, A.G. Kalinichev, X. Hou, et al., Mater. Struct. 38 (2005) 449e458. [28] R.T. Cygan, J.-J. Liang, A.G. Kalinichev, J. Phys. Chem. B 108 (2004) 1255. [29] H. J. C. Berendsen, J. P. M. P., W. F. Gunsteren, J. Hermans, Reidel: Dordrecht. (1981) 331. [30] E. Lindahl, B. Hess, D. van der Spoel, J. Mol. Model 7 (2001) 306. [31] D.M. York, T.A. Darden, L.G. Pedersen, J. Chem. Phys. 99 (1993) 8345e8348. [32] X. Wang, J. Liu, H. Du, et al., Langmuir 26 (2009) 3407e3414. [33] Q. Du, E. Freysz, Y.R. Shen, Phys. Rev. Lett. 72 (1994) 238. [34] V. Ostroverkhov, G.A. Waychunas, Y.R. Shen, Phys. Rev. Lett. 94 (2005) 046102. [35] V. Ostroverkhov, G.A. Waychunas, Y.R. Shen, Chem. Phys. Lett. 386 (2004) 144. [36] M.J. Wojcik, V. Buch, J.P. Devlin, J. Chem. Phys. 99 (1993) 2332e2344. [37] Segelstein D J, University of Missouri-Kansas City. 2011. [38] X. Wei, P.B. Miranda, C. Zhang, Y.R. Shen, Phys. Rev. B 66 (2002) 085401. [39] J. Bjornstrom, A. Martinelli, J.R.T. Johnson, et al., Chem. Phys. Lett. 380 (2003) 165e172. re, J. Chem. Phys. 18 (1950) 88e92. [40] P.A. Gigue [41] M.J. Loeffler, B.D. Teolis, R.A. Baragiola, J. Phys. Chem. B 110 (2006) 6911e6915. [42] M.J. Loeffler, R.A. Baragiola, J. Phys. Chem. A 115 (2011) 5324e5328. [43] T.P.M. Goumans, A. Wander, W.A. Brown, et al., Phys. Chem. Chem. Phys. 9 (2007) 2146e2152. [44] J. Yang, E.G. Wang, Phys. Rev. B 73 (2006) 035406. [45] Y.W. Chen, I.H. Chu, Y. Wang, et al., Phys. Rev. B 84 (2011) 155444. [46] S.H. Lee, P.J. Rossky, J. Chem. Phys. 100 (1994) 3334e3345. [47] G. Sutmann, R. Vallauri, J. Mol. Liq. 98 (2002) 215e226. [48] Novak A, Springer Berlin Heidelberg. (1974) 177e216. [49] J.D. Kubicki, D. Sykes, G.R. Rossman, Phys. Chem. Miner. 20 (1993) 425e432. [50] S.W. Peterson, H.A. Levy, Acta Crystallogr. 10 (1957) 70e76. pez, Chem. Phys. 332 (2007) 341e347. [51] M.T.C. Martins-Costa, M.F. Ruiz-Lo
Contents lists available at ScienceDirect
Journal of Molecular Structure journal homepage: http://www.elsevier.com/locate/molstruc
Vibrational spectra and molecular dynamics of hydrogen peroxide molecules at quartz/water interfaces Ye-qing Lv a, b, Shi-li Zheng a, Shao-na Wang a, Wen-yi Yan a, Yi Zhang a, Hao Du a, * a
National Engineering Laboratory for Hydrometallurgical Cleaner Production Technology, Key Laboratory of Green Process and Engineering, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, PR China b University of Chinese Academy of Sciences, Beijing 100049, PR China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 4 November 2015 Received in revised form 11 December 2015 Accepted 29 January 2016 Available online 1 February 2016
The influence of H2O2 on the water vibration at quartz interface was examined using sum-frequency generation (SFG) spectroscopy, and the effect of H2O2 concentration has been systematically studied. Further, the number density and radical distribution of water molecules, H2O2 molecules, and quartz surface silanol groups were calculated using molecular dynamics (MD) simulation to provide molecular level interpretation for the SFG spectra. It is concluded from this study that the hydrogen peroxide molecules prefers to donate H-bonds to the in-plane silanol groups rather than accepting H-bonds from out-of-plane silanol groups, as evidenced by the strengthening of the peak located at 3400 cm1 assigned to “liquid-like” hydrogen-bonding network. The SFG results have been supported by the MD calculation results, which demonstrate that the relative intensity of the peak located at 3400 cm1 to that of located at 3200 cm1 increases monotonously with the increase in the number of hydrogen peroxide in the first hydration shell of silanol. © 2016 Elsevier B.V. All rights reserved.
Keywords: Hydrogen peroxide Quartz Vibrational spectra SFG MD
1. Introduction Hydrogen peroxide, H2O2, is one of the environmentally friendly (‘green’) oxidants whose degradation products are oxygen and water, and has been widely used in many industrial fields, such as water treatment [1], ozone decomposition [2], heterogeneous catalytic reactions [3,4], and skin disinfection [5]. However, H2O2 is not stable due to its tendency to disproportionate exothermally into molecular oxygen and water under the influence of catalytic impurities (e.g., metal ions), elevated temperatures, pH levels, solar radiation, and airborne particles. In this regard, continuous efforts have been made to design new compounds for safe storage and handling of H2O2. Some of the more successful compounds include solid peroxohydrates, which are used mainly as bleaching agents in washing powders [6], and ureaehydrogen peroxide (1:1) adduct, which is used in hair bleaching, skin disinfection, teeth whitening, and as an efficient oxidant in organic synthesis [7,8]. Recently, a new method, solegel technique, featuring the incorporation of H2O2 into a silica xerogel matrix was successfully
* Corresponding author. E-mail address: [email protected] (H. Du). http://dx.doi.org/10.1016/j.molstruc.2016.01.085 0022-2860/© 2016 Elsevier B.V. All rights reserved.
developed for the safe handling and storage of H2O2 [9,10]. A large number of researches have been reported in terms of the stability of hydrogen peroxide in silica gel, focusing on the release and reaction activity of hydrogen peroxide [10e12]. However, limited studies have been reported with respect to the interaction between hydrogen peroxide and silica gel from the molecular level, which is of great importance for in-depth understanding of the properties of hydrogen peroxide incorporated in silica xerogel. Spectroscopic techniques including IR spectrum and Raman, as well as molecular simulations have been utilized to examine the structure of hydrogen peroxide on the surface of silica gel [9]. It is generally agreed that the fundamental structure of silica consists of three dimensional tetrahedral SiO4 units arranged in a variety of siloxane structures, both linear and cyclic, with hydrophilic hydroxyls on the surface [13]. Any hydrogen-bonding sensitive molecule close to the silica surface would be anchored to its surface by hydrogen bonds, and the hydroxyl functions, SieOH, are believed to be responsible for the adsorption properties of silica [14,15]. Owing to its favorable geometry, hydrogen peroxide may form a relatively strong hydrogen bond with the oxygen atom in the siloxane bridge, SieOeSi. Further, the presence of H2O molecules, strongly linked to H2O2, facilitate the generation of cluster geometry by forming a less strained ring, thus reinforcing the hydrogen bonding network, as
Y.-q. Lv et al. / Journal of Molecular Structure 1113 (2016) 70e78
well as the binding of hydrogen peroxide to silica surface [9]. Nevertheless, previous studies generally focus on the interpretation of interfacial structure based on indirect spectra examination due to the limitation of the experimental methods, and direct observation of H2O2-silica interaction still remains a challenge, which could be resolved by performing surface sensitive spectroscopic examination. In this regard, sum-frequency generation (SFG) spectroscopy would provide valuable information for further understanding of the influence of H2O2 on silica interfacial phenomena. SFG spectroscopy is based on a second-order nonlinear optical process that is forbidden in media with inversion symmetry, thus, it is highly interface-sensitive [16]. It has been recognized as an ideal tool for the investigation of liquid/solid interfacial phenomena [17]. Examination of hydrophilic quartz surface has provided rich information with respect to surface water structures and waterequartz interactions. It has been well established that the SFG peak located at 3200 cm1 corresponds to the well-ordered, tetrahedrally coordinated, symmetric OeH stretching mode, defined as the ice-like band because of the internal OeH and intermolecular H-bond OeH$$$O network are close to those encountered in ice. And the assignment of the 3400 cm1 band is generally attributed to more randomly oriented (asymmetric) OeH stretching vibrations (liquid-like hydrogen-bonding network) [18]. Interpretation of SFG spectra of complex system is quite challenging due to the difficulty to resolve the contribution from different molecular vibrations observed in the spectra. Molecular simulation in this regard could serve as a complimentary method for the determination of interfacial water geometries and contributions of modeled molecules to SFG spectra. The silicaewater interface had been examined using density functional theory based molecular dynamics simulations [19]. It has been pointed that the silanol groups on the water/quartz interface can be separated into two families, naming out-of-plane and in-plane. The out-of-plane silanols donate a strong and a short H-bond to a nearby water molecule, which is responsible for intensification of the 3200 cm1 vibrational band observed in SFG experiments. The in-plane silanols, however, accept one H-bond from a nearby water molecule. These water molecules are not strongly bound to the surface, and they are responsible for the liquid-like OeH band in the SFG spectrum, localizing in around the 3400 cm1 [19]. Interactions between hydrogen peroxide and water molecules have been extensively studied using computational methods, focusing on the structural of H2O2-(H2O)n complexes [20,21]. However, fundamental researches on the silica-H2O2 interactions are limited. In view of the forgoing, the purpose of this study is to systematically study the influence of H2O2 on the water vibration at quartz interface using SFG spectroscopy and molecular dynamics (MD) simulation, providing fundamental information for better understanding of the H2O2/SiO2/H2O surface interactions. 2. Methods 2.1. SFG spectroscopy A picosecond Nd:YAG laser (PL2251, Ekspla) was used to pump an optical parametric generation/optical parametric amplification/ difference frequency generation (OPG/OPA/DFG) system to generate tunable infrared radiation in the 2.3e10 mm range. The second harmonic output of the YAG (yttrium-aluminum-garnet) laser (532 nm) was the source of the visible beam. The two beams are about 4.0 and 6.0 mJ, respectively, and they were introduced to co-focus on the quartz sample surface to generate SFG signal (Fig. 1). The produced SFG beam was reflected from the quartz surface and filtered through irises and a monochromator,
71
Fig. 1. Experimental setup for SFG studies of the structure of hydrogen peroxide solution at a quartz prism surface.
eventually being detected by a photomultiplier tube. The SFG signals were normalized to the intensities of the visible and the infrared beam. The prism was fixed on top of the cell using screws as illustrated in Fig. 1, and the amount of liquid in the cell was about 2 mL. All measurements were carried out at constant temperature (22 ± 1 C). The incident angle of the visible light was about 70 , which is close to the critical angle of total reflection (qc) of quartz/ water (72 ) so that a strong surface field can be generated. The incident angle of the infrared light was set at 50 , which is far from qc in order to avoid a large change in the Fresnel factor when the IR frequency is scanned in the absorption range of the OH stretching vibration of water [22]. The output/input polarization combination for the spectra is s output, s visible input, and p infrared input, respectively. This polarization combination (ssp) is sensitive only to the vibrational mode normal to the surface of the isotropic media. The following equations express the SFG intensity (ISFG)
2 ð2Þ ISFG fcNR þ A0 expði4Þ=ðu u0 þ iG0 Þ ð2Þ
where u is the infrared frequency, cNR is the non-resonant contribution to the surface non-linear susceptibility, and u0, A0, 4, and G0 are the resonant frequency, transition amplitude, phase difference between the resonant and non-resonant terms and homogeneous width, respectively [16,23e25]. Before each experiment, the hemicylindrical fused quartz prism was cleaned in “piranha” solution (3:1 v/v concentrated H2SO4/30% H2O2; Caution: extremely oxidative and should be handled carefully) for 30 min and then rinsed thoroughly with deionized water. It was then dried at 110 C for 5 min and irradiated in a UV/ozone cleaner for 30 min. At the beginning of each experiment, the SFG spectrum of pure water/quartz interface was measured to confirm that the quartz surface was contamination free. Following that, the SFG spectra of the interface in contact with the H2O2 solutions (concentrations of 0.5 M, 1.0 M, 1.5 M, and 2.0 M) were measured sequentially from the lowest concentration to the highest concentration. For each solution with a particular concentration, the cell and quartz prism were rinsed thoroughly with the solution before spectroscopic measurements to guarantee that the bulk solution in the cell had the desired electrolyte concentration. All SFG spectra were normalized to the SFG signal reflected from a quartz crystal in order to reduce the effects of laser fluctuation, change of beam overlap, and IR absorption (~10%) of the fused quartz window. A nonlinear least-squares routine was used to fit the SFG spectra [22]. The quality of the fit was evaluated by r2 values (>0.98), and the best fits are shown as solid lines in the SFG spectra presented in the following sections.
72
Y.-q. Lv et al. / Journal of Molecular Structure 1113 (2016) 70e78
2.2. Molecules dynamic simulation A MD simulation program DL_POLY_214 [26] was used to study the structure of water and hydrogen peroxide molecules at the quartz surface under four different hydrogen peroxide concentrations (0.5 M, 1.0 M, 1.5 M, and 2.0 M). For these systems, a simple cubic cell containing water and hydrogen peroxide molecules and the quartz mineral was defined with periodic boundary conditions. The simple point charge (SPC/E) water model, incorporated with the CLAYFF force field [27] for quartz was the basis for the simulation. The intermolecular potential used in the simulations was given as a combination of the LennardeJones and the Coulomb electrostatic interactions, and can be expressed as:
D h . E 6 i 12 þ qi qj rij Upair ¼ S S 4ε sij rij sij rij i
j
where ε is the energy parameter, s is the size parameter, q is the charge, and r is the distance between species i and j. LorentzeBerthelot mixing rules were applied to calculate the potential parameters of pairs.
pffiffiffiffiffiffiffi εij ¼ εi εj sij ¼ si þ sj 2 The intermolecular potential parameters are listed in Table 1 [28e30]. The initial configuration of the quartz mineral was constructed using lattice parameters provided by American Mineralogist Crystal Structure Database. Initially, the quartz crystal was simulated in a NPT (constant number, pressure, and temperature) assemble with the pressure fixed at 0.1 Mpa and the temperature fixed at 300 K for 100 ps. After adding SPC/E water and hydrogen peroxide molecule into the system, simulation was performed in a NVT (constant number, volume, and temperature) assemble using Hoover's thermostat. The Leap-frog method with a time step of 1 fs was used to integrate the particle motion. The Ewald sum has been used to account for the electrostatic interactions [27,31,32]. A final simulation time of 1 ns (106 steps each of 1 fs) including a 500 ps equilibration period was performed. The final results were analyzed based on the production of the 500 ps simulation after the equilibration period. 3. Results and discussion 3.1. SFG The interfacial water structure at a hydrophilic quartz surface was first examined and the result is shown in Fig. 2. From Fig. 2, two broad bands of comparable intensities centered at ~3200 and ~3400 cm1 are observed. The positions of the two bands and their relative intensities are in good agreement with the results reported by Shen et al. [33e35]. The 3200 cm1 band corresponds to the
Table 1 Potential parameters for quartz/water/hydrogen peroxide interaction. Species
Charge (e)
ε (kcal/mol)
s (Å)
Ref.
Silicon Bridging oxygen Hydroxyl hydrogen Surface oxygen Water hydrogen Water oxygen Hydrogen peroxide hydrogen Hydrogen peroxide oxygen
2.10 1.05 0.425 0.95 0.41 0.82 0.4976 0.4976
1.84 106 0.1554 0 0.1554 0 0.1554 0 0.1554
3.706 3.166 0 3.166 0 3.166 0 3.166
[28] [28] [28] [28] [29] [29] [30] [30]
Fig. 2. SFG spectra of water/quartz interfaces, the solid line represents the best fit.
well-ordered, tetrahedrally coordinated, symmetric OeH stretching mode (ice-like hydrogen-bonding network), which is characteristic of perfect hexagonal arrangement found in the SFG spectrum of an ice/air interface [36]. The assignment of the 3400 cm1 band is generally attributed to more randomly oriented (asymmetric) OeH stretching vibrations (liquid-like hydrogenbonding network) [37]. This is supported by the fact that as the temperature rises, the hexagonal ice structure distorts, significantly increasing the intensity of this band [38]. Further analysis of Fig. 2 shows that there's a little peak at around 2920 cm1, which is in line with the observation reported by Bjornstrom et al. using Raman spectroscopy [39]. It has been suggested that water can exist in silica gel pores either as the socalled “bulk like” water (water droplets in the pores) or as a thin film adsorbed on the silica surface. Water molecules penetrated in the pores undergo transformation from bulk water to a thin film strongly adhering to the hydrophilic surface of the pores, creating red shift of the SiOeH stretching vibrations towards the 2950e2850 cm1 region [39]. The influence of hydrogen peroxide concentration on the SFG spectra has been studied and the results are shown in Fig. 3. From Fig. 3, it is clear that in general, the addition of H2O2 does not change the SFG spectra significantly, as evidenced by the presence of the featuring water vibrational peaks centered at ~3200 and ~3400 cm1. However, in comparison with pure water/quartz system, the intensity of the peak located at 2800e3000 cm1 increases obviously, which is due to the strong hydrogen bonding of quartz surface silanol groups to the nearby water molecules, correlating with the elongation of the SiOeH bond [39], as well as the contribution from H2O2 vibrations, in good agreement with the infrared spectrums of hydrogen peroxide reported in the literature [40e42]. Further analysis of Fig. 3 shows that the intensity of the peak located at 3400 cm1 (I2) relative to that of the peak located at 3200 cm1 (I1) becomes more substantial as the H2O2 concentration increases. In 0.5 M H2O2 solution, I1 is larger than I2, and as the concentration of hydrogen peroxide solution increases, I2 gradually increases and finally exceeds I1. The change of the relative peak intensity (I2/I1) has been summarized in Table 2. From Table 2, it can be seen that the relative peak intensity increases monotonically from 0.90 to 1.11 as the concentration of hydrogen peroxide solution increases from 0.5 to 2.0 M. This result could be understood from possible surface-induced ordering or disordering of interfacial hydrogen peroxide molecules. When the hydrogen peroxide is added, structures of water H-bond donors and acceptors in the interfacial layer will be affected, which can be seen from the SFG
Y.-q. Lv et al. / Journal of Molecular Structure 1113 (2016) 70e78
73
(a)
(b)
(c)
(d) Fig. 4. Oxygen (dotted line) and hydrogen (solid line) number density profiles of silanol groups along z-axis. The color representation of the snapshot from simulation is as follows: Si, yellow; O, red; and H, white. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article).
Fig. 3. SFG spectra of (a) 0.5 M H2O2, (b) 1.0 M H2O2, (c) 1.5 M H2O2, and (d) 2.0 M H2O2 on the quartz surface, the solid lines represent the best fits.
Table 2 The influence of H2O2 concentration on the relative peak intensity (I2/I1) at quartz surface. CH2O2 (M, mol/L)
0.5
1.0
1.5
2.0
I1 ( 103) I2 ( 103) I2/I1
2.60 2.34 0.90
2.42 2.44 1.01
2.44 2.61 1.07
2.17 2.40 1.11
spectrum shown in Fig. 3. There is an increase in the liquid-like hydrogen-bond as evidenced by the dominance of the 3400 cm1 peak. However, it is interesting to notice that the 3200 cm1 peak intensity is almost unchanged, suggesting that the amount of icelike hydrogen-bond appears less affected by the addition of hydrogen peroxide. The reasons for the change of SFG spectra with respect to the H2O2 solution concentration will be discussed in detail in the following sections. 3.2. MD simulation In order to further understand the interactions between H2O2 and quartz as well as the influence of H2O2 on interfacial hydrogen bonding structures, MD simulations have been performed. The water/quartz system was first examined, and the number density profiles of silanol groups are shown in Fig. 4. From Fig. 4, it can be seen that the surface oxygen atoms mainly locate at 10.88 Å, and the hydrogen atoms, however, have two main positions, 9.63 and 10.87 Å, suggesting that the silanol groups on the surface can be separated into two families, out-of-plane and in-plane. The configuration of surface silanol groups can be seen from
the snapshot of simulation shown in Fig. 4. It is clear that out-ofplane silanol groups extend into the water phase, with hydrogen atoms locating at 9.63 Å, while the in-plane silanol groups locate in the plane of the quartz surface with the hydrogen atoms locating at 10.87 Å [14,43e45]. The number density profiles of liquid water are shown in Fig. 5. The number density distribution is, to a good approximation, symmetric with respect to the center of the simulation cell, with a slight asymmetry to be attributed to the limited duration of the simulation run and to the limited sample size, and the results are in good agreement with the literature [19,46]. From Fig. 5, it is concluded that water is strongly adsorbed at the quartz surface, as evidenced by the presence of sharp water peaks at the vicinity of quartz surface. In order to examine the interaction between the interfacial water molecule with the quartz surface silanol groups, angle a, which donates the angle of the water dipole moment with respect to the quartz surface normal (schematically shown Fig. 6) has been analyzed, with the results summarized in Fig. 7. It is obvious from Fig. 7 that the majority of water molecules in the first hydration shell of quartz surface have two main orientations, referring to angle a with value of 53 and 130 , respectively. These two species of molecules with different dipole orientations are considered to be generated from the formation of H-bond with out-of-plane silanols (53 ) and in-plane silanols (130 ). In order to study the surface H-bonds in detail, RDFs between the water hydrogen (Hw) and surface silanol oxygen (Os), gHw-Os(r), as well as water oxygen (Ow) and silanol hydrogen (Hs), gOw-Hs(r), were investigated and the results are shown in Fig. 8, From Fig. 8, it can be seen that there are two obvious peaks in both figures, due to the H-bonds with the out-of-plane and in-plane silanols. In gHwOs(r), the position of the first peak is 1.78 Å, which stems from the water molecule donating one hydrogen atom to the in-plane silanols. The length of Hw/Os H-bond is similar to that of the Hw/Ow H-bond, which is 1.80 Å [47]. It should be noted that the H-bond
74
Y.-q. Lv et al. / Journal of Molecular Structure 1113 (2016) 70e78
Fig. 7. Water dipole moment (angle a) relative number density distribution along the surface normal in the first hydration shell of quartz surface. The solid lines represent the best fits.
Fig. 5. Oxygen (solid line) and hydrogen (dotted line) number density profiles of liquid water along z-axis in the water/quartz system. The color representation of the snapshot from simulation is as follows: Si, yellow; O, red; and H, white. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article).
distance correlates with the corresponding vibrational frequencies (according to Badger's rule). A similar correlation has been previously observed for hydroxyls in molecules and for aluminosilicate cluster calculations [48,49]. This H-bond will contribute to the 3400 cm1 vibrational band observed in SFG experiments, which had also been concluded using density functional theory based molecular dynamics simulations [19]. The distance of Ow$$$Hs is 1.62 Å, and this H-bond originates from the out-of-plane silanol groups, donating a strong and a short H-bond to a nearby water molecule. The Ow$$$Hs length, however, is similar (still shorter) to the short Ow$$$Hw of 1.74e1.75 Å for ice, according to neutron diffraction measurement [50]. As mentioned above, the H-bond distance correlates with the corresponding frequencies, therefore, this local strong H-bond network is
Fig. 6. Schematic illustration describing the orientation of the water dipole moment (a).
responsible for the 3200 cm1 vibrational band observed in SFG experiments. Based on the above analysis, the structure of water molecule at the quartz surface can be obtained and the schematic illustration of the dipole moment orientation of water molecules in the first hydration shell of quartz surface is shown in Fig. 9. On the basis of their respective covalent Ow$$$Hw and H-bond lengths, two types of interfacial water molecules H-bonded to the quartz surface can be clearly identified, which are liquid-like water and ice-like water. It can be seen that liquid-like water molecules will donate a H-bond to the in-plane surface silanols and have Hw$$$Os H-bonds of 1.78 Å and dipole moment orientation of 130 with respect to the quartz surface normal. The ice-like water molecules in contrast will accept a Hbond from the out-of-plane surface silanols and have a very short Ow$$$Hs H-bond of 1.62 Å with dipole moment orientation of 53 . The effect of H2O2 has further been studied, and the number density profiles of hydrogen peroxide molecules are shown in Fig. 10. The origin of coordinate is the position of silanol oxygen atoms. It is observed from Fig. 10 that the hydrogen peroxide molecule once approach to the quartz surface will be strongly
Fig. 8. Radical distribution functions between the surface silanol (Os-oxygen atom; Hshydrogen atom) and the water solvent (Ow-oxygen atom; Hw-hydrogen atom) in the water/quartz system. The above functions are (a) HweOs and (b) OweHs, respectively.
Y.-q. Lv et al. / Journal of Molecular Structure 1113 (2016) 70e78
75
Fig. 9. Schematic illustrations of the dipole moment orientation of water molecules in the first hydration shell of quartz surface. The color representation is as follows: Si, yellow; O, red; and H, white. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article).
Fig. 10. (a) Oxygen and (b) hydrogen number density profiles of hydrogen peroxide away from the quartz surface in the hydrogen peroxide solution/quartz system.
adsorbed, evidenced by the pronounced peaks close to the quartz surface. Further, it can be seen that as the distance away from the surface reaches 6 Å, the number density of hydrogen peroxide molecules is nearly unchanged. It is interesting to find out from Fig. 10 that as the concentration of hydrogen peroxide increases, the height of the peak close to the quartz surface increases accordingly, indicating the number of hydrogen peroxide adsorbed at the surface increases. Comparing with the number density profile of water molecules (Fig. 5), it is not difficult to find out that the hydrogen peak of hydrogen peroxide close to the quartz surface is more pronounced. Meanwhile, there is only one sharp peak in oxygen number density profile originated from hydrogen peroxide (Fig. 10(a)), suggesting that the two oxygen atoms in hydrogen peroxide molecules have the same distance away from the quartz surface. In other words, the hydrogen peroxide molecules once being adsorbed, prefer to orient themselves parallel to the surface. In order to understand how the hydrogen peroxide molecules interact with the silanols, the RDFs between the surface silanol and the water and hydrogen peroxide molecules are presented in Fig. 11. It can be seen that the RDFs in Fig. 11(a) and (b) are similar to that
have been shown in Fig. 8, suggesting that the interactions between the surface silanol and the water molecules are much less affected by the presence of hydrogen peroxide due to its limited quantity and low surface adsorption. To further confirm the effect of hydrogen peroxide to the interfacial water structures, the water dipole moment (angle a) relative number density distributions along the quartz surface normal in the first hydration shell of quartz surface under four different hydrogen peroxide concentrations (ranging from 0.5 to 2.0 M), are calculated and shown in Fig. 12. From Fig. 12, it is obvious that for all the systems studied, the presence of hydrogen peroxide molecules have negligible influence on the interfacial water molecule dipole moment orientations, which is due to the limited adsorption of hydrogen peroxide molecules relative to the water molecules. For the RDFs between the surface silanol and the hydrogen peroxide molecules, it is observed from Fig. 11(c) and (d) that the positions of the first peak are 1.83 and 1.67 Å, respectively. The first peak in Fig. 11(c) is due to the hydrogen peroxide molecules donating H-bonds to the in-plane surface silanols and the first peak in Fig. 11(d) is due to the hydrogen peroxide molecules accepting H-bonds from the out-of plane silanols. The H-bonds formed due to
76
Y.-q. Lv et al. / Journal of Molecular Structure 1113 (2016) 70e78
Fig. 11. Radical distribution functions between the surface silanol and the hydrogen peroxide (OHP-oxygen atom; HHP-hydrogen atom)and water molecules. (a) OseHw, (b) HseOw, (c) Os-HHP, and (d) Hs-OHP.
the interaction between the surface hydrogen peroxide molecules and quartz surface silanols are longer than the H-bonds between water molecules and silanols, which are 1.78 and 1.62 Å, respectively. Marilia and co-workers [51] have pointed that the OeH bonds in H2O2 will be substantially lengthened in solution (by about 0.02 Å), and the OeH bond will be weakened consequently, which is in good agreement with our results. Further, it is clear that the first peak in Fig. 11(c) is higher than the second one, while opposite result has been observed in Fig. 11(d), suggesting that the hydrogen peroxide molecules adsorbed at the quartz surface are more likely to donate H-bonds to the in-plane surface silanol groups. Previous quantum mechanics and molecular mechanics (QM/MM) study has concluded that H2O2 is a better proton donor than H2O, however a weaker proton acceptor [51]. Therefore, it is conclude from the above analysis that the hydrogen peroxide molecules prefer to donate H-bonds to the in-plane silanol groups rather than accept H-bonds from out-of-plane silanol groups; thus, contribute to the 3400 cm1 peak. Based on the above analysis, the structure of hydrogen peroxide at quartz surface can be obtained, and the schematic illustration of the interactions between hydrogen peroxide/water molecules with the surface silanols is shown in Fig. 13. It is found out that the interactions between the surface silanols and the water molecules are much less affected by the presence of hydrogen peroxide due to limited adsorption. And the adsorbed hydrogen peroxide molecules prefer to orient themselves parallel to the surface and donate their hydrogen atoms to the in-plane silanols, contributing to the 3400 cm1 peak, in agreement
with the results obtained from density functional quantum simulation [9]. Another characteristic in Fig. 11(c) and (d) is that although the position of peaks remains unchanged, the height varies with the concentration of hydrogen peroxide, caused by the increase in hydrogen peroxide molecules adsorbed at the quartz surface as evidenced in Fig. 10. The coordination numbers of different atoms around silanol groups in 0.5 M, 1.0 M, 1.5 M, and 2.0 M hydrogen peroxide solutions are calculated, and the results are summarized in Table 3. From Table 3, it can be seen that the number of water molecules in the first hydration shell of silanol groups remains nearly unchanged, and decreases only slightly with the increase in hydrogen peroxide concentration, in good agreement with the results observed in Figs. 11 and 12. Further analysis of the number of hydrogen peroxide molecules in the first hydration shell of surface silanol groups can find that the coordination number increases monotonously as the hydrogen peroxide concentration increases. It is known that hydrogen peroxide molecule, donating its hydrogen atoms to the in-plane silanols, contributes to the intensification of the 3400 cm1 peak in the SFG spectra. The correlation between the coordination number of hydrogen peroxide in the first hydration shell of silanol groups with relative peak intensity (I2/I1) from SFG experiments was analyzed and the results are presented in Fig. 14. It is obvious I2/I1 increases monotonously with the molecular numbers of hydrogen peroxide in the first hydration shell of silanol groups, which donates hydrogen atoms to in-plane silanols and contributes to the 3400 cm1 intensity in the SFG spectra.
Y.-q. Lv et al. / Journal of Molecular Structure 1113 (2016) 70e78
77
Table 3 The coordination number of different atoms around silanol groups in 0.5 M, 1.0 M, 1.5 M, and 2.0 M hydrogen peroxide solutions. C (mol/L)
n (the number of molecules in the first hydration shell) OseHHP
0.5 1.0 1.5 2.0
Fig. 12. Water dipole moment (angle a) relative number density distribution along the surface normal in the first hydration shell of quartz surface under four different hydrogen peroxide concentrations, ranging from 0.5 to 2.0 M. The solid lines represent the best fits.
4. Conclusion In this study, the molecular structure of hydrogen peroxide molecules at the silica surface has been examined using SFG
4.35 5.46 9.88 1.09
OseHw 3
10 103 103 102
2.60 2.60 2.61 2.55
HseOHP 1
10 101 101 101
1.27 1.98 2.60 2.88
HseOw 3
10 103 103 103
2.55 2.55 2.54 2.54
101 101 101 101
Fig. 14. The change of relative intensity (I2/I1) with the coordination number of hydrogen peroxide in the first hydration shell of silanol groups.
spectroscopy, and it is observed that as the concentration of hydrogen peroxide solution increases, there is an increase in the intensity of 3400 cm1 peak; however, the intensity of 3200 cm1 peak remains nearly unchanged. Molecular dynamics simulation study further reveals that the surface silanol groups can be separated into two families, out-ofplane and in-plane. The out-of-plane silanols donate a strongly
Fig. 13. Schematic illustration of the interactions between hydrogen peroxide and water molecules and surface silanols. The color representation is as follows: Si, yellow; O, red; and H, white. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article).
78
Y.-q. Lv et al. / Journal of Molecular Structure 1113 (2016) 70e78
hydrogen-bond to the nearby water molecule with the dipole moment orientation of 53 with respect to the surface normal, attributing to the 3200 cm1 peak intensity in the SFG spectra. The in-plane silanols, in contrast, accept a weakly hydrogen-bond from the nearby water molecule and the dipole moment orientation of water molecule centered at 130 , attributing to the 3400 cm1 peak intensity in the SFG spectra. Because H2O2 is a better proton donor than H2O but a weaker proton acceptor, and consequently, the H-bonds between the hydrogen peroxide molecules and silanol groups is longer than that of the water/silanol H-bonds. As the hydrogen peroxide is added to the water/quartz system, the hydrogen peroxide molecules prefer to parallel to the surface and donate H-bonds to the in-plane silanol groups rather than accept H-bonds from out-of-plane silanol groups, and thus, contribute to the water vibrational peak located at 3400 cm1. Acknowledgments We gratefully acknowledge the financial support from the Major State Basic Research Development Program of China (973 program) under grant No. 2013CB632601 and 2013CB632605, National Natural Science Foundation of China under grant No. 51274178 and 51404227. References [1] R.A. Patil, S.B. Kausley, P.L. Balkunde, et al., J. Water Health 11 (2013) 443e456. [2] Am Water Works Res F, Langlais B, Reckhow D A, et al., CRC Press. 1991. € m, J. Colloid Interface Sci. 130 (1989) [3] P. Tengvall, H. Elwing, I. Lundstro 405e413. [4] I.I. Salame, T.J. Bandosz, J. Colloid Interface Sci. 210 (1999) 367e374. [5] J.C. Kim, K.U. Lee, W.C. Shin, et al., Colloids Surfaces B Biointerfaces 36 (2004) 161e166. [6] A. McKillop, W.R. Sanderson, Tetrahedron 51 (1995) 6145e6166. [7] H.E. Strassler, W. Scherer, J.R. Calamia, N. Y. State Dent. J. 58 (1992) 30e35. [8] R.S. Varma, K.P. Naicker, Org. Lett. 1 (1999) 189e192. _ [9] J. Zeglinski, G.P. Piotrowski, R. Pie˛ kos, J. Mol. Struct. 794 (2006) 83e91. _ ski, A. Cabaj, M. Strankowski, et al., Colloids Surfaces B Biointerfaces [10] J. Zegli n 54 (2007) 165e172. [11] S. Bednarz, B. Rys, D. Bogdał, Molecules 17 (2012) 8068e8078. [12] F. Sudur, B. Pleskowicz, N. Orbey, Ind. Eng. Chem. Res. 54 (2015) 1930e1940.
[13] G.M. Rignanese, J.C. Charlier, X. Gonze, Phys. Chem. Chem. Phys. 6 (2004) 1920e1925. [14] A.V. Bandura, J.D. Kubicki, J.O. Sofo, J. Phys. Chem. C 115 (2011) 5756e5766. [15] J. Rubio, J.A. Kitchener, J. Colloid Interface Sci. 57 (1976) 132e142. [16] Y.R. Shen, Nature 337 (1989) 519e525. [17] Q. Du, R. Superfine, E. Freysz, et al., Phys. Rev. Lett. 70 (1993) 2313. [18] V. Ostroverkhov, G.A. Waychunas, Y.R. Shen, Chem. Phys. Lett. 386 (2004) 144e148. [19] M. Sulpizi, M.P. Gaigeot, M. Sprik, J. Chem. Theory Comput. 8 (2012) 1037e1047. [20] A.D. Kulkarni, R.K. Pathak, L.J. Bartolotti, J. Chem. Phys. 124 (2006), 1e214309. [21] A.D. Kulkarni, R.K. Pathak, L.J. Bartolotti, J. Phys. Chem. A 109 (2005) 4583e4590. [22] S. Ye, S. Nihonyanagi, K. Uosaki, Phys. Chem. Chem. Phys. 3 (2001) 3463e3469. [23] F. Vidal, A. Tadjeddine, Rep. Prog. Phys. 68 (2005) 1095. [24] G.L. Richmond, Chem. Rev. 102 (2002) 2693e2724. [25] S. Gopalakrishnan, D. Liu, H.C. Allen, et al., Chem. Rev. 106 (2006) 1155e1175. [26] W. Smith, T.R. Forester, J. Mol. Graph. 14 (1996) 136e141. [27] R.J. Kirkpatrick, A.G. Kalinichev, X. Hou, et al., Mater. Struct. 38 (2005) 449e458. [28] R.T. Cygan, J.-J. Liang, A.G. Kalinichev, J. Phys. Chem. B 108 (2004) 1255. [29] H. J. C. Berendsen, J. P. M. P., W. F. Gunsteren, J. Hermans, Reidel: Dordrecht. (1981) 331. [30] E. Lindahl, B. Hess, D. van der Spoel, J. Mol. Model 7 (2001) 306. [31] D.M. York, T.A. Darden, L.G. Pedersen, J. Chem. Phys. 99 (1993) 8345e8348. [32] X. Wang, J. Liu, H. Du, et al., Langmuir 26 (2009) 3407e3414. [33] Q. Du, E. Freysz, Y.R. Shen, Phys. Rev. Lett. 72 (1994) 238. [34] V. Ostroverkhov, G.A. Waychunas, Y.R. Shen, Phys. Rev. Lett. 94 (2005) 046102. [35] V. Ostroverkhov, G.A. Waychunas, Y.R. Shen, Chem. Phys. Lett. 386 (2004) 144. [36] M.J. Wojcik, V. Buch, J.P. Devlin, J. Chem. Phys. 99 (1993) 2332e2344. [37] Segelstein D J, University of Missouri-Kansas City. 2011. [38] X. Wei, P.B. Miranda, C. Zhang, Y.R. Shen, Phys. Rev. B 66 (2002) 085401. [39] J. Bjornstrom, A. Martinelli, J.R.T. Johnson, et al., Chem. Phys. Lett. 380 (2003) 165e172. re, J. Chem. Phys. 18 (1950) 88e92. [40] P.A. Gigue [41] M.J. Loeffler, B.D. Teolis, R.A. Baragiola, J. Phys. Chem. B 110 (2006) 6911e6915. [42] M.J. Loeffler, R.A. Baragiola, J. Phys. Chem. A 115 (2011) 5324e5328. [43] T.P.M. Goumans, A. Wander, W.A. Brown, et al., Phys. Chem. Chem. Phys. 9 (2007) 2146e2152. [44] J. Yang, E.G. Wang, Phys. Rev. B 73 (2006) 035406. [45] Y.W. Chen, I.H. Chu, Y. Wang, et al., Phys. Rev. B 84 (2011) 155444. [46] S.H. Lee, P.J. Rossky, J. Chem. Phys. 100 (1994) 3334e3345. [47] G. Sutmann, R. Vallauri, J. Mol. Liq. 98 (2002) 215e226. [48] Novak A, Springer Berlin Heidelberg. (1974) 177e216. [49] J.D. Kubicki, D. Sykes, G.R. Rossman, Phys. Chem. Miner. 20 (1993) 425e432. [50] S.W. Peterson, H.A. Levy, Acta Crystallogr. 10 (1957) 70e76. pez, Chem. Phys. 332 (2007) 341e347. [51] M.T.C. Martins-Costa, M.F. Ruiz-Lo