Hydration of ammonia, methylamine, and methanol in amorphous solid water

Hydration of ammonia, methylamine, and methanol in amorphous solid water

Chemical Physics Letters 645 (2016) 27–31 Contents lists available at ScienceDirect Chemical Physics Letters journal homepage: www.elsevier.com/loca...

673KB Sizes 1 Downloads 97 Views

Chemical Physics Letters 645 (2016) 27–31

Contents lists available at ScienceDirect

Chemical Physics Letters journal homepage: www.elsevier.com/locate/cplett

Hydration of ammonia, methylamine, and methanol in amorphous solid water Ryutaro Souda International Center for Materials Nanoarchitectonics, National Institute for Materials Science, 1-1 Namiki, Tsukuba, Ibaraki 305-0044, Japan

a r t i c l e

i n f o

Article history: Received 16 October 2015 In final form 11 December 2015 Available online 18 December 2015

a b s t r a c t Interactions of polar protic molecules with amorphous solid water (ASW) have been investigated using temperature-programmed desorption and time-of-flight secondary ion mass spectrometry. The ammonia and methylamine are incorporated into the interior of porous ASW films. They are caged by water molecules and are released during water crystallization. In contrast, the methanol–water interaction is not influenced by pores of ASW. The methanol additives tend to survive water crystallization and are released during ASW film evaporation. The hydration of n-hexane in ASW is influenced significantly by methanol additives because n-hexane is accommodated in a methanol-induced hydration shell. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Interactions of small functionalized organic molecules with water are useful to elucidate biomolecular systems in their aqueous environments. Two important molecules in this respect are alcohols and amines. Because they have hydrophilic and hydrophobic moieties, their interaction with water is complicated. The increase in the entropy of water–alcohol mixture is much less than that expected from the ideal solution. This phenomenon has been explained by the formation of highly ordered clathrate-like water structures around the aliphatic group (the so-called ‘iceberg’ model) [1]. In fact, the neutron diffraction study suggested that a distorted cage is formed around the methanol molecule for a dilute methanol solution in water [2]. Furthermore, more recent neutron diffraction [3] and X-ray emission [4] data suggest that mixing of a concentrated methanol–water solution is incomplete at the molecular level. The hydration and protonation behaviors of methylamine in water have been discussed mainly based on Monte Carlo and molecular dynamics simulations [5–8]. The first hydration shell of an amino group contains three or fewer water molecules participating in hydrogen bonding. A strong hydrophobic association of methyl groups is suggested, but the hydration structure in aqueous methylamine solution can differ from that in the methanol–water mixture. To date, intermolecular interactions or hydration of adspecies have been investigated extensively using thin films of amorphous solid water (ASW) at cryogenic temperatures [9–24]. The water

E-mail address: [email protected] http://dx.doi.org/10.1016/j.cplett.2015.12.028 0009-2614/© 2015 Elsevier B.V. All rights reserved.

molecules become mobile at the glass-transition temperature (Tg = 136 K) [25], as demonstrated by experiments of time-offlight secondary ion mass spectrometry (TOF-SIMS) [26]. The crystallization kinetics of water is discussed based on temperatureprogrammed desorption (TPD) of simple molecular additives in ASW [27–31]. Nonpolar molecules such as CCl4 embedded underneath thin ASW films are released explosively at the crystallization temperature of water via the ‘molecular volcano’ mechanism [28]. The volcano peak is also observed when N2 , O2 , CO, H2 S, OCS, CO2 , C2 H2 , SO2 , CS2 , and CH3 CN molecules are adsorbed onto porous ASW films, although the peak is absent for HCOOH, CH3 OH, and NH3 adspecies [31]. The result of NH3 is obscure because the TPD spectrum of ammonia is not separable from that of water, but it is categorized intuitively as a group of polar protic molecules. The specificity of methanol might be ascribed to formation of a type II clathrate hydrate [32]. Small amounts of methanol additives can modify the properties of ASW [21,26]. However, no report in the literature describes a systematic study that has elucidated the interactions of polar protic molecules with ASW. The roles of hydrogen bond formation at the polar group and water cage formation around the hydrophobic moiety are the most intriguing questions related to hydration of amphiphiles. As described herein, the ammonia and methylamine additives interacting with porous and nonporous ASW films are studied in comparison to water–methanol interactions using TOF-SIMS and TPD. The uptake and diffusion of additives in the ASW film are investigated using TOF-SIMS. In addition, the hydration–dehydration processes of additives in the thin film interior are discussed based on TPD spectra. The roles not only of the hydrogen bond formation with water molecules but also of pores of ASW films for hydrophobic caging are specifically examined. The H/D exchange

28

R. Souda / Chemical Physics Letters 645 (2016) 27–31

rate between adspecies and water is examined based on TPD as a measure of the hydrogen bond formation in the hydration state. The nonpolar n-hexane additive is coadsorbed with methanol and methylamine additives. Their TPD spectra are measured to shed light on hydrophobic hydration and the association of these additives in the ASW film interior. 2. Experiment

3. Results and discussion The coverage of ASW was determined based on evolution curves of secondary ion intensities as a function of exposure [20–22]. It requires ca. 2.5 L (Langmuir: 1 L = 1 × 10−6 Torr s) to form a monolayer (ML) of water on Ni(1 1 1) at temperatures below 100 K. All additives studied here (NH3 , CH3 NH2 , CH3 OH, and C6 H14 ) were deposited at 20 K on or underneath the ASW film at exposure of 1 L. Figure 1 shows temperature-programmed TOF-SIMS intensities from an ammonia-adsorbed ASW film. The H2 18 O molecule (20 L or 8 ML) was deposited onto Ni(1 1 1) at 20 K, forming a porous ASW film. Then the NH3 molecule was adsorbed onto the film surface. The NH3 adspecies disappears from the surface at T > 130–150 K, as revealed from steep decay of the NH4 + intensity. The increase of the Ni+ ion at 160 K indicates that the ASW film dewets the Ni(1 1 1)

Figure 1. TOF-SIMS intensities of the ammonia adsorbed porous ASW film as a function of the substrate temperature. The H2 18 O molecule (20 L) was deposited on Ni(1 1 1) at 20 K. Then the NH3 molecule (1 L) was adsorbed onto it at 20 K. The temperature ramp rate was 5 K min−1 .

14000 NH3 (16 amu) on Ni(111) ASW (20 L @120K) ASW (20 L @20K)

12000

Intensity (arb. units)

Experiments were performed in an ultrahigh vacuum (UHV) chamber with base pressure of <1 × 10−10 Torr. Actually, TOF-SIMS measurements were made using a primary beam of 2-keV He+ ions generated in an electron-impact-type ion gun (IQE 12/38; Specs GmbH). The ion beam was incident to the sample surface at an angle of 70◦ after chopping into pulses using electrostatic deflectors. To extract low-energy secondary ions efficiently, a bias voltage (±500 V) was applied to the sample and a grounded mesh was placed approximately 4 mm in front of the surface. Secondary ions ejected perpendicularly to the surface were detected using a microchannel plate after passage through a field-free TOF tube. The fluence of He+ in TOF-SIMS measurements was restricted below 1 × 1012 ions cm−2 to minimize thin film decomposition. The TPD spectra were recorded using a quadrupole mass spectrometer (QMS; IDP 300S; Hiden Analytical Ltd.) placed in a differentially pumped housing. A retractable orifice was placed approximately 3 mm distant from the sample to detect molecules desorbed from the surface. A Ni(1 1 1) surface was used as a substrate. It was heated several times in UHV to approx. 1300 K by electron bombardment from behind. The substrate was mounted on a Cu cold finger extended from a closed-cycle helium refrigerator. Then it was cooled to 20 K. The cold finger temperature was monitored close to the sample position using Au(Fe)-chromel thermocouples. It was controlled using a digital temperature controller and a cartridge heater attached to the finger. The temperature was ramped at a rate of 5 K min−1 for both TPD and TOF-SIMS measurements. Thin films were deposited onto the clean Ni(1 1 1) surface by backfilling the UHV chamber with gaseous samples admitted through high precision leak valves. Liquid samples of water (H2 16 O, H2 18 O, and D2 O), methanol, and n-hexane were degassed using several freezepump-thaw cycles. Gaseous ammonia was admitted from a glass bottle without further purification. Methylamine vapor was created from its aqueous solution (40%) after several pumping repetitions. The vapor includes ca. 20% of H2 O, as estimated from comparison of the water TPD peak area of a multilayer film formed by the deposition of methylamine with that of pure water at the same exposure. Surface cleanliness of Ni(1 1 1), as well as purity of other films, was confirmed in situ by the absence of impurity peaks in TOF-SIMS spectra.

10000

18

H2 O (20 amu x1/30)

8000 6000 4000 2000 0 80

100

120

140

160

180

200

T (K) Figure 2. TPD spectra of NH3 (1 L) and H2 18 O (20 L) molecules from the porous ASW film deposited at 20 K. Also shown are TPD spectra of NH3 (1 L) from the nonporous ASW film (prepared by deposition of 20 L H2 18 O at 120 K) and Ni(1 1 1) substrate. The temperature ramp rate was 5 K min−1 .

substrate at this temperature. Dewetting of a pure ASW film is associated with water crystallization because nucleation occurs in a liquid-like phase [26]. Consequently, the NH3 adspecies have no effects on the crystallization kinetics of water. Figure 2 displays TPD spectra of the H2 18 O and NH3 molecules desorbed from the porous ASW film prepared in the same manner that shown in Figure 1. For comparison, TPD spectra of 1 L NH3 adsorbed onto Ni(1 1 1) and a nonporous ASW film (formed by deposition of 20 L H2 18 O at 120 K) are also shown. The NH3 molecule desorbed in the gas phase is monitored using a 16 amu signal to reduce the contribution from the background H2 O molecule. The NH3 molecule desorbs from the porous ASW film over a wide temperature range with a characteristic peak at ca. 160 K. At this temperature, the H2 18 O TPD spectrum exhibits a shoulder as a result of water crystallization. Based on the TPD result, the decay of the TOF-SIMS NH4 + intensity at T > 130 K (Figure 1) is not ascribable to desorption of NH3 from the ASW film surface. The NH3 molecules forming a peak at 160 K come from the ASW film interior during water crystallization. Desorption occurs after the

R. Souda / Chemical Physics Letters 645 (2016) 27–31

6000

(a)

(a)

4

H3O

Intensity (arb. units)

10

3

+

+

2

CH3

10

5

10

4

10

3

+

(b)

2000

0 1500

(b) 1000

H2O (18 amu)

500

CH3NH2 (31 amu)

10

CH3NH2 (31 amu) CH3NHD (32 amu)

H

Ni

10

4000

+

H (CH3NH2)

+

Intensity (arb. units)

10

29

2

0 60

80

100

120

140

160

180

200

220

T (K) Figure 3. Temperature-programmed TOF-SIMS intensities (a) and TPD spectra (b) for methylamine (1 L) embedded underneath the H2 O film (20 L deposited at 20 K).

NH3 adspecies disappears from the film surface completely. The onset of NH3 desorption from the nonporous ASW film deposited at 120 K agrees with that from the Ni(1 1 1) substrate, but the TPD spectrum is broadened considerably in comparison with that from Ni(1 1 1). Probably, this is true because ammonia forms hydrogen bonds with water molecules. The desorption onset of NH3 shifts to higher temperature from the porous ASW film, suggesting that more numerous water molecules take part in the hydrogen bonding with the ammonia adspecies on the surface of porous ASW. The adspecies might be incorporated in pores via surface diffusion. This is the case for nonpolar adspecies for which mobility is independent of hydrogen bonds [22]. However, uptake of the ammonia adspecies occurs after the pore collapse at 120 K, as presented in Figure 1. Probably, this is explainable as that the hydrogen-bonded ammonia tends to be incorporated in the film interior after mobility occurs for water at T > Tg . The ammonia adspecies is released at 160 K without reappearance at the film surface. Such species are thought to be hydrated in the film interior, whereas the unhydrated species remaining near the surface region forms a broad peak (110–155 K) in TPD. In fact, the sharp peak at 160 K becomes dominant relative to the broad peak when the ammonia adlayer is capped with the ASW film (not shown). Figure 3 displays the experimentally obtained results of (a) TOFSIMS and (b) TPD obtained for the methylamine (1 L) adsorbed Ni(1 1 1) substrate, which is capped with the porous ASW film (20 L H2 18 O). The CH3 + and H+ (CH3 NH2 ) ions from methylamine tend to increase in intensity at T > 130 K because of diffusion of methylamine molecules to the surface. This behavior is probably induced by mobile water molecules because methylamine forms hydrogen bonds with water. The Ni+ ion increases at 155–160 K as a result of the film dewetting during water crystallization. In contrast to the CH3 + ion, the H+ (CH3 NH2 ) ion does not decay after water multilayer film evaporation. Probably, the protonation of methylamine occurs more efficiently for chemisorbed species on the Ni(1 1 1) substrate than the physisorbed species bound to the water molecule [23]. The segregated methylamine tends to remain near the surface region

80

100

120

140

160

180

200

T (K) Figure 4. TPD spectra for methylamine (1 L at 20 K) deposited onto (a) porous and (b) nonporous ASW films, respectively, prepared by deposition of D2 O molecules (20 L) at 20 K and 120 K onto Ni(1 1 1).

until water crystallization occurs, as revealed from the CH3 + intensity. The result contrasts starkly to that of ammonia (Figure 1). The amphiphilicity of methylamine is probably responsible for this behavior: The amino group forms hydrogen bonds with water, whereas the methyl moiety can be excluded from the hydrogenbond network of water on the surface. The TPD spectrum reveals that thermal desorption of methylamine commences at 140 K, but most molecules are released during film dewetting (i.e., water crystallization) at ca. 160 K. To gain more insights into hydrogen bonding between the adspecies and water, isotope exchange measurements were made. Figure 4 displays TPD spectra for the CH3 NH2 adspecies (1 L) on the (a) porous and (b) nonporous ASW films prepared respectively by deposition of 20 L D2 O molecules at 20 and 120 K. The shapes of TPD methylamine spectra from porous and nonporous ASW films differ considerably. A narrow TPD peak of methylamine from the porous ASW film at around 160 K resembles that depicted in Figure 3b for methylamine embedded underneath the ASW film, indicating that the molecules can be incorporated in the ASW film interior irrespective of the deposition order of the adspecies. Most of the incorporated methylamine molecules are released at ca. 160 K, although the tail to the high-temperature side might be readily apparent. In contrast, a broad spectrum of methylamine occurs without the peak at 160 K from the nonporous ASW film, suggesting that the methylamine is not incorporated into the film interior. The proton transfer reaction is induced by hydrogen bond formation between methylamine and water. The H/D exchange ratios, as estimated from the TPD peak area of CH3 NHD relative to CH3 NH2 , are 0.31 and 0.20, respectively, for porous and nonporous ASW films, suggesting that methylamine forms a larger number of hydrogen bonds with water on the porous ASW film. However, the H/D exchange ratio between NH3 and D2 O is not estimated precisely because the QMS signal of the H/D exchanged ammonia overlaps with that of heavy water. The intensity ratio of NH2 D to

30

R. Souda / Chemical Physics Letters 645 (2016) 27–31

5

4

10

5

10

H2O (18 amu) CD3OH (35 amu) CD3OD (36 amu)

4

10

3

2

10

1

10 100

(a)

H2O (18 amu) CH3OH (31 amu) C6H14 (57 amu)

10

3

10

120

140

160

180

200

T (K) Figure 5. TPD spectra of water and deuterated methanol for the CD3 OD (1 L) adsorbed porous H2 O film (20 L) prepared by deposition onto Ni(1 1 1) at 20 K.

Intensity (arb. units)

Intensity (arb. units)

10

2

10

1

10

5

10

4

10

(b)

H2O (18 amu) CH3NH2 (31 amu) C6H14 (57 amu)

3

NH3 species from NH3 (1 L) adsorbed onto the porous D2 O film (20 L at 20 K) is roughly estimated as about 0.8, for the TPD peak at 160 K. Earlier studies [23,24] based on TOF-SIMS have examined NH3 and CH3 NH2 adsorption on porous D2 O films. Results revealed that the amount of isotope exchanged species (e.g., NH2 D and CH3 NHD) increases concomitantly with increasing temperature. The completely H/D exchanged species such as ND3 and CH3 ND2 remain on the Ni(1 11 ) substrate after D2 O film evaporation. These results suggest that complete H/D exchange occurs on the porous D2 O film. In contrast, the TPD result shows that a considerably small amount of adspecies undergoes isotope exchange on porous and nonporous D2 O films because adspecies probed by TOF-SIMS are distinct from those probed by TPD. The complete H/D exchange occurs for chemisorbed residues on the Ni(1 1 1) substrate in TOF-SIMS measurements. However, most of the physisorbed species desorb from the ASW film without undergoing efficient H/D exchange. For comparison, the H/D exchange is also investigated for methanol additives interacting with ASW. Figure 5 displays TPD spectra for the CD3 OD (1 L) adsorbed porous H2 O film (20 L). In this case, a broad peak is observed in methanol TPD without a sharp peak during water crystallization at around 160 K. The occurrence of a shoulder in the H2 O TPD spectrum at around this temperature is indicative of water crystallization. Therefore, numerous methanol adspecies can survive the phase transition of water. They eventually desorb along with evaporation of water, as evidenced by the main TPD peak at 175 K. The yield of CD3 OH relative to CD3 OD is 4.9. Consequently, the H/D exchange occurs much more efficiently for methanol than for ammonia and methylamine. It is particularly interesting that TPD spectra of water and methanol obtained using the nonporous H2 O film (20 L deposited at 120 K) are almost identical to those presented in Figure 5 in terms not only of the spectral shapes but also of the CD3 OH to CD3 OD intensity ratio (not shown explicitly). These results indicate that pores of ASW have no effect on interactions between methanol and water. The methanol is likely to be hydrated such that the OH group enters the hydrogen bond network of water directly from the ASW film surface, thereby resulting in the high H/D exchange rate. In contrast, pores of ASW films are known to play a decisive role in hydrophobic hydration of nonpolar additives because they are caged by the water molecules in interstitial sites [22]. The caged species are released (dehydrated) during crystallization of water at 160 K, whereas methanol additives forming a hydrogen-bond network of water can survive water crystallization. Consequently, the hydration state of ammonia and methylamine differs from that of methanol and rather resembles hydrophobic hydration of nonpolar additives in terms of the effects of ASW pores and the occurrence of the sharp dehydration peak. Probably, the hydrogen bonding of the ammonia and methylamine

10

2

10

1

10 100

120

140

160

180

200

T (K) Figure 6. (a) TPD spectra of water, methanol, and n-hexane for methanol (1 L) and nhexane (1 L) additives embedded underneath the ASW film (20 L of H2 O at 20 K), and (b) those of water, methylamine, and n-hexane for methylamine (1 L) and n-hexane (1 L) additives embedded underneath the ASW film (20 L of H2 O at 20 K).

additives with water is weaker than that of methanol. The H/Dexchanged methylamine yields are low and comparable to those shown in Figure 4a and b because the circumstances are similar between entrapped molecules interacting with the cage wall surface and adsorbed ones interacting with the free surface of ASW. In the framework of polyamorphism [33], the local structure of ASW resembles that of crystalline water [34], so that hydrophobic hydration is promoted by the formation of ice-like cages. If methanol modifies the hydrogen bond network of water, then the hydration behaviors of additives can also be influenced by methanol. To assess this possibility, interactions of additives in ASW are examined based on TPD. Figure 6 presents a comparison of TPD spectra for (a) n-hexane (1 L) on methanol (1 L) and (b) n-hexane (1 L) on methylamine (1 L). They were adsorbed sequentially on the Ni(1 1 1) substrate, and 20 L of water molecules were deposited on them at 20 K. The n-hexane additives are dehydrated at ca. 160 K during water crystallization [22], but the peak position of n-hexane is shifted to a higher temperature by methanol additives, as depicted in Figure 6a. The peak at 160 K is almost absent. The main peak occurs at 175 K simultaneously with water and methanol TPD peaks. In contrast, the n-hexane peak appears at 160 K in Figure 6b, together with the main peak of methylamine, indicating that methylamine has no effects on hydration of the coadsorbed n-hexane molecule. These results are associated with different hydration structures of methanol and methylamine in ASW. The amount of methanol additives is small relative to the water matrix, but most of the n-hexane additives are influenced by methanol additives. Probably, dimerization occurs via hydrophobic association between methanol and n-hexane, so that the n-hexane additive tends to be accommodated in the hydration shell modified by methanol additives in ASW. The fact that the trapped n-hexane molecule can survive water crystallization and desorb together with water and methanol suggests that the hydration structure induced by methanol is retained across water crystallization.

R. Souda / Chemical Physics Letters 645 (2016) 27–31

Intensity (arb. units)

5

10

4

10

H2O (18 amu) CH3NH2 (30 amu) CH3OH (32 amu) C6H14 (57 amu)

3

10

2

10

1

10 100

120

140

160

180

200

T (K) Figure 7. TPD spectra of water, methylamine, methanol, and n-hexane for methanol (1 L), methylamine (1 L), and n-hexane (1 L) additives embedded underneath the ASW film (20 L of H2 O) deposited at 20 K.

In this respect, results of transmission electron microscopy and electron diffraction studies of H2 O–CH3 OH mixtures have suggested that a type II clathrate hydrate of methanol is formed after water crystallization [32]. The clathrate hydrate is expected to be formed for hydrophobic guests because the hydrogen-bonded water lattice should not be disrupted. However, this is not the case for methanol additives, as demonstrated in this study. Moreover, inelastic neutron scattering studies have demonstrated that the hydrogen-bond network of water is broken by methanol doping in ASW whereas the ice-like or clathrate-like structure is reinforced by doping of hydrophobes such as SF6 [35]. Based on these experimentally obtained facts, it is likely that a methanol-rich grain-boundary phase (probably amorphous) segregates from crystallites of pure water at 160 K. As a consequence, the n-hexane additives can be trapped by methanol-induced hydration cages irrespective of the water phase transition. In contrast to methanol, the hydration cage of water is fundamentally not influenced by methylamine additives, so that n-hexane is caged by water solely or jointly with methylamine. The broadness of the TPD spectrum of methylamine relative to that of n-hexane might be an indication of hydrogen bond formation with water. Interactions between methanol and methylamine additives in water are explored further using n-hexane as a probe. The TPD spectra for n-hexane (1 L), methanol (1 L), and methylamine (1 L) additives capped with the H2 O overlayer (20 L) are portrayed in Figure 7. In this case, the methylamine and methanol are monitored respectively by 30 and 32 amu signals because of overlapping of the main 31 amu signal. The peak at 160 K becomes evident in the methanol TPD spectrum relative to that in Figure 6a whereas the methylamine spectrum is fundamentally the same as that without methanol, as depicted in Figure 6b. The methanol spectrum shape becomes similar to that of methylamine. The n-hexane TPD peak appears at 160 K and resembles that in Figure 6b. The result suggests strongly that the methanol-induced hydration cage is quenched by association of the methanol and methylamine additives to form hydrogen bonds. As a consequence, all additives are accommodated fundamentally in the hydration shells of the pure water molecules, thereby forming main TPD peaks at 160 K. 4. Conclusion The hydration of ammonia, methylamine, methanol, and nhexane additives and their mutual interactions in thin ASW films were examined in this study. The roles of pores in hydration of additives were examined from comparison of the TPD spectra from porous and nonporous ASW films. The ammonia and methylamine adspecies are thought to be caged by hydrogen-bonded water molecules, as evidenced by the fact that pores play an important

31

role in their entrapment in ASW. The hydration cages of the water molecules collapse during water crystallization, thereby producing sharp TPD peaks of the additives at ca. 160 K. The results resemble hydrophobic hydration of nonpolar additives. In fact, a small H/D exchange rate of methylamine with heavy water (approximately 0.3) can be an indication of the ineffective hydrogen-bond formation as a result of hydrophobic hydration. Probably, the additives can form hydrogen bonds with free OH groups in a cage without modification of the hydrogen-bond network of the water molecules. In contrast, pores of the ASW films have no effects on the interaction with the methanol additives because methanol enters the hydrogen bond network of the water molecules directly. This behavior also manifests itself in the much higher H/D exchange rate of methanol (approx. 4.9) than of methylamine. Results show that the methanol additives tend to survive water crystallization at 160 K because mixing occurs with water via hydrogen bonding. Because of this behavior, the hydration properties of ASW are modified to a considerably degree by the methanol additives. Nonpolar n-hexane molecules can survive water crystallization if methanol additives coexist, suggesting that n-hexane can be incorporated in the hydration shell induced by methanol. No apparent effects have been identified on modification of n-hexane hydration by methylamine additives. However, methanol-induced hydration of n-hexane is inactivated if methylamine additives coexist, suggesting that the associative interaction between methylamine and methanol additives prevails over the water–methanol interaction in the ASW film interior. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35]

H.S. Frank, M.W. Evans, J. Chem. Phys. 135 (1945) 507. A.K. Soper, J.L. Finney, Phys. Rev. Lett. 71 (1993) 4346. S. Dixit, J. Crain, W.C.K. Poon, J.L. Finney, A.K. Soper, Nature 416 (2002) 829. J.H. Guo, Y. Luo, A. Augustsson, S. Kashtanov, J.E. Rubensson, D.K. Shuh, Phys. Rev. Lett. 91 (2003) 157401. W.J. Dunn, P.I. Nagy, J. Phys. Chem. 94 (1990) 2099. R.C. Rizzo, W.L. Jorgensen, J. Am. Chem. Soc. 121 (1999) 4827. P.G. Kusalik, D. Bergman, A. Laaksonen, J. Chem. Phys. 113 (2000) 8036. H. Hesske, K. Gloe, J. Phys. Chem. A 111 (2007) 9848. H. Ogasawara, N. Horimoto, M. Kawai, J. Chem. Phys. 112 (2000) 8229. S.C. Park, H. Kang, J. Phys. Chem. B 109 (2005) 5124. S.C. Park, K.W. Maeng, T. Pradeep, H. Kang, Angew. Chem. Int. Ed. 40 (2001) 1497. S.C. Park, T. Pradeep, H. Kang, J. Chem. Phys. 113 (2000) 9373. J. Cyriac, T. Pradeep, J. Phys. Chem. C 111 (2007) 8557. J. Günster, G. Liu, J. Stultz, D.W. Goodman, J. Chem. Phys. 110 (1999) 2558. S. Krischok, O. Höfft, J. Günster, J. Stultz, D.W. Goodman, V. Kempter, Surf. Sci. 495 (2001) 8. J. Günster, G. Liu, J. Stultz, S. Krischok, D.W. Goodman, J. Phys. Chem. B 104 (2000) 5738. J. Günster, S. Krischok, V. Kempert, J. Stults, D.W. Goodman, Surf. Rev. Lett. 9 (2002) 1511. J. Cyriac, T. Pradeep, H. Kang, R. Souda, R.G. Cooks, Chem. Rev. 112 (2012) 5356. R.G. Bhuin, R.R. Methikkalam, B. Vivaraman, T. Pradeep, J. Phys. Chem. C 119 (2015) 11524. R. Souda, J. Phys. Chem. C 117 (2013) 26969. R. Souda, J. Phys. Chem. B 110 (2006) 17524. R. Souda, J. Phys. Chem. B 109 (2005) 21879. R. Souda, Surf. Sci. 547 (2003) 403. R. Souda, J. Chem. Phys. 119 (2003) 2774. G.P. Johari, A. Hallbrucker, E. Mayer, Nature 330 (1987) 552. R. Souda, Phys. Rev. Lett. 93 (2004) 235502. R.S. Smith, B.D. Kay, Nature 398 (1999) 788. R.S. Smith, C. Huang, E.K.L. Wong, B.D. Kay, Phys. Rev. Lett. 79 (1997) 909. R.S. Smith, C. Huang, B.D. Kay, J. Phys. Chem. B 101 (1997) 6123. R.A. May, R.S. Smith, B.D. Kay, Phys. Chem. Chem. Phys. 13 (2011) 19848. M.P. Collings, M.A. Anderson, R. Chen, J.W. Dever, S. Viti, D.A. Williams, M.R.S. McCoustra, Mon. Not. R. Astron. Soc. 354 (2004) 1133. D. Blake, L. Allamandola, S. Sandford, D. Hudgins, F. Freund, Science 254 (1991) 548. O. Mishima, H.E. Stanley, Nature 396 (1998) 329. J.L. Finney, A. Hallbrucker, I. Kohl, A.K. Soper, D.T. Bowron, Phys. Rev. Lett. 88 (2002) 225503. O. Yamamuro, T. Matsuo, I. Tsukushi, O. Onoda-Yamamuro, Can. J. Phys. 81 (2003) 107.