Coexistence of Kosmotropic and Chaotropic Impacts of Urea on Water

Subscriber access provided by AUSTRALIAN NATIONAL UNIV

Article

Coexistence of Kosmotropic and Chaotropic Impacts of Urea on Water as Revealed by Terahertz Spectroscopy Keiichiro Shiraga, Yuichi Ogawa, Koichiro Tanaka, Takashi Arikawa, Naotaka Yoshikawa, Masahito Nakamura, Katsuhiro Ajito, and Takuro Tajima J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.7b11839 • Publication Date (Web): 30 Dec 2017 Downloaded from http://pubs.acs.org on December 31, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry B is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 45 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 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Coexistence of Kosmotropic and Chaotropic Impacts of Urea on Water as Revealed by Terahertz Spectroscopy Keiichiro Shiraga†,*

[email protected]

Yuichi Ogawa‡

[email protected]

Koichiro Tanaka||,#

[email protected]

Takashi Arikawa||

[email protected]

Naotaka Yoshikawa||

[email protected]

Masahito Nakamura§

[email protected]

Katsuhiro Ajito§

[email protected]

Takuro Tajima§

[email protected]

†

RIKEN Center for Integrative Medical Sciences (IMS), Suehiro-cho, Tsurumi-ku, Yokohama, Kanagawa 230-0045, Japan

‡

Graduate School of Agriculture, Kyoto University, Kitashirakawa-oiwakecho, Sakyo-ku, Kyoto 606-8502, Japan

||

Department of Physics, Kyoto University, Kitashirakawa-oiwakecho, Sakyo-ku, Kyoto 606-8502, Japan

#

Institute for Integrated Cell-Material Sciences (iCeMS), Kyoto University, Sakyo-ku, Kyoto 606-8501, Japan.

§

NTT Device Technology Labs, NTT Corporation, Morinosato Wakamiya, Atsugi, Kanagawa 243-1098, Japan

* Corresponding author: Fax +81 45 503 7014, [email protected] 1 ACS Paragon Plus Environment

The Journal of Physical Chemistry 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 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 45

Abstract Whether urea can serve as a kosmotrope or chaotrope has long been a topic of debate. In this study, broadband THz spectroscopy (from 0.2 to 12 THz) of aqueous solutions of urea was used to characterize the hydration state and the hydrogen bond structure of water around urea. Three low-frequency vibration modes of urea were found around 2 THz, 4 THz, and above 12 THz. After eliminating the contribution of these modes, the “urea-vibration-free” complex dielectric constant was decomposed into the relaxation modes of bulk water and the oscillation modes of water. When hydration water is defined to be reorientationally retarded relative to bulk, our analysis revealed that the hydration number is 1.9 independent of the urea concentrations up to 5 M and this number is close agreement with that of water constrained by the strong acceptor hydrogen bonds of urea oxygen. Regarding the hydrogen bond structure, it was found that the tetrahedral-like water structure is mostly preserved (though the hydrogen-bond lifetime is significantly shortened) but the population of non-hydrogen-bonded water molecules fragmented from the network is markedly increased, presumably due to the urea’s NH2 inversion. These experimental results point to coexistence of apparently two contradictory aspects of urea: dynamical retardation (the kosmotropic aspect) by the –CO group and slight structural disturbance (the chaotropic aspect) by the –NH2 group.

2 ACS Paragon Plus Environment

Page 3 of 45 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 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1.

Introduction The urea–water system has attracted attention because of its specific effect of a denaturant of

protein1 and an inhibitor of micellar aggregation2. On the basis of thermodynamic data, Frank and Franks3 attributed these peculiar characteristics to changes in the structure of water induced by urea. In particular, urea is hypothesized to increase the fraction of dense and distorted water species at the cost of the ice-like ordered one, which points to urea having a chaotropic (structure breaking) nature in the water structure.3 However, this description has been challenged by recent computational4-10 and experimental11-22 studies. Whereas some studies classify urea as a chaotrope (structure-breaker),4,11-16 others provide conflicting evidence supporting a komotropic (structure making) nature5,17 or little influence on the water structure6-10,18-22. A primary reason for these conflicting results is the complex dynamical structure of liquid water mediated by intermolecular hydrogen bonds (HBs): since the continuous lifetime of water–water HBs is typically of sub-picosecond order,23 the tetrahedral-like water structure continuously fluctuates on a sub-picosecond or picosecond timescale. In this circumstance, the observed water structure is dependent on the experimental time window employed24, and observing the behavior of water on these timescales leads to the most direct characterization of water HB dynamics. Polarization-resolved pump-probe spectroscopy19 and optical Kerr effect (OKE) spectroscopy14,25,26 are experimental approaches that meet above requirements. Nonetheless, a pump-probe 3 ACS Paragon Plus Environment


Page 4 of 45

experiment has difficulty in rigorously distinguishing anisotropy decay of the water OH stretching component from that of urea NH, which results in an ambiguous observation of water dynamics. Furthermore, weak OKE signals originating from polarizability changes make it difficult to recognize tiny changes in the dynamics and structure induced by the solute because of the relatively small polarizability of water. Terahertz (THz) spectroscopy is an alternative approach to characterizing the dynamical structure of water around urea because it directly observes sub-picosecond and picosecond dipole fluctuations. In particular, the complex dielectric constant, ̃, of liquid water in the THz region consists of Debye relaxation and damped harmonic oscillation (DHO) modes reflecting the reorientational dynamics of bulk water27 and the structural properties of the HB network28. Therefore, the hydration state and HB structure can be characterized on the basis of the solute-induced changes in the Debye and DHO modes29-35. However, since these water modes extend over a wide frequency range, broadband THz spectroscopy covering a range from sub-THz to above 10 THz is necessary for accurate extraction of the Debye and DHO modes. Despite the potential of revealing the dynamical structure of water around urea, only a few THz spectroscopic studies have been performed on urea–water interactions15,21. In these earlier studies, urea in water solvent was found to have inherent low-frequency vibration modes around 2~3 THz (67~100 cm-1) and 4~5 THz (133~167 cm-1) to a non-negligible extent14,21,25,26, unlike most macromolecules, which exhibit little dielectric response in the THz range compared with water29-35. 4 ACS Paragon Plus Environment

Page 5 of 45 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 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Idrissi and co-workers performed a molecular dynamics (MD) simulation and found that the 2~3 THz and 4~5 THz modes could be ascribed to translational and angular motions of solvated urea, respectively26. On the other hand, Funkner et al. tentatively assigned these modes to rattling motions of urea in the surrounding water cage.21 Although these findings imply that the low-frequency urea modes are associated with the motion of urea itself, firm evidence for this has not been provided yet. Since the low-frequency urea modes hinder selective observation of water relaxations and DHOs in the THz region, quantitative identification of the urea modes is essential to characterization of water around urea. A possible way to quantitatively identify the low-frequency urea modes is replacing water with other solvents such as dimethyl sulfoxide (DMSO), which has high solubility for urea but a relatively small absorption in the THz region35. Aiming to examine the impact of urea on the dynamical structure of water, therefore, we performed broadband THz spectroscopy in the range from 0.2 to 12 THz. In addition to those of urea aqueous solutions, the complex dielectric constants of urea DMSO solutions were determined in order to identify the low-frequency urea vibration modes (urea ). The hydration state and the HB structure around urea were analyzed by decomposing the urea-vibration-free complex dielectric constants (̃ − urea ) into the constituent Debye and DHO modes of water.

2.

Experimental Section



Page 6 of 45

2.1. Sample preparation. Urea aqueous and DMSO solutions were prepared by dissolving urea powder (purity ≥99.0 %, Sigma-Aldrich) into demineralized water and DMSO (Wako Pure Chemical Industries, Ltd.) at final urea concentrations of =1, 3 and 5 M. Then, density measurements were performed with a DMA 58 density meter from Anton Paar to derive the stoichiometric molar concentration of water water and DMSO DMSO in the urea solutions. 2.2. Determination of complex dielectric constants.

The complex dielectric constants of pure

water, DMSO, and urea solutions at 300 K were determined over the frequency range from 0.2 to 12 THz by using three different spectroscopic systems: terahertz continuous-wave attenuated total reflection spectroscopy (THz CW-ATR: 0.2−0.9 THz)36, terahertz time-domain attenuated total reflection spectroscopy (THz TD-ATR: 0.9−4.0 THz)37 and far-infrared Fourier-transform attenuated total reflection spectroscopy (FIR FT-ATR: 4−12 THz). The THz CW-ATR measurement employs a homodyne detection system to determine the real and imaginary parts of the complex dielectric constant between 0.2 and 0.9 THz to a frequency resolution of 10 GHz. In this system, two 1.55-µm-band lasers (a wavelength-fixed distribution feedback laser at 193.13 THz and a tunable laser at 193.33−194.03 THz) were amplified by an erbium doped optical fiber amplifier, and the combined laser beam was separated into two optical paths: one to a uni-traveling carrier photodiode (UTC-PD) emitter, the other to an InGaAs-based bowtie photoconductive antenna (PCA) detector. The emitted polarized high-power THz continuous waves gen6 ACS Paragon Plus Environment

Page 7 of 45 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 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


erated from the biased UTC-PD based on photomixing were led to an ATR Dove prism (inner reflection angle 51.6°, made of silicon) maintained at 300±0.1 K. The homodyne detection scheme allowed us to simultaneously record power and phase spectra, by mixing the propagating THz waves and the combined beat signal of the two lasers whose phase was modulated by a 20-kHz sawtooth function36. Then, substitution of the measured power and phase spectra into Fresnel’s equation enabled the real and imaginary parts of the complex dielectric constant to be determined. From 0.9 to 4.0 THz, we used a THz TD-ATR spectroscopy system equipped with the same silicon Dove prism as in the THz CW-ATR measurement. A mode-locked Ti: sapphire laser with a pulse duration of approximately 10 fs at a repetition rate of 80 MHz, a center wavelength of 800 nm, and an average power of 350 mW (Femtolasers, Ltd., femtosource twin synergy) was used for generation and detection of the THz pulses. Ultrashort THz pulses were emitted by optical rectification in a [110]-oriented GaP with a thickness of 300 µm38 and detected by a low-temperature-grown GaAs-based PCA. The power and phase shift spectra were calculated by a Fourier transform of the temporal THz pulses; then, the complex dielectric constant of the sample was derived37. The FIR FT-ATR measurements were performed with a FARIS-1s (Jasco, Ltd.) for the higher frequencies up to 12 THz. A ceramic heater and deuterated triglycine sulfate were used as a light source and detector, respectively, and the silicon ATR prism at 300±0.2 K was put at the focal position of the FIR light at an incident angle of 45°. The measured polarized reflectance spectrum was subjected to a Kra7 ACS Paragon Plus Environment


Page 8 of 45

mers−Kronig transform39 to calculate the phase shift spectra, and the complex dielectric constant between 4 and 12 THz was determined. The details of the Kramers−Kronig transformation are explained in the Supporting Information (S1).

3.

Results & Discussion 3.1. Identification of low-frequency urea modes.

Figures 1a,b show the complex dielec-

tric constant of pure water and urea aqueous solutions at 300 K. In good conformity to previous FIR spectroscopy21, we recognized three isosbestic points in the imaginary part around 1.5, 3.2 and 8.5 THz (see Figure S2 in the Supporting Information for clarity). The presence of these isosbestic points implies that the total dielectric responses of urea aqueous solutions can be described as a concentration-weighted superposition of water and urea modes. However, quantitative observation of the low-frequency urea vibration modes was hampered by the large background absorption of water. In order to settle this troubling issue, the complex dielectric constant of urea in DMSO was also determined, as shown in Figures 1c,d. It was reported that the complex dielectric constant of pure DMSO is constituted of two Debye relaxations below the sub-THz region40, a libration around 2 THz41, and intramolecular modes (C-S-C and S=O bending) in the 9−12 THz region42, but the imaginary part between 3 and 7 THz gradually developed with increasing urea concentration, giving rise to an isosbestic point around 3 THz. This observation confirms that the low-frequency urea vibration


Page 9 of 45 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 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


mode is present in DMSO as well as in water. To gain further insight into these urea vibration modes, we determined the difference molar extinction coefficient ∆⁄ (see the Supporting Information S2 for the derivation process). As can be seen in Figure 2a, the difference molar extinction coefficient of the urea aqueous solution ∆aq. ⁄ shows an asymmetric and broad mode around 4 THz and another intense mode above 12 THz. As deduced from the concentration independence of ∆aq. ⁄ up to 5 M, these urea modes are not associated with urea dimers (or other oligomers) because dimer formation would lead to a quadratic increase of these modes with urea concentration21. Additionally, aiming to understand temperature dependence of these urea modes, pure water and urea aqueous solution 5 M were also measured at 320 K (see the Supporting Information S3), and ∆aq. ⁄ was derived in the same manner. The little temperature dependence of ∆aq.⁄ , as can be seen in the inset of Figure 2a, is suggestive that these urea modes are not significantly affected by the water HB network, which is very sensitive to temperature. All these observations are consistent with a previous FIR spectroscopy by Funkner et al.21, who tentatively assigned the ~4 THz mode to a rattling motion of urea in the surrounding water cage. Within the scope of their tentative assignment, this ~4 THz urea mode is expected to disappear if the water solvent is replaced with DMSO, since DMSO is a rather unstructured liquid43. However, as shown in Figure 2b, the difference molar extinction coefficient of urea DMSO solution ∆DMSO ⁄ was found to have a concentration- and temperature-independent 9 ACS Paragon Plus Environment


Page 10 of 45

peak around 4 THz, which strongly casts doubt on their assignment of the 4 THz mode. On the other hand, a classical MD simulation study by Idrissi et al. claimed that the angular velocity autocorrelation function of urea gives rise to a peak around 3~4 THz and interpreted this band as librational motion of a single urea molecule.26 Given that the libration frequency of such a single molecule remains largely unaffected by the surrounding environment41,42, the temperature- and concentration-independent nature of the 4 THz mode observed in both aqueous and DMSO solutions is compatible with the scenario of libration dynamics of a single urea molecule. In addition to the 4 THz mode, ∆DMSO ⁄ implies the existence of another urea mode above 12 THz (note that the sharp dispersion at 10 THz comes from a blueshift of the asymmetric S=O bending of DMSO induced by urea and is not related to the urea mode itself). On the basis of the earlier MD simulation, this high-frequency mode can be assigned to an intramolecular vibration21 such as NH2 twisting44, NH2 inversion45, and/or CN bending44,45 of urea. Thus, the difference, ∆aq. − ∆DMSO , is expected to reveal the urea vibration mode that is inherent in water: as displayed in Figure 2c, the subtracted absorption spectrum shows a relatively weak and damped mode around 2 THz and an intense mode above 12 THz. While the latter is attributed to a redshift of the water libration mode46 (see the Supporting Information S5), the former 2 THz mode is supposed to be an inherent mode of urea dissolved in water, such as a water–urea cocluster vibration or urea rattling in the water cage. The water–urea cocluster vibration scenario seems to conflict with lack of temperature dependence of 10 ACS Paragon Plus Environment

Page 11 of 45 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 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


this 2 THz mode (see Figure 2c), because a temperature increase of 20 K significantly weakens water–urea HBs, which involves a peak shift and/or spectrum broadening. On the other hand, the rattling scenario is more acceptable for the following reason: though the 20 K temperature increase actually shortens the water–water HB lifetime, the restructuring time of water (i.e. > picoseconds) is so long for this damped 2 THz urea vibration that the water cage can be regarded as maintained during the oscillation. Therefore, the urea rattling scenario, rather than the water–urea cocluster vibration, seems to be a more conceivable assignment of the low-frequency urea vibration mode at 2 THz. For a quantitative understanding of these urea vibration modes, the complex susceptibilities of the 2THz 4THz urea rattling urea around 2 THz, the urea libration urea around 4 THz and the intramo-

12THz lecular vibration of urea urea above 12 THz were determined by fitting with DHO functions

(see the Supporting Information S6). Figure 3 shows the difference complex dielectric constant ∆̃aq. of urea aqueous solution at 5 M and its constituent complex susceptibilities of the urea vibration modes. From this figure it is evident that the total urea vibration mode, urea = 2THz 4THz 12THz urea + urea + urea , is not enough to reproduce ∆̃aq. . Accordingly, the shaded

area in Figure 3, ∆̃aq. − urea , is considered to reflect the modified Debye and DHO modes of water. Since such modified water modes are associated with non-bulk-like dynamics and structure of water perturbed by urea, the changes in the complex dielectric constant of the urea aqueous solution need to be further examined (in the next section). 11 ACS Paragon Plus Environment


3.2. Decomposition of the complex dielectric constant.

Page 12 of 45

With the urea vibration modes

revealed in the previous section, the “urea-vibration-free” complex dielectric constant, ̃ − urea , was derived (Figure 4a,b). It is broad and seemingly featureless, but several different sub-picosecond/picosecond dynamics of water are behind it27,28. For pure water, the complex dielectric constant ̃ can be better understood by decomposing it into the Debye and DHO modes as follows27,28:

̃ = w1 +w2 + S + L + " ̃ =

∆w1 ∆w2 ∆%S ∆%L + + & + & + " & 1 + #$w1 1 + #$w2 S − + #'S L − & + #'L

(1)

where, w1 is a slow relaxation of bulk water, w2 is a fast relaxation of bulk water, S is intermolecular stretching of water, L is libration of water, and " is the high-frequency limit of the real part. w1 and w2 are modeled by the Debye function with the relaxation strength ∆w1w2 and relaxation time $w1w2 , and S and L described by a DHO function consisting of the vibration strength ∆%SL , resonant frequency SL , and damping constant 'SL . The slow relaxation mode w1 located around 20 GHz has been intensively investigated with microwave dielectric spectroscopy47, and it is widely known to be a collective reorientation of hydrogen-bonded (HB) bulk water and to be sensitive to temperature and isotopic effects. In contrast, the fast relaxation w2 in the sub-THz region has not been sufficiently examined owing to experimental difficulty in this region, and thus, the assignment of this mode is still controversial48,49. 12 ACS Paragon Plus Environment

Page 13 of 45 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 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


However, based on the experimental evidence that the temperature- and isotopic-dependence of the w2 relaxation time $w2 nicely matches up with the theoretical collisional relaxation time of gaseous water molecules27,50, the origin of w2 is considered to be an individual relaxation of non-hydrogen-bonded (NHB) bulk water that transiently appears in the HB network27. Moreover, the intermolecular stretching S around 5 THz is a delocalized translation motion of hydrogen-bonded water51-53, and its dynamics are sensitive to fluctuations of the water HB network28. Finally, the libration mode (L : a hindered H...O rotational motion) has an asymmetric band shape54 caused by motional anisotropy of water 55. Two libration modes originating from different rotation axes are, in fact, distinguished (~12 THz and ~20 THz) in the FIR spectrum, but we have treated these two modes as one because only the low-frequency tail is observable in our measurement range (0.2– 12 THz). Previous dielectric spectroscopy studies of urea aqueous solutions20,56 revealed the existence of an additional Debye mode assigned to the urea–water cocluster reorientation in the lower GHz region. Since the contribution of this additional Debye mode is not negligible above 0.2 THz, the “urea-vibration-free” complex dielectric constant, ̃ − urea , should be fitted with a partially modified function from Eq. (1);

̃ − urea = )*+ + w1 +w2 + S + L + " ̃ − urea =

∆u-w ∆w1 ∆w2 + + 1 + #$u-w 1 + #$w1 1 + #$w2 13 ACS Paragon Plus Environment


̃ − urea = +

∆%S ∆%L + + " S & − & + #'S L & − & + #'L

Page 14 of 45

(2)

where, )*+ is the urea–water cocluster Debye mode with the relaxation strength ∆u-w and relaxation time $u-w . The Levenberg–Marquardt nonlinear least-squares method was used for fitting Eq. (1) and (2) under the constraint $w1 =7.93 ps

27,57

, assuming that bulk water molecules are re-

orientationally unperturbed by the solute in the urea aqueous solution. In addition, since the relaxation frequency of the )*+ mode (~7 GHz)20,56,58 is far outside our measurement range, its constituent parameters ∆u-w and $u-w were cited from Hayashi et al. at 298 K 20. Small temperature mismatch (300 K in this study vs. 298 K) is insignificant, as discussed in the Supporting Information S7. Figures 4c,d show the fitting results for ̃ − urea at 5 M and its decomposed Debye and DHO modes: the experimental results (empty circles) are well reproduced by the sum of the best-fitted complex susceptibilities (black dotted line) over the whole measured frequency range. The hydration state and HB structure around urea will thus be discussed in the following sections on the basis of the best-fitted parameters. 3.3. Hydration state.

Although the relaxation mode of hydration water is involved in the

)*+ process, the formation of urea–water coclusters makes it difficult to separate out the relaxation components of hydration water. On the other hand, if hydration water is defined to be orientationally retarded compared with bulk29,30, the amount of hydration water can be indirectly derived 14 ACS Paragon Plus Environment

Page 15 of 45 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 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


from the reduced bulk water relaxation strength (∆bulk = ∆w1 + ∆w2). Figures 5a,b show the urea concentration dependence of the w1 and w2 relaxation strengths in comparison with the analytical strength estimated from the stoichiometric water molar concentration water under the assumption that all the water molecules in the urea aqueous solution behave as bulk water (hereafter referred as “water dilution effect”). It can be seen that the w1 relaxation strength ∆w1 falls far below the virtual line, while the w2 relaxation strength ∆w2 gradually increases as the urea concentration becomes higher. Nevertheless, the increase in ∆w2 is so small compared with that of ∆w1 that the overall bulk water relaxation strength, ∆bulk = ∆w1 + ∆w2 , is smaller than the virtual values estimated from the water dilution effect. This suggests that a subset of bulk water in the urea aqueous solution is replaced with hydration water and no longer contributes to the bulk water processes. For a quantitative assessment, first of all, the molar concentration of bulk water bulk is derived as follows34,35:

bulk =

∆w1 ⁄0HB + ∆w2 ⁄0NHB 4water ∆w1 0⁄0HB + ∆w2 0⁄0NHB 5water

(3)

where, ∆w1w2 0 is the w1 (w2) relaxation strength of pure water, 4water is the density of water, and 5water is the molecular weight of water. 0HB =2.9 and 0NHB =1.0 are the Kirkwood correlation g factor of HB and NHB bulk water at 300 K, respectively (see the Supporting Information S8 for the derivation processes). By taking a strategy that all the water molecules orientationally distinguished from bulk water are defined as hydration water, the molar concentration of hydration water hyd can 15 ACS Paragon Plus Environment


Page 16 of 45

be expressed as hyd = water − bulk . Accordingly, the number of hydration water molecules per single urea solute (hydration number, 9hyd) is described as

9hyd =

hyd

(4)

As displayed in Figure 5c, the resulting hydration number was revealed to be 9hyd ≈1.9, in close agreement with the result of a previous dielectric spectroscopy studies by Hayashi et al. (9hyd ≈2.3)20 and Agieienko and Buchner (9hyd ≈1.85)56 (note that their definition of hydration water is essentially identical to, but slightly different from ours). However, all these hydration numbers are somewhat smaller than the coordination number 9c =4.1~6.3 of water around urea deduced from the geometrical definition12,44. This inconsistency implies that only a fraction of the water molecules (9hyd ⁄9c =30~46 %) in the first solvation shell can be classified as hydration water, the rest being reorientationally indistinguishable from bulk water. Previous computational and experimental studies have shown that the highly directional urea–water HB interaction induces pronounced orientational retardation of water with the dipole vector direction of urea, compared to that with the out-of-plane and in-plane axis directions7,56. This is partly because the acceptor HB of the oxygen atom in urea, with a average number of 1.5~1.9 8,57, is stronger than water–water HBs (Ourea–Hwater > Owater–Hwater), while the Owater–Hurea and Hwater–Nurea HBs are estimated to be weaker than Owater–Hwater ones5,8,56,57. The formation of stronger acceptor HBs of urea oxygen is corroborated by the positive excess molar heat capacity of the –CO group59. Thus, it can be expected that the retarded orientation of hydration 16 ACS Paragon Plus Environment

Page 17 of 45 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 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


water (9hyd ≈1.9) observed in the present study is mainly caused by the strongly impeded water around the urea oxygen atom. Such retarded water dynamics associated with formation of strong HB are, in some cases, recognized as the kosmotropic nature of urea5. Another suggestive finding from Figure 5c is that the hydration number 9hyd is almost concentration independent up to 5 M. Since urea dimer or oligomer formation involving urea–urea HBs at the expense of urea–water ones will lead to a decrease in the hydration number with increasing urea concentration, the constant 9hyd in this study is interpreted as a negligible aggregation of urea. The absence of urea aggregation is consistent with the low-frequency urea vibration modes that linearly increased with urea concentration (Figure 2) and with earlier studies concluding that a vast number of urea molecules are monomers in aqueous solution21,60. 3.4. HB structure of water around urea.

In accordance with the decomposed Debye and

DHO modes in the THz region, the HB structure of water around urea can be characterized in two ways: (1) HB fragmentation and (2) disordering of the tetrahedral HB structure. Issue (1) can be discussed by focusing on the strength of the w2 relaxation mode w2 , because ∆w2 , which is proportional to the population of NHB water molecules27 on the basis of the assignment proposed by Yada et al.27, serves as an indicator of the HB fragmentation effect of the solute. Meanwhile, most water molecules in the liquid phase are in the tetrahedral-like HB network, and its dynamical fluctuations give rise to the infrared activity of intermolecular stretching S . Therefore, issue (2) can 17 ACS Paragon Plus Environment


Page 18 of 45

be addressed by solute-induced changes of the S mode. From Figure 5b, it is notable that ∆w2 significantly exceeds the virtual line representing the water dilution effect at any urea concentration. Under the assumption that this w2 mode is originated from individual relaxation of NHB water, this result implies an increased number of NHB water molecules due to the urea solute. As a first step to gaining a more quantitative understanding, the molar concentration of NHB bulk water NHB is calculated as34,35

NHB =

∆w2 ⁄0NHB

∆w1 ⁄0HB + ∆w2 ⁄0NHB bulk

(5)

where, bulk is the molar concentration of bulk water determined by Eq. (3). Then, the number of solute-induced NHB water molecules (hereafter referred to as the NHB number: ∆9NHB ) is derived as34,35

∆9NHB =

NHB

water − 12THz mode

1.5 1.0 0.5 High-frequency limit

0.0 1

(b)

0.1

0.1

ACS Paragon Plus 1 Environment

Frequency [THz]

10

8

(a)

Page 42 of 45

(c)

Urea-water relaxation w1 relaxation w2 relaxation Intermolecular stretch Libration

6

Re[ε]

6 4

ΔVS

2

0

10

(b)

Pure water Urea aq. 1.0M Urea aq. 3.0M Urea aq. 5.0M

6 4

Δεw2

4

2

ΔVL+ε∞

(d) Im[ε]

Im[ε]

Re[ε]

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

8


Experiment Fitting

1

2 0 0.1

1

Frequency [THz]

0.1

0.1 ACS Paragon Plus Environment 10

1

Frequency [THz]

10

Page 43 of 45 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 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


ACS Paragon Plus Environment


ACS Paragon Plus Environment

Page 44 of 45

Page 45The of 45 Journal Hydration of Physical water Chemistry 1 2 3 4 5 6

ACS Paragon Plus Environment Non-hydrogen bonded water

Coexistence of Kosmotropic and Chaotropic Impacts of Urea on Water

Recommend Documents