Correlation Between Chain Architecture and Hydration Water

Subscriber access provided by ORTA DOGU TEKNIK UNIVERSITESI KUTUPHANESI

Article

Correlation Between Chain Architecture and Hydration Water Structure in Polysaccharides Michael Grossutti, and John R Dutcher Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.6b00026 • Publication Date (Web): 09 Feb 2016 Downloaded from http://pubs.acs.org on February 19, 2016

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.

Biomacromolecules 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 31

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

Biomacromolecules

Correlation Between Chain Architecture and Hydration Water Structure in Polysaccharides

Submitted to Biomacromolecules

Michael Grossutti and John R. Dutcher

∗

Department of Physics, University of Guelph Guelph, ON N1G 2W1

KEYWORDS: polysaccharides, infrared absorption, water sorption, hydration, network water

∗

Corresponding author: John Dutcher, Department of Physics, University of Guelph, Guelph, Ontario, Canada N1G 2W1; Tel: 519-824-4120, ext. 53950; Fax: 519-836-9967; Email: [email protected] 1 ACS Paragon Plus Environment

Biomacromolecules

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

ABSTRACT

The physical properties of confined water can differ dramatically from those of bulk water. Hydration water associated with polysaccharides provides a particularly interesting example of confined water, since differences in polysaccharide structure provide different spatially confined environments for water sorption. We have used attenuated total reflection infrared (ATR-IR) spectroscopy to investigate the structure of hydration water in films of three different polysaccharides under controlled relative humidity (RH) conditions. We compare the results obtained for films of highly branched, dendrimer-like phytoglycogen nanoparticles to those obtained for two unbranched polysaccharides, hyaluronic acid (HA) and chitosan. We find similarities between the water structuring in the two linear polysaccharides, and significant differences for phytoglycogen. In particular, the results suggest that the high degree of branching in phytoglycogen leads to a much more well-ordered water structure (low density, high connectivity network water), indicating the strong influence of chain architecture on the structuring of water. These measurements provide unique insight into the relationship between the structure and hydration of polysaccharides, which is important for understanding and exploiting these sustainable nanomaterials in a wide range of applications.

2 ACS Paragon Plus Environment

Page 2 of 31

Page 3 of 31

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

Biomacromolecules

Introduction Polysaccharides are an important class of naturally occurring biopolymers that have attracted increasing recent interest because of their potential to replace synthetic polymers derived from depleting petroleum sources, while addressing consumer concerns about the environmental impact of current technologies and products. Polysaccharides are commonly used in a wide range of industrial and biomedical applications, e.g., as moisturizers, rheology modifiers, and emulsifiers,1,2 with the utility of these hydrophilic biopolymers determined to a large extent by their interaction with water. It is clear that a detailed understanding of polysaccharide hydration is key to the development of new products and technologies based on these sustainable materials. Water binds to the polar groups of polysaccharides, which typically results in a threedimensional transient hydrogen bond network in which bonds are constantly breaking and reforming.3 Both the chemical composition and the structure of the polysaccharide affects the hydrogen bond network of the sorbed water molecules. In addition, confinement of the sorbed interstitial water leads to structure and dynamics that differs from that of bulk water.4-6 Infrared (IR) absorption spectroscopy is well suited to the characterization of water structure in polysaccharides, providing insight into the strength and arrangement of chemical bonds. The O-H stretching mode of water is sensitive to the surrounding hydrogen bond environment. In general, hydrogen bonding between two water molecules leads to a weakening of the covalent water O-H bond, causing a decrease in the O-H stretching frequency. Furthermore, the strength of hydrogen bonding depends on the O-H···O bond angle.3,7,8 A hydrogen bond is said to be linear when the hydrogen atom sits along the axis defined by the two oxygen atoms (i.e. the HOO bond angle is 0°). Thus, the O-H stretching frequency also depends


Biomacromolecules

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

on the orientation of the hydrogen bonds, and is sensitive to distortions of hydrogen bond linearity. An increase (decrease) in the number, linearity, or strength of hydrogen bonding results in a decrease (increase) in the frequency of the v(OH) IR absorption band. Thus, the IR spectrum of water provides semi-quantitative molecular level information regarding the structural arrangement of the water hydrogen bond network.8-10 In the present study, we used attenuated total reflection infrared (ATR-IR) spectroscopy to measure IR absorption spectra of films of three naturally occurring polysaccharides. We compare the results obtained for two linear molecules, hyaluronic acid (HA) and chitosan, with those obtained for a highly branched, dendrimeric molecule, phytoglycogen. The physical properties of HA and chitosan have been well characterized,11-21 but those of phytoglycogen are not as well known. Although significant work has been done to understand the basic structure of glycogen-like molecules,22-28 as well as the action of enzymes,29-31 a detailed understanding of the interaction between water and the highly branched, hydrophilic structure of the particles remains elusive. Recently, we used neutron scattering to study the structure and hydration of monodisperse phytoglycogen nanoparticles.32 We found that the phytoglycogen nanoparticles are very highly hydrated, containing between 250 and 285% of their weight in water, and that the dynamics of hydration water inside the nanoparticles was significantly slowed down. The neutron scattering study did not, however, provide information about the structure of the hydration water, and this is the focus of the present manuscript. By calculating the difference between ATR-IR spectra measured at different relative humidity (RH) values and that measured at a reference low RH value, we eliminate the chemical specificity of the OH band feature, allowing us to isolate and compare the structuring of water in the different polysaccharides. We find striking similarities between water structuring in the two linear polysaccharides, and


Page 4 of 31

Page 5 of 31

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

Biomacromolecules

significant differences between the linear molecules and highly branched phytoglycogen. In particular, the phytoglycogen nanoparticles exhibited a high degree of highly ordered network water connectivity, and a large increase in the fraction of disordered multimer water clusters with increasing RH, whereas the water structure for HA and chitosan was found to be relatively insensitive to changes in RH. These measurements suggest that the high degree of branching in phytoglycogen leads to a much more well-ordered water structure, indicating the strong influence of chain architecture on the structuring of water. The results of the present study provide new insights that are key to fully understanding and exploiting these materials in new technologies and therapies.

Experimental Section Polysaccharide Materials Three different polysaccharides were used in the present study: phytoglycogen, hyaluronic acid (HA) and chitosan. Phytoglycogen is a highly branched polysaccharide, produced in the form of monodisperse nanoparticles by some varieties of plants. It consists of linear chains of α-1,4 linked D-glucose units that undergo α-1,6 branching every 10-12 monomer units.33-35 Monodisperse phytoglycogen nanoparticles (hydrated molecular weight MW = 14.7 × 106 g/mol)32 were extracted and purified from sweet corn (provided by Mirexus Biotechnologies Inc.). Further details on the extraction, purification and characterization of the phytoglycogen nanoparticles can be found in ref. 32. Hyaluronic acid (HA) is a linear polysaccharide composed of alternating β-1,4 linked D-glucoronic acid and β-1,3 linked N-acetyl-D-glucosamine units. The hydration properties of HA have been extensively studied because of its use in medical and cosmetic applications such as moisturizers, lubricants, wound healing, and bodily fluid


Biomacromolecules

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

substitutes.11-17 HA (molecular weight MW = 1.3 × 106 g/mol) was obtained from Making Cosmetics Inc. and was used as received. Chitosan is a linear polysaccharide composed of randomly distributed β-1,4 linked D-glucosamine (deacetylated unit) and N-acetyl-Dglucosamine (acetylated unit), and it is characterized by its degree of acetylation (DA), with DA < 50% corresponding to chitosan and DA > 50% corresponding to the related polysaccharide, chitin. Its high cost of production typically limits chitosan to specialty, niche applications such as wound dressings, antimicrobial coatings and drug delivery.21,36-41 Chitosan (molecular weight MW = 110 × 103 - 250 × 103 g/mol), DA = 24%) was obtained from Sigma Aldrich and used as received.

ATR-IR Experimental Methods Solutions of 1% w/w phytoglycogen nanoparticles and hyaluronic acid (sodium salt) were prepared using Milli-Q water (resistivity of 18.2 MΩ-cm), and 1% w/w chitosan in 5% acetic acid. To prepare the polysaccharide films (thickness > 3 µm), solutions were spread onto a ZnSe-diamond ATR crystal and the water was evaporated under a stream of dry nitrogen gas. All spectra were collected on a Bruker Vertex 70 FTIR spectrometer equipped with an MCT-A detector using an incident angle of 45°. The spectra were acquired at a resolution of 4 cm-1. This ensures data point collection every 2 cm-1, and allows for the determination of the band center position with an uncertainty of 0.5 cm-1.42 The ATR-IR spectra are expressed in absorbance units defined as A = -log(I/I0), where I and I0 are the infrared single beam intensities of the film at controlled relative humidity and the bare ATR crystal background, respectively. The relative humidity (RH) was controlled by adjusting the relative amounts of constant flows of dry nitrogen gas and water-saturated nitrogen gas over the films, with the RH value measured using a


Page 6 of 31

Page 7 of 31

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

Biomacromolecules

Sensirion SHT71 humidity sensor.18 Polysaccharide ATR-IR spectra were collected in RH steps of ~10% between 0-90% humidity at ~25 °C. The sample was allowed to equilibrate for at least 20 minutes at each RH setting prior to collecting the ATR-IR spectrum. Longer time-scale experiments were also performed to determine the equilibration time necessary to achieve steady-state spectra. The broad O-H stretching absorbance band of sorbed water is the result of overlapping components. Fourier self-deconvolution was used to determine the frequencies of each component (band enhancement factors of 1.6-2.0 were used). The absorbance for a given component was calculated by integrating in a 40 cm-1 window around the component center, as determined by the Fourier self-deconvolution procedure. The center of gravity of the v(OH) band was determined for the 3800-2950 cm-1 frequency range.

Results and Discussion Water molecules can form up to four hydrogen bonds, which allows for different transient local structural arrangements.43-45 Numerous experiments and simulations show evidence of different local structures in bulk liquid water that arise from variations in the arrangement of hydrogen bonds.43-45 The ATR-IR spectrum of the O-H stretching mode of bulk liquid water is shown in Fig. 1. The shape of the v(OH) band for bulk liquid water is complex, as it depends on the transient hydrogen bond network of the water molecules, and it consists of two prominent shoulders near 3225 cm-1 and 3400 cm-1, and a much weaker high frequency shoulder near 3600 cm-1. Simple models attributing the characteristic shape of the v(OH) band to subpopulations of water molecules in different hydrogen bonding environments have been developed and used to describe both bulk and confined water structure.8-10,46-60 The low


Biomacromolecules

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

frequency shoulder observed near 3225 cm-1 is attributed to collective in-phase O-H stretching vibrations of tetrahedrally coordinated water molecules, i.e. water pentamers, experiencing strong intermolecular coupling via linear hydrogen bonds.9,10,46 The shoulder observed near 3400 cm-1 is attributed to out-of-phase O-H stretching vibrations of water molecules which are more weakly coupled in a distorted tetrahedral structure that includes some nonlinear (or absent) hydrogen bonds, i.e. distorted pentamers and tetramers.9,10,46 Both of these components are associated with network water, i.e. water in a hydrogen bonded network. The weak high frequency shoulder near 3600 cm-1 is attributed to water molecules in a highly disturbed hydrogen bond network. This type of water is sometimes termed multimer water,9,51,52 which consists of water clusters containing two or less hydrogen bonds per water molecule.9,10,46-48,52 The structure of the water hydrogen bonding network and the relative populations of the different water types can be modified from that of bulk liquid water upon sorption by polysaccharides. In the present study, we use this well established model of water structuring to interpret the shape of the v(OH) band arising from interstitial water molecules sorbed by phytoglycogen, HA, and chitosan films.


Page 8 of 31

Page 9 of 31

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

Biomacromolecules

Figure 1: The ATR-IR spectrum of the O-H stretching region measured for bulk liquid water at ~25 °C. The arrows indicate the positions of the network and multimer water sub-populations.

ATR-IR Spectroscopy of Polysaccharide Films In Fig. 2a-c we show the relative humidity (RH) dependence of ATR-IR spectra of phytoglycogen, HA, and chitosan films within the 3800-2600 cm-1 range, corresponding to the v(OH) band. The IR absorbance in this region is dominated by the O-H stretching vibrations of sorbed water, but also contains contributions due to O-H stretching vibrations arising from the pyranose ring hydroxyl groups (~3600-3000 cm-1), various C-H stretching vibrations (~30002800 cm-1) and, in the case of HA and chitosan, N-H stretching vibrations (~3300-3000 cm-1). Because of their different chemical compositions, the spectra shown in Fig. 2a-c for the different polysaccharides have significantly different shapes. For each polysaccharide, the IR absorbance in the 3600-3000 cm-1 region increases with increasing RH (Fig. 2), corresponding to the uptake of water. As described above, each of the ATR-IR spectra presented in Fig. 2a-c is a convolution of IR bands arising from the vibrational modes of the sorbed water and the polysaccharide. To remove the contributions of the polysaccharide vibrational modes from the IR spectra, difference spectra were calculated by subtracting the absorbance spectrum measured at the lowest RH (< 4% RH, which we refer to as 0% RH) from spectra collected at higher RH values. In this way, information specific to the nature of the sorbed water can be obtained. This approach has been used to study water in HA films and other systems.9,13-17,49-51


Biomacromolecules

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 10 of 31

Figure 2: ATR-IR absorbance spectra of a) phytoglycogen, b) hyaluronic acid, and c) chitosan films deposited on a ZnSe-diamond internal reflection element; and difference spectra [∆A = A(RH%) - A(0%)] for d) phytoglycogen e) hyaluronic acid, and f) chitosan. The arrows indicate the progression of the spectra with increasing relative humidity.

In Fig. 2d - f, we show difference spectra for the phytoglycogen, hyaluronic acid, and chitosan films as a function of RH. These sequences of spectra are remarkably similar for the two linear polysaccharides, and significantly different for the highly branched phytoglycogen, suggesting differences in the water hydrogen bonding network. The center of gravity (COG) of these v(OH) bands reflects the average frequency distribution of the O-H stretching vibrations, and serves as a measure of the average strength of hydrogen bonding for the sorbed interstitial water molecules. In Fig. 3 we show the COG for the v(OH) band of the water sorbed by each of the polysaccharide films. An increase (decrease) in the frequency of the v(OH) band is indicative 10 ACS Paragon Plus Environment

Page 11 of 31

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

Biomacromolecules

of a decrease (increase) in the degree of hydrogen bonding. For phytoglycogen, the v(OH) COG remains nearly unchanged at 3263 cm-1 before increasing slightly (by 3 cm-1) above 80% RH. The results for HA are similar, with the v(OH) COG remaining nearly constant at 3317 cm-1 throughout the studied RH range, which is consistent with previous literature studies of HA.17 In contrast, after remaining essentially constant as the RH was increased from 0 to 40% RH, chitosan exhibits a significant (40 cm-1) increase in the v(OH) band COG between 40% and 60% RH, and remains essentially constant for larger RH values. The significant increase in the v(OH) band COG with increasing RH between 40% and 60% RH for the chitosan film can be attributed to the occurrence of a humidity driven glass transition. The sorption of water by hydrophilic polymers can have a plasticizing effect, increasing the mobility of the polymer chains and decreasing the glass transition temperature (Tg) of the polymer.61,62 The resulting changes to the mechanical and barrier properties of hydrophilic polymers have been exploited in a variety of applications including food processing and packaging, and controlled delivery of bioactives.19 At constant temperature, water sorption can cause a glass transition to occur. In the case of chitosan, previous studies using dynamic mechanical thermal analysis and differential scanning calorimetry have shown that a glass transition occurs at ~25°C for water contents of ~15-20% by mass.19,20 Our gravimetric water sorption isotherm (25°C) data for chitosan (Fig. S1, Supporting Information) show that the film is ~15-20% water by mass at ~50% RH, suggesting that a glass transition occurs near this RH value. Therefore, our observed increase in the v(OH) band COG with increasing RH between 40% and 60% RH for the chitosan film (Figure 3), corresponding to a decrease in the degree of hydrogen bonding of the interstitial water molecules, is consistent with previously published Tg data for chitosan.19,20


Biomacromolecules

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 12 of 31

It is interesting to note that at large RH values, after the glass transition has occurred in chitosan, the band shape and COG of the v(OH) band for both linear polysaccharides are similar to each other and to those for bulk liquid water (COG ~3320 cm-1). This suggests that the average hydrogen bond strength of the water sorbed by the linear polysaccharides is similar and more closely resembles bulk liquid water than that sorbed by highly branched phytoglycogen, which exhibits stronger average hydrogen bonding.

Figure 3: The center of gravity of the v(OH) band of sorbed water in films of phytoglycogen (red), hyaluronic acid (blue), and chitosan (black).

To further quantify the spectral changes, it is useful to calculate absorbance ratios that compare the intensity of the IR components that arise from the different water sub-populations. The absorbance of each water band component was calculated by integrating a 40 cm-1 window about each of the three component centers, as identified by Fourier self-deconvolution. The integrated area is proportional to the number of absorbing water molecules in the spectral range.9 Using these absorbance values, the spectral parameters Rnetwork (Rnetwork = A3225/A3400) and Rmultimer 12 ACS Paragon Plus Environment

Page 13 of 31

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

Biomacromolecules

(Rmultimer = A3600/A3225) were calculated for each spectrum. These parameters provide information about the relative populations of the different water types and are known to be sensitive to the structural arrangement of hydrogen bonds.9,10,47-49,52-58 The parameter Rnetwork describes the hydrogen bond structure of the network water, and serves as a measure of the distortion of the hydrogen bond network and the degree of intermolecular coupling between water molecules. For example, the cooling of bulk water causes a progressive increase in Rnetwork as the water molecules adopt a more rigid tetrahedral structure with less hydrogen bond defects.48,53 The parameter Rmultimer provides a measure of the relative sub-populations of water molecules in a disrupted hydrogen bond network (multimer water) to those in a well-ordered hydrogen bond network (network water). Because of the complex shape of the v(OH) band, the parameters Rnetwork and Rmultimer serve as a convenient means of comparing hydration between the different polysaccharides. In Fig. 4 we plot Rnetwork and Rmultimer as a function of RH for the phytoglycogen, HA, and chitosan films. We see that the Rnetwork values calculated for the linear polysaccharides are similar, and significantly smaller than those measured for highly branched phytoglycogen. We also see that the Rmultimer values for the linear polysaccharides are significantly larger than for phytoglycogen, indicating a more disordered hydrogen bonding network for the linear polysaccharides. These general differences in Rnetwork and Rmultimer between the linear and highlybranched polysaccharides reflect the clear differences in the v(OH) band shape that are evident in Fig. 2d-f. We now discuss the Rnetwork and Rmultimer values for the three polysaccharides in more detail. For phytoglycogen, there is a small increase in Rnetwork from ~2.15 at 0% RH to a plateau value of ~2.3 at 40% RH, suggesting that the sorbed water fills in the well-ordered primary


Biomacromolecules

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 14 of 31

hydration shell hydrogen bond network. The corresponding value of Rmultimer increases continuously from ~0.21 to ~0.27 with increasing humidity, showing that the multimer water fraction is increasing with increased water sorption. The greatest increase occurs at low RH between 0-30% RH, after which it remains essentially constant before increasing again above 80% RH. It is interesting to note that both Rnetwork and Rmultimer increase most steeply between 040% RH. This implies that in the early stages of water sorption, the water molecules are incorporated into both the well-ordered (network) and disordered (multimer) water subpopulations. The Rnetwork and Rmultimer values for the linear polysaccharides HA and chitosan agree closely with one another, reflecting the resemblance between their respective difference spectra (Fig. 2e and f). For HA, Rnetwork and Rmultimer remain essentially constant at values of ~1.27 and ~0.37 respectively throughout the entire RH range. This suggests that neither the network water connectivity nor the multimer water fraction changes for HA with increased water sorption, which is consistent with the essentially constant value of the COG of the v(OH) band (Fig. 3). For chitosan, Rnetwork remains essentially constant at a value of ~1.4 as the RH is increased to ~40% RH, and then decreases to a value of ~1.2 at 60% RH, after which it remains essentially constant. This suggests that the order and connectivity of the network water is decreased following the glass transition that occurs as the RH is increased from 40% RH to 60% RH, as discussed earlier. Rmultimer for chitosan increases significantly (~0.34 to ~ 0.4) in passing through the glass transition, suggesting that increased polysaccharide chain motion leads to an increased fraction of multimer water. Together, the values of Rnetwork and Rmultimer show that the chitosan glass transition induces disorder in the water structure by both perturbing the network water connectivity and increasing the multimer water fraction. This is consistent with a decrease in the


Page 15 of 31

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

Biomacromolecules

average hydrogen bonding strength, as indicated by the decrease in the v(OH) band COG for the chitosan film (Fig. 3). Such behavior has been observed for sorbed water at the gel-to-liquid phase transition of phospholipid multibilayers, for which the disordering of the phospholipid polar headgroups perturbs the water structure.9 We also note that the values of Rnetwork measured for all three polysaccharides are larger than the value of Rnetwork ~0.95 measured for bulk liquid water at 25 °C, with the values for phytoglycogen being significantly larger. The enhanced values of Rnetwork for the polysaccharides, in particular phytoglycogen, indicate stronger hydrogen bonding that could correspond to a greater fraction of linear hydrogen bonds relative to that of bulk liquid water. 7,8

Figure 4: a) Rnetwork and b) Rmultimer for water sorbed by phytoglycogen (red) and hyaluronic acid (blue) and chitosan (black) films.

The difference spectra in Fig. 2d-f and the spectral parameters in Fig. 4 contain information about the average interstitial water structure in the polysaccharide films at a given RH. Further details of the water structure can be obtained by calculating difference spectra between successive RH steps (Fig. S2, Supporting Information). In this way, incremental 15 ACS Paragon Plus Environment

Biomacromolecules

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 16 of 31

changes in sorbed water structure can be resolved. This approach has been used to gain insight into the hydration characteristics of biological polymers and lipid films.9,17 The RH dependence of the incremental ∆Rnetwork and ∆Rmultimer values for the three polysaccharides are shown in Fig. 5. These plots magnify the changes in water structure reported in Fig. 4, emphasizing the changes induced by each RH step, and we focus on the main features in these incremental data. For phytoglycogen, ∆Rnetwork has a maximum value at ~30% RH, indicating that at low RH values the sorbed water molecules are contributing to an increasingly ordered and connected hydrogen bond network. This occurs together with an increase in ∆Rmultimer at low RH values, indicating that both highly ordered and disordered water structure forms. For RH >80%, there is an increase in ∆Rmultimer and a corresponding decrease in ∆Rnetwork. This indicates that the water sorbed at large RH values is disordered, which is consistent with the observed decrease in the average strength of hydrogen bonding by sorbed water molecules in the phytoglycogen film at high RH (as shown by the increase in the v(OH) band COG frequency in Fig. 3). For HA, the incremental ∆Rnetwork and ∆Rmultimer parameters exhibit the same behavior (and numerical values) as the average water structure parameters presented in Fig. 4. This is consistent with the observed insensitivity to RH for water sorbed by the HA film and indicates that the incremental water is incorporated into a similar hydrogen bonding environment throughout the sorption process. The RH dependence of the incremental ∆Rnetwork and ∆Rmultimer parameters for chitosan are similar to those for the other linear polysaccharide HA, with the exception that, near the chitosan glass transition (~50% RH), both Rnetwork and Rmultimer deviate from the values observed for the HA film. The decrease in ∆Rnetwork and the corresponding increase in ∆Rmultimer near the glass transition show that the water molecules that are sorbed during the glass transition are 16 ACS Paragon Plus Environment

Page 17 of 31

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

Biomacromolecules

joining a more disordered hydrogen bond network. These observations are consistent with the notion of increased chain motion and conformational rearrangement inducing disorder in the water hydrogen bond network. Finally, it is interesting to note that the incremental ∆Rmultimer values for the three polysaccharides essentially converge at ~90% RH. This suggests that, as the polysaccharides approach saturation at very high humidity, the fraction of water molecules sorbing as multimer water becomes very similar.

Figure 5: The parameters a) ∆Rnetwork and b) ∆Rmultimer for the incremental water sorbed between RH steps by phytoglycogen (red), hyaluronic acid (blue), and chitosan (black) films.

Together, the data presented in Figs. 4 and 5 provide a detailed description of water structure in the polysaccharide films as a function of RH. They demonstrate that the interstitial water in phytoglycogen experiences stronger hydrogen bonding, as indicated by both greater network water connectivity and less disordered multimer water, compared to the interstitial water in the linear polysaccharides.


Biomacromolecules

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 18 of 31

Previous studies on confined interstitial water in polymer, micelle, liposome, and nanoporous glass systems have found differences in water structure compared to bulk liquid water. These differences have been found to depend on the size of the interstitial spatial environment occupied by the water molecules,53-58 and are insensitive to interstitial chemical properties such as charge and polar group identity.46,54-57 Generally, the confinement of water in small spaces introduces hydrogen bond defects. In reverse micelles and liposomes, network water connectivity has been observed to decrease as the water molecules were confined in progressively smaller spaces (a few nm).53,54,56 In cross-linked poly(acrylamide) gels, network water connectivity decreases as cross-linking density is increased. As with the micellar and liposomal systems, this behavior was attributed to the decreased interstitial volume in the highly cross-linked gels that induces hydrogen bond defects.55,58,59,63 However reducing the interstitial volume does not always lead to reduced water connectivity. The opposite trend was observed for water confined in cylindrical glass nanopores; network water connectivity was found to increase as the cylinder radius was decreased (from 320 nm to 8 nm).51 In this system, the reduction in interstitial space is most pronounced in one dimension (the cylindrical pore radius). For the cylindrical nanopores, increased confinement stabilized the network water hydrogen bond structure across long distances along the cylinder axis.51 These results suggest that the significant differences in water structure observed between the highly-branched phytoglycogen and the linear polysaccharides HA and chitosan in the present study are predominately due to differences the interstitial spatial environment, as dictated by the chain architecture, rather than differences in the pyranose ring substituents. The denser highly branched phytoglycogen chain architecture may better promote, via narrower interstitial


Page 19 of 31

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

Biomacromolecules

spacing, a more ordered water hydrogen bond network along the axis of the polysaccharide chain than the larger spacing associated with the linear chain architecture. The large differences in water structure that we have observed between the linear and highly branched polysaccharides have direct implications for their applications. The larger values of Rnetwork for phytoglycogen (Figure 4), indicating stronger bonding of water with the highly branched polysaccharide than for the linear polysaccharides, means that phytoglycogen will less readily sorb and desorb water in response to changes in RH. This provides an advantage for the use of phytoglycogen as a long term moisturizing agent in cosmetics, with more consistent delivery of moisture in environments in which considerable changes in RH occur. The high degree of water structuring for phytoglycogen will also contribute to its other remarkable physical properties such as the low viscosity and exceptional stability of aqueous dispersions.32

Conclusions Polysaccharide hydration, a key property that determines their physical characteristics and applications, depends on different sorption processes, such as the strong binding of water to various polysaccharide polar groups, and the extent of water layering that is determined by the spatial arrangement of the monomers. The relative contributions of different sorption processes can vary with RH, and this is also reflected in the structure of the sorbed interstitial water. In this work, ATR-IR spectroscopy was used to investigate the structure of water in films of phytoglycogen, hyaluronic acid, and chitosan under controlled relative humidity conditions. This detailed comparison has revealed strong similarities between the results for the linear polysaccharides, and significant differences between the results for the linear polysaccharides and highly branched phytoglycogen. The data show that the interstitial water sorbed by phytoglycogen is, on average, more strongly hydrogen bonded, with a greater fraction of the 19 ACS Paragon Plus Environment

Biomacromolecules

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 20 of 31

water molecules incorporated into a well-ordered hydrogen network, compared to the interstitial water in the linear polysaccharides (Rnetwork is, on average, a factor of 2 larger for phytoglycogen than for the two linear polysaccharides). This has allowed us to attribute significant differences in the ordering of the sorbed water to differences in chain architecture, which provides important insight into the physical properties of sustainable polysaccharide materials and their use in a wide range of applications.

Supporting Information Four pages of Supporting Information containing additional technical details and 2 supporting data figures are included. This material is available free of charge via the Internet at http://pubs.acs.org.

Acknowledgements Mirexus Biotechnologies Inc. generously supplied the monodisperse phytoglycogen nanoparticles. We thank Jacek Lipkowski and Phil Whiting for their valuable comments on the manuscript. This work was funded by grants from the Ontario Ministry of Agriculture (OMAF) and the Natural Sciences and Engineering Research Council (NSERC) of Canada. J.R.D. is the recipient of a Senior Canada Research Chair in Soft Matter and Biological Physics.

References 1. Williams, P.A. Renewable Resources for Functional Polymers and Biomaterials: Polysaccharides, Proteins and Polyesters. RSC Publishing, 2011. 2. Kaplan, D.L. Biopolymers from renewable resources. Springer, 1998. 3. Ball, P. Chem. Rev. 2008, 108, 74-108.


Page 21 of 31

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

Biomacromolecules

4. Brovchenko, I.; Oleinikova, A. Interfacial and Confined Water. Elsevier, 2008. 5. Pissis, P.; Kyritsis, P. J. Polym. Sci. Polym. Phys. 2013, 51, 159-175. 6. Bellissent-Funel, M. C. J. Phys.-Condens. Mat. 2001, 13, 9165-9177. 7. Sharp, K. A.; Vanderkooi, J. N. Acc. Chem. Res. 2010, 43, 231−239. 8. Disalvo, E. A.; Frias, M. A. Langmuir 2013, 29, 6969−6974. 9. Binder, H. Eur. Biophys. J. 2007, 36, 265-279. 10. Liu, D.; Ma, G.; Levering, L.M.; Allen, H.C. J. Phys. Chem. B 2004, 108, 2252-2260. 11. Kogan, G.; Soltes, L.; Stern, R.; Gemeiner, P. Biotechnol. Lett. 2007, 29, 17-25. 12. Alber, C.; Engblom, J.; Falkman, P.; Kocherbitov, V. J. Phys. Chem. B 2015, 119, 42114219. 13. Marechal, Y. J. Mol. Struct. 2004, 700, 217-223. 14. Haxaire, K. Marechal, Y.; Milas, M. Rinaudo, M. Biopolymers 2003, 72, 10-20. 15. Haxaire, K. Marechal, Y.; Milas, M. Rinaudo, M. Biopolymers 2003, 72, 149-161. 16. Maréchal, Y., M. Milas, and M. Rinaudo. Biopolymers 2003, 72, 162-173. 17. Servaty, R.; Schiller, J.; Binder, H.; Arnold, K. Int. J. Biol. Macromol. 2001, 28, 121-127. 18. Murray, C.A.; Dutcher, J.R. Biomacromolecules 2006, 7, 3460-3465. 19. Lazaridou, A.; Biliaderis, C.G. Carbohyd. Polym. 2002, 48, 179-190. 20. Ratto, J.; Katakeyama, T.; Blumstein, R.B. Polymer 1995, 36, 2915-2919. 21. Kumar, M.N.V.R. React. Funct. Polym. 2000, 46, 1-27. 22. Manners, D. J. Carbohyd. Polym. 1991, 16, 37-82. 23. Meléndez, R.; Meléndez-Hevia, E.; Canela, E. I. Biophys. J. 1999, 77, 1327-1332. 24. Meléndez, R.; Meléndez-Hevia, E.; Mas, F.; Mach, J.; Cascante, M. Biophys. J. 1998, 75, 106-114. 25. Powell, P. O.; Sullivan, M. A.; Sheehy, J. J.; Schulz, B. L.; Warren, F. J.; Gilbert, R. G. PloS One 2015, 10, e0121337. 21 ACS Paragon Plus Environment

Biomacromolecules

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

26. Huang, L.; Yao, Y. Carbohyd. Polym. 2011, 83, 1665-1671. 27. Wong, K.-S.; Kubo, A.; Jane, J.-l.; Harada, K.; Satoh, H.; Nakamura, Y. J. Cereal Sci. 2003, 37, 139-149. 28. Yun, S.-H.; Matheson, N. K. Carbohyd. Res. 1993, 243, 307-321. 29. Kajiura, H.; Takata, H.; Kuriki, T.; Kitamura, S. Carbohyd. Res. 2010, 345, 817-824. 30. Putaux, J.-L.; Potocki-Véronese, G.; Remaud-Simeon, M.; Buleon, A. Biomacromolecules 2006, 7, 1720-1728. 31. Takata, H.; Kajiura, H.; Furuyashiki, T.; Kakutani, R.; Kuriki, T. Carbohyd. Res. 2009, 344, 654-659. 32. Nickels, J.D.; Atkinson, J.; Papp-Szabo, E.; Stanley, C.; Diallo, S.O.; Perticaroli, S.; Baylis, B.; Mahon, P.; Ehlers, G.; Katsaras, J.; Dutcher, J.R. Biomacromolecules 2016, DOI: 10.1021/acs.biomac.5b01393. 33. Powell, P.O.; Sullivan, M.A.; Sweedman, M.C.; Stapleton, D.I.; Hasjim, J.; Gilbert, R.G. Carbohyd. Polym. 2014, 101, 423-431. 34. Putaux, J.; Bule, A.l Borsali, R. Chanzy, H. Int. J. Biol. Macromol. 1999, 26, 145-150. 35. Ball, S.; Guan, H.-P.; James, M.; Myers, A.; Keeling, P. Mouille, G. Bule, A.l Colonna, P. Preiss, J. Cell 1996, 86, 349-352. 36. Fan, W.; Yan, W.; Xu, Z.; Ni, H. Colloids and Surfaces B: Biointerfaces 2012. 90, 21-27. 37. Qi, L.; Xu, Z.; Jiang, X.; Hu, C.; Zou, X. Carbohydrate Research 2004, 339, 2693-2700. 38. Xu, Y.; Du, Y. International Journal of Pharmaceutics 2003, 250, 215-226. 39. Janes, K.A.; Fresneau, M.P.; Marazuela, A.; Fabra, A.; Alonso, M.J. J. Controlled Release 2001, 73, 255-267. 40. Jayakumar, R.; Prabaharan, M.; Kumar, P.T.S.; Nair, S.V.; Tamura, H. Biotechnology advances 2011, 29, 322-337. 41. Ohkawa, K. Minato, K.; Kumagai, G.; Hayahi, S.; Yamamoto, H. Biomacromolecules 2006, 7, 3291-3294. 42. Griffith, P.R.; de Haseth, J.E. Fourier Transform Infrared Spectrometry; John Wiley and Sons: New York, 1986.


Page 22 of 31

Page 23 of 31

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

Biomacromolecules

43. Nilsson, A.; Pettersson, L.G.M. Nature Communications 2015, 6, 1-11. 44. Nilsson, A.; Pettersson, L.G.M. Chem. Phys. 2011, 389, 1-34. 45. Mahoney, M.W.; Jorgensen, W.L. J. Chem. Phys. 2000, 112, 8910-8922. 46. Marinov, V.S.; Nickolov, Z.S.; Matsuura, H. J. Phys. Chem. B 2001, 105, 9953-9959. 47. Leikin, S.; Parsegian, V.A.; Yang, W.-H.l Walrafen, G.E. Proc. Natl. Acad. Sci. USA 1997, 94, 11312-11317. 48. Walrafen, G.E.; Chu, Y.C. J. Phys. Chem. 1995, 99, 11225-11229. 49. Ide, M.; Mori, T.l Ichikawa, K.; Kitano, H. Tanaka, M.; Mochizuki, A.; Oshiyama, H.; Mizuno, W. Langmuir 2003, 19, 429-435. 50. Ping, Z.H.; Nguyen, Q.T.; Chen, S.M.; Zhou, J.Q.; Ding, Y.D. Polymer 2001, 42, 8461-8467. 51. Le Caër, S.; Pin, S.; Esnouf, S.; Raffy, Q.; Renault, J.Ph.; Brubach, J.-B.; Creff, G.; Roy, P. Phys. Chem. Chem. Phys. 2011, 13, 17658-17666. 52. Uchida, T.; Osawa, M.; Lipkowski, J. J. Electroanal. Chem. 2014, 716, 112-119. 53. Bey Temsamani, M.; Maeck, M; Hassani, E.; Hurwitz, H.D. J. Phys. Chem. B 1998, 102, 3335-3340. 54. Lafleur, M.; Pigeon, M.; Pezolet, M. J. Phys. Chem. 1989, 93, 1522-1526. 55. Maeda, Y.; Kitano, H. Spectrochim. Acta A 1995, 51, 2433-2446. 56. Brubach, J.-B.; Mermet, A.; Filabozzi, A.l Gerschel, A.; Lairez, D.; Krafft, M.P., Roy, P. J. Phys. Chem. B 2001, 105, 430-435. 57. Boissiere, C.; Brubach, J.-B.; Mermet, A.; de Marzi, G.; Bourgaux, C.; Prouzet, E.; Roy, P. J. Phys. Chem. B 2002, 106, 1032-1035. 58. Terada, T.; Ibaba, T.; Kitano, H.; Maeda, Y.; Tsukida, N. Macromol. Chem. Phys. 1994, 195, 3261-3270. 59. Maeda, Y.; Idem M.; Kitano, H. J. Mol. Liq. 1999, 80, 149-163. 60. Brubach, J.-B.; Mermet, A,; Filabozzi, A.; Gersschel A.; Roy, P. J. Chem. Phys. 2005, 122, 184509. 61. Hancock, B.C.; Zografi, G. Pharm. Res. 1994, 11, 471-477.


Biomacromolecules

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

62. Xiado, Y.; Carter, B.P.; Schmidt, S.J. J. Food Science 2011, 76, E78-E89. 63. Terada, T.; Maeda, Y.; Kitano, H. J. Phys. Chem. 1993, 97, 3619-3622.


Page 24 of 31

Page 25 of 31

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

Biomacromolecules

For Table of Contents Use Only


Biomacromolecules

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

Figure 1: The ATR-IR spectrum of the O-H stretching region measured for bulk liquid water at ~25 °C. The arrows indicate the positions of the network and multimer water sub-populations. 84x63mm (220 x 220 DPI)

ACS Paragon Plus Environment

Page 26 of 31

Page 27 of 31

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

Biomacromolecules

Figure 2: ATR-IR absorbance spectra of a) phytoglycogen, b) hyaluronic acid, and c) chitosan films deposited on a ZnSe-diamond internal reflection element; and difference spectra [∆A = A(RH%) - A(0%)] for d) phytoglycogen e) hyaluronic acid, and f) chitosan. The arrows indicate the progression of the spectra with increasing relative humidity. 168x95mm (220 x 220 DPI)


Biomacromolecules

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

Figure 3: The center of gravity of the v(OH) band of sorbed water in films of phytoglycogen (red), hyaluronic acid (blue), and chitosan (black). 164x127mm (150 x 150 DPI)


Page 28 of 31

Page 29 of 31

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

Biomacromolecules

Figure 4: a) Rnetwork and b) Rmultimer for water sorbed by phytoglycogen (red) and hyaluronic acid (blue) and chitosan (black) films. 164x69mm (220 x 220 DPI)


Biomacromolecules

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

Figure 5: The parameters a) ∆Rnetwork and b) ∆Rmultimer for the incremental water sorbed between RH steps by phytoglycogen (red), hyaluronic acid (blue), and chitosan (black) films. 306x124mm (150 x 150 DPI)


Page 30 of 31

Page 31 of 31

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

Biomacromolecules


Correlation Between Chain Architecture and Hydration Water

Recommend Documents