Notice: Wiley Online Library will be unavailable on Saturday 27th February from 09:00-14:00 GMT / 04:00-09:00 EST / 17:00-22:00 SGT for essential maintenance. Apologies for the inconvenience.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
A polyamide consists of a repetition of an ethylene sequence and an amide group. The former is hydrophobic, whereas the latter is hydrophilic. The polymer is usually partially crystalline. The hydrophobicity and hydrophilicity of the constituents and the crystallinity of the polymer are the three most important factors that determine the water sorption property. To understand the effect of the sorbed water on the physicochemical properties, we must study the interactions on a molecular basis. Infrared spectroscopy is one of the most powerful methods of studying the molecular mechanism involved in the interactions between polymers and water because the frequencies and shapes of the OH absorptions of water are very sensitive to the interactions.1, 2 In addition, recent advances in Fourier transform infrared (FTIR) spectrometers have made it possible to sensitively detect very small absorption changes caused by the interactions.3
The influence of the interactions with water on the properties of polyamides has been studied by several authors.4, 5 Vergelati et al.4 reported that water induced the crystalline transition of nylon-6, 6. Fukuda et al.5 applied near-infrared spectroscopy to the study of the sorption behavior of water in nylon-6. Kusanagi and Yukawa6 studied the infrared spectra of the water sorbed in various polymers, including nylon-6. They found that the OH stretching absorptions of the sorbed water especially varied to a large extent in the frequencies from one polymer to another.
We have been studying the interactions of polymers with water by following the FTIR spectra of a water-containing sample during dehydration under the flow of dry air. This method is useful for studying the interactions because very small spectral changes caused by the interactions can be revealed by the spectral subtraction technique. It has been recently found by this method that water in perfluorinated Nafion membranes exists as oxonium ions in equilibrium with free water, sulfonate ions, and sulfonic acid.7 In this study, we have applied this method to the study of the interactions of nylon-6 with water. An analysis of the observed spectral changes has led to the conclusion that water penetrates the amorphous portion of the sample to hydrate the free (i.e., non-hydrogen-bonded CO or NH) amide groups. This is the first case, to our knowledge, in which the interacting counterpart for water has been definitely identified.
The source pellet sample of nylon-6 used in this study was T-814, a product of Toyobo Co., Ltd., which has common characteristics. The film samples for infrared spectral measurements were prepared by the drawing of the melt of the pellets. A 5-μm-thick film sample, which gave absorptions of good quality even for the very strong amide I and II bands, was mainly used in this study. A 17-μm-thick sample, which could hold more water for a longer time, was also used to measure the spectrum of the sorbed water in a more hydrated state.
The FTIR spectrometer used in this study was a Nicolet Magna 760 equipped with a Globar and a KBr beam splitter.7, 8 The measurements were made at a resolution of 4 cm−1 and with 50 scans.
The film sample of nylon-6 was immersed in water at room temperature for about 30–60 min for sufficient hydration9 just before the measurements. It was provided for the spectral measurements immediately after the removal of the surface water. The FTIR spectra of the sample were continuously measured in one experiment through the whole process to the most possible dehydration under a purge of dry air. The spectral changes in nylon-6 that were caused by the interactions with water were separated by the subtraction of the spectrum at the most possible dehydration from those taken during the process of dehydration, the subtraction factor being 1.0 for all cases.
RESULTS AND DISCUSSION
Absorptions of the Sorbed Water and the NH Group in the 3600–3000-cm−1 Region
Figure 1 shows the initial and final spectra of a series of spectra for a 5-μm-thick sample of nylon-6 taken for the entirety of the dehydration process for 72 min. The two spectra are very similar on the whole, except that there are some differences in the 3600–3300- and 3250–3000-cm−1 regions. The sharp and very strong band at 3299 cm−1 is assigned to the NH stretching of the hydrogen-bonded NH in the crystalline portion and in the ordered amorphous portion, whereas the 3066-cm−1 band is assigned to the overtone of the amide II band.10 The swelling at 3466 cm−1 for spectrum a, which is not observed at all for spectrum b, is assigned to the OH stretching absorption of the sorbed water. The two bands at 2934 and 2868 cm−1 are assigned to the CH stretching of the ethylene sequence.
The spectral changes in nylon-6 caused by the interactions with the sorbed water, which are not clear in Figure 1, are distinctively revealed in the difference spectra in Figure 2. The separated absorptions decrease in intensity with the progression of dehydration, and this indicates that these are caused by the interactions of nylon-6 with water.
The absorption with double peaks at 3466 and 3426 cm−1 in Figure 2, which is assigned to the water sorbed in nylon-6,6 has a hollow between them. Their relative intensities slightly change with the progress of dehydration. The two peaks are not so clear for spectrum d, and a difference spectrum similarly obtained for a 17-μm-thick sample of high dehydration is shown in Figure 3 for comparison. In this spectrum, the peak at 3474 cm−1 is a little stronger than the other just oppositely to spectrum a in Figure 2. The absorptions of the sorbed water in nylon-6 have significantly lower frequencies than those in many other polymers,6 and the spectrum is so distorted that the separation between the two peaks is only about 40 cm−1. This suggests that the sorbed water should strongly interact with nylon-6. An absorption due to the OH deformation of water, in addition, is expected to appear around 1650 cm−1.11 For nylon-6, however, a very strong absorption of amide I overlaps the expected position, and it is not practically possible to separate the absorption due to water.
According to Figure 2, the absorptions at 3248 and 3093 cm−1, together with a shoulder at 3187 cm−1, decrease in intensity with the progress of dehydration, like the absorptions due to the water. In contrast, the band at 3297 cm−1, which is the remainder of the absorption due to the NH stretching at 3299 cm−1 in Figure 1, shows a somewhat different dependence on the dehydration. It gradually decreases in intensity until spectrum c but almost disappears for spectrum d. This seems to indicate that the sorbed water has some influence on the intensity of the 3299-cm−1 band.
There appears to be a deep valley between 3426 and 3297 cm−1 in the series of difference spectra in Figure 2. This feature should be associated with the interactions of the sorbed water with the NH groups, as follows. As discussed later, the water should hydrate the free amide group in the amorphous portion. The hydration should reduce the population of the free NH, which is expected to appear at a frequency in the 3440–3300-cm−1 region.10, 12 When the spectrum of the dehydrated sample that has regenerated the free NH is subtracted from the spectra of the hydrated sample during dehydration, as in Figure 2, there should appear a minus peak in the 3440–3300-cm−1 region, as observed later for amide I and II bands. However, the skirt of the absorption of the sorbed water that extends to 3300 cm−1 overlaps the frequency region to deform the expected minus peak to the deep valley, as observed. On the other hand, the clear band at 3248 cm−1 in Figure 2 should be assigned to the NH stretching of the hydrated NH group. The frequency is lower by 51 cm−1 than the sharp absorption at 3299 cm−1. The observation of the amide II absorption of the hydrated NH group, which we discuss later, is in complete agreement with this interpretation.
There appears an absorption at 3093 cm−1 in the difference spectra in Figure 2, which decreases in intensity with the progress of dehydration, like the absorption at 3248 cm−1. We discuss the band later in relation to the effect of the hydration on the amide II band.
Here it should be noted that hydration causes some spectral changes in the absorptions of the CH2 groups around 2934 and 2868 cm−1 according to the difference spectra shown in Figure 2. This seems to indicate that the hydration of the amide groups causes some changes also in the structure of the ethylene sequences in the neighborhood of the hydrated portion.
Amide I and Amide II Bands
Figure 4 shows the 1800–1000-cm−1 region of the same two infrared spectra of nylon-6 shown in Figure 1. The very strong absorption at 1641 cm−1 is assigned to the amide I band, which has a main contribution of the CO stretching.10 Another strong band at 1542 cm−1 is assigned to the amide II band, to which the NH deformation mainly contributes.10 The weak absorption at 1265 cm−1 is associated with the amide group and is sometimes called an amide III band.10 The two spectra in Figure 4 are so similar that they cannot be distinguished from one another, some thickening of the spectral line being observed in the 1650–1500-cm−1 region. Even in this case, the spectral changes caused by the interactions of nylon-6 with the sorbed water are distinctively separated in the difference spectra, as discussed later.
Figure 5 shows the 1800–1000-cm−1 region of the same difference spectra shown in Figure 2. The spectral changes caused by the interactions are so clear that there appear minus and plus peaks around each of the amide I and II bands.13
The minus peak at 1675 cm−1 and the plus peak at 1628 cm−1 appear above and below the amide I band at 1641 cm−1, respectively. For spectrum a, for example, the minus peak has a depth of about 0.04 in absorbance, whereas the plus peak has a height of about 0.075. The magnitude of the two peaks decreases as the dehydration proceeds, and this verifies that they are caused by the interactions with the sorbed water.
The appearance of the minus and plus peaks may be explained in terms of the hydration of the free CO groups and the regeneration of the free groups as follows. The water sorbed in the nylon-6 sample should partly hydrate the free CO group to form the CO::H2O complex, which results in the decreased population of the free CO group. Dehydration, on the contrary, should release the water from the hydrate to regenerate the free CO group. If the spectrum of the most dehydrated sample is subtracted from those of the hydrated sample taken during dehydration, as in Figure 5, a minus peak should appear at the frequency of the absorption of the free CO group, and there should appear a plus peak at the frequency of the absorption of the hydrated CO group. As the dehydration proceeds, the magnitude of the minus and plus peaks should become smaller. The free CO group of the amide should have an absorption in the 1700–1650-cm−1 region.10, 12 The frequency of 1675 cm−1 for the minus peak agrees well with the expected position of the free CO. On the other hand, the plus peak at 1628 cm−1 should be assigned to the hydrated CO group. The frequency of this band is lower by 47 cm−1 than that of the free CO group. The observed magnitude of the shift is rather large for the hydrogen bonding.14 The absorption of the hydrated CO group is even lower by 13 cm−1 than the one at 1641 cm−1 of the hydrogen-bonded CO as CO::HN in the crystalline portion in nylon-6.
The appearance of the plus and minus peaks for the amide II band at 1542 cm−1 can similarly be explained. In this case, however, their position with respect to the amide II band is just opposite to the case of the amide I band because a deformation vibration should shift to a higher frequency by hydrogen bonding oppositely to the case of the stretching vibration.15 Consequently, a plus peak appears at 1566 cm−1 and a minus peak appears at 1529 cm−1 in a series of difference spectra shown in Figure 5. The frequency of the minus peak at 1529 cm−1 is within the 1550–1510-cm−1 region of the free NH group.10 The plus peak at 1566 cm−1 should be assigned to the hydrated NH group, the NH stretching counterpart of which appears at 3248 cm−1, as discussed previously. For the amide II band, the upshift by hydration is 38 cm−1, whereas the shift caused by the hydrogen bonding with a CO group is 12 cm−1 from 1529 to 1541 cm−1. The magnitude of the shift is significantly larger for the hydration.
We have already assigned the absorptions associated with the hydrated CO and NH groups according to the difference spectra. The absorptions in the 3200–3000-cm−1 region are related to an overtone of the amide II and a combination of the amide I and amide II for the hydrated amide. In Figure 2, there appears a clear absorption at 3093 cm−1, which is higher by 27 cm−1 than the 3066-cm−1 band, this being the overtone of the amide II band at 1542 cm−1.10 The 3093-cm−1 band is assigned to the overtone of the amide II at 1566 cm−1 for the hydrated amide. The upshift by hydration is reasonable for the assignment.
There appears a shoulder around 3187 cm−1 in the difference spectrum in Figure 2, whereas the spectrum in Figure 1 also shows a shoulder at 3194 cm−1. This may be assigned to the combination of the amide I band and the amide II band, the added frequency (3183 cm−1) agreeing with the observed frequency (3194 cm−1). The shoulder at 3187 cm−1 in the difference spectrum (Fig. 2) is similarly assigned to the combination between the amide I and amide II bands of the hydrated amide group. The frequency of the shoulder is near the one at 3194 cm−1 in Figure 1. This is reasonable because the shift by hydration is almost canceled for the combination.
Interactions of Amide Groups with Water
We have already discussed the effect of the hydration on the NH stretching and the amide I and II bands on the basis of the difference spectra. All the changes observed for the relevant absorptions have consistently been interpreted as being due to water penetrating the amorphous portion of a nylon-6 sample to hydrate the free amide groups.
The effect of the hydration on infrared absorptions of NH and CO groups is summarized in Table 1. The magnitude of the shift of the NH stretching absorption is larger by 51 cm−1 for the NH::OH2 hydration than the NH::OC hydrogen bonding, whereas that of the amide II band is similarly larger by 24 cm−1. However, the magnitude of the shift of the amide I band caused by the CO::H2O hydration is larger by 13 cm−1 than that by the CO::HN hydrogen bonding. Therefore, the hydration evidently has a larger shifting effect on the frequencies of the NH and CO absorptions than the NH::OC hydrogen bonding. This means that the hydrogen bonding between the sorbed water and the amide groups is significantly stronger than that between NH and CO of the amide groups in the crystal.
Table 1. Effect of the Hydration on the Absorptions of CO and NH Groups
3400–3300 (hollow, overlapping with the absorption of the sorbed water)
υ(CO) (amide I)
1675 (minus peak)
1628 (plus peak)
δ(NH) (amide II)
1566 (plus peak)
1529 (minus peak)
A possible hydration structure is
where W is a water molecule and the CO and NH groups should be hydrated as CO::H2O and NH::OH2, respectively. The structure has two kinds of nonequivalent water molecules, the one being hydrated to CO and the other to NH. The two kinds of water molecules may have different OH stretching frequencies. In the spectra in Figures 2 and 3, the absorptions of water have two close peaks. The absorptions of water can be associated with the different coordinating types. From the information available at present, however, it is not possible to decide whether the two peaks should be assigned to the two hydrogen-bonding types or the symmetric and antisymmetric OH stretching vibrations of water.
The hydration structure of water given as I is of the first hydration sphere that directly interacts with the amide groups. To get some more information on the water in the hydration sphere outside the first hydration sphere, we measured the FTIR spectra of a hydrated 17-μm-thick sample, which could hold more water. The separated spectra of the water by dehydration are shown in Figure 6. According to the spectra, the sorbed water at first shows only one peak at 3419 cm−1, with an obscure shoulder at the slope. As the dehydration proceeds, the shoulder becomes a little clearer at a frequency above 3465 cm−1, and the one at 3419 cm−1 gradually shifts up. With further dehydration, the shoulder becomes a peak, as in spectrum c. Finally, in spectrum d, the sorbed water shows the two peaks at 3474 and 3433 cm−1, the separation being only 41 cm−1. This spectrum is reproduced as an ordinate-expanded one in Figure 3, in which the peak at 3474 cm−1 is a little stronger than the other at 3433 cm−1. This should be the spectrum of the water that directly interacts with the amide group in the first hydration sphere.
The spectral change from a to d in Figure 6 seems to indicate that the water in the second hydration shell should have a spectrum in which the component of the lower frequency is stronger than the other and the frequency shifts down, becoming somewhat similar to the spectrum of liquid water,16 as the water layer is further away from the first hydration sphere. This may mean that in the second hydration sphere the interactions between water molecules should play a more important role in determining the structure.
For the purpose of studying the interactions of nylon-6 with water, we have followed the changes in the infrared spectra of a hydrated nylon-6 sample during dehydration. The sample should be as thin as 5 μm to give sufficiently good infrared spectra for investigation because the absorptions due to the amide group, which is the interacting counterpart of water, are so strong. The spectral change is hardly observable even between the initial and final spectra taken during dehydration (Figs. 1 and 4). The application of spectral subtraction has clearly revealed the changes caused by the interactions with respect to the NH stretching and the amide I and II bands. The changes have reasonably been interpreted as the results of the hydrogen bonding of water with free CO and NH groups in the amorphous portion. Therefore, we have definitely identified the interacting counterpart of water in polymers. The importance of understanding the interactions of water in polymers on a molecular basis has long been recognized.17 The found interaction is at the first hydration sphere around an amide group. To understand the interactions of water in polyamides more extensively, we should study how the interactions in the first hydration sphere are transferred to the second hydration sphere. The infrared spectra of a thicker sample (Fig. 6) suggests that the structure of the water over the first hydration sphere should gradually change with increasing hydration, although the corresponding changes in the amide I and II bands could not be observed because it was too thick.
The study of the interactions of an amide group with water is important, not only for understanding the interactions of nylon-6 with water but also because an amide linkage is the skeletal structure of proteins. The interactions of water with proteins are of great importance for life.18 Proteins consist of various hydrophilic groups, and the interactions with water should not be so simple as they are for nylon-6. The investigation of the interaction of an amide group with water is still the first step to understanding the interactions of water with proteins. In fact, the changes in the infrared spectra of a hydrated collagen sample are similar to those of nylon-6 in many important respects.19
R. Iwamoto thanks the member companies (A-ICS Corp.; Kuraray Co., Ltd.; Thermo Nicolet Japan, Inc.; JT Engineering, Inc.; Shimadzu Corp.; Nippon Gosei Ind. Co., Ltd.; Toagosei Co., Ltd.; Toyobo Research Center Co., Ltd.; and Yokogawa Electric Corp.) of the Near-Infrared Spectroscopy (NIRS) project for their sponsorship of this study.