5.4. Fast relaxation component
The time constant of the fast component is a few picoseconds in Fig. 6(b) and is not observed in liquid water [32,33]. Therefore, the fast component in the present study is considered to be due to the motion of the interfacial water at the surfactant. The motion of water was reported to slow down to around 100 MHz owing to hydration [43]. The water motion on such a timescale cannot be measured in the present study, and hence, the mechanism of the fast component is considered to be different from the slowing down of the bulk water motion of ~8 ps. We believe that there exists a local vibration-like motion of water molecules subject to the surfactant. Studies by ultrafast spectroscopy of the OH stretching vibration of water in AOT reverse micelles have demonstrated that the hydrogen-bond network of water is strongly disrupted in the hydration layer at the interface [7–10]. Accordingly, the hydrogen-bond network of water at the interface will be less extended than that in the core of the reverse micelle. Moreover, the water motion at the interface is restricted by the interaction with the surfactant. Under these conditions, the water at the interface will exhibit a localized vibration-like motion.
The time constant of the fast component increases with decreasing temperature above Tc, as shown in Fig. 6(b). The slowing down of the slow component in the core could affect the motion of the interfacial water in the restricted space, as indicated in literature [7,8]. On the other hand, the time constants below Tc are smaller than those above Tc and independent of temperature. It is reasonable to attribute this result to the size reduction of the reverse micelle below Tc. However, the study of reverse micelles of smaller sizes is necessary to discuss this result in detail and will be reported on elsewhere.
As described in Section 5.1, the water expelled from the reverse micelle is dispersed in the nonpolar solvent below Tc. This leads to the fact that another interfacial water layer at the solvent appears below Tc, in addition to that at the surfactant. Moreover, the surface area at the solvent is comparable to that at the surfactant because most of the bulk-like water molecules are extracted from the reverse micelle. However, the amplitude of the fast component does not increase below Tc compared with that above Tc, as shown in Fig. 6(c). Studies of water at hydrophobic liquid surfaces suggest that the hydrogen-bond network of water at the surface is weakly formed and that the onset of bulk-phase behavior occurs very close to the interface [45–47]. Hence, it is likely that the interfacial water at the solvent mostly serves as bulk-like water and does not show the local motion that is observed in the interfacial water at the surfactant. As a result, the interfacial water at the solvent does not contribute to the fast relaxation component below Tc.
6. Conclusion
We have made temperature-dependent measurements of THz-TDS and DLS for an AOT reverse micelle with a Stokes radius of ~12 nm. Two relaxation processes of water with ~12 ps and ~3 ps at room temperature have been found; the slow component is due to the collective motion of bulk-like water, whereas the fast one is due to the local motion of the interfacial water at the surfactant. The slowing down of the slow component with decreasing temperature is more pronounced than that in liquid water above Tc (=286 K).