Big Rip in SO(1,1) phantom universe

Yi-Huan Wei
DOI: https://doi.org/10.48550/arXiv.gr-qc/0502077
2005-02-17
Abstract:For the inverse linear potential, the SO(1,1) field behaves as phantom for late time and the Big Rip will occur. The field approaches zero as time approaches the Big Rip, here. For this potential the phantom equation of state takes the late-time minimum $w_\Phi=-3$. We give some discussions that the Big Rip in the SO(1,1) model may be treated as either the transition point of universe from expansion to extract phase or the final state. In the latter picture of the universe, the field has the $T$ symmetry and the scale factor possesses the $CT$ symmetry, for which the SO(1,1) charge $\bar{Q}$ plays a crucial role.
General Relativity and Quantum Cosmology
What problem does this paper attempt to address?
The main problem that this paper attempts to solve is to explore the influence of container materials on the Raman scattering signals and photoluminescence phenomena of water samples and their water - ethanol mixtures. Specifically, the author hopes to verify the following two objectives through systematic experiments: 1. **Prove the photoluminescence phenomenon caused by contaminants introduced by the container**: Research shows that different types of containers (such as fused silica, polypropylene, and soda - lime glass) will introduce trace contaminants into water samples. Under specific conditions, these contaminants will produce significant photoluminescence effects, thereby affecting the intensity of Raman scattering signals. 2. **Question the "intrinsic luminescence" explanation of water proposed by Lobyshev et al.**: By comparing the experimental results, the author hopes to re - evaluate the previous hypothesis that water itself has intrinsic luminescence characteristics and provide new insights. ### Specific problems and methods To achieve the above - mentioned goals, the researchers carried out the following tasks: - **Select samples**: Use pure water and 50% water - 50% ethanol mixed solutions as experimental samples. - **Select containers**: Store the samples in containers made of three different materials respectively: fused silica (Quartz glass), polypropylene (Polypropylene), and soda - lime glass (Soda - lime glass type III). - **Experimental methods**: - **Raman scattering experiment**: Use an argon - ion laser for excitation (with a wavelength of 514.5 nm), and record the Stokes and anti - Stokes bands to distinguish Raman lines and photoluminescence. - **Photoluminescence experiment**: Use different excitation wavelengths (such as 488 nm) to measure the photoluminescence response of the samples. - **Conductivity measurement**: Measure the conductivity of the samples to evaluate the presence and concentration of contaminants. ### Main findings - **Pure water samples**: - The water sample in the soda - lime glass container shows a significant photoluminescence effect, especially near 2110 cm⁻¹, with an intensity 98% higher than that of the water sample in the fused silica container. - The photoluminescence phenomenon depends on the excitation wavelength and is mainly concentrated in the internal and intermolecular vibration mode regions of water molecules. - **Water - ethanol mixed samples**: - The water - ethanol mixed sample shows a stronger photoluminescence effect in the soda - lime glass container, especially near 1360 cm⁻¹. - Reducing the laser power can completely eliminate this photoluminescence effect, while the Raman mode remains unchanged. ### Conclusions By combining Raman scattering, photoluminescence, and conductivity experiments, the researchers conclude that the optical properties of high - purity water samples may be significantly interfered with by trace contaminants from the container. In particular, soda - lime glass containers are the least suitable for storing high - purity water samples because they are prone to release contaminants and cause photoluminescence phenomena. In contrast, fused silica containers are the best choice because they can minimize the interaction between the contents and the container. In addition, the researchers also point out that at the current experimental stage, due to the lack of further characterization means (such as ICP - MS), it is still not possible to finally determine whether water has intrinsic luminescence characteristics. However, the existing results show that it is very difficult to obtain completely uncontaminated water samples, and even trace contaminants will produce a weak photoluminescence effect.