We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Polarized electron beam production via laser wakefield acceleration in pre-polarized plasma is investigated by particle-in-cell simulations. The evolution of the electron beam polarization is studied based on the Thomas–Bargmann–Michel–Telegdi equation for the transverse and longitudinal self-injection, and the depolarization process is found to be influenced by the injection schemes. In the case of transverse self-injection, as found typically in the bubble regime, the spin precession of the accelerated electrons is mainly influenced by the wakefield. However, in the case of longitudinal injection in the quasi-1D regime (for example, F. Y. Li et al., Phys. Rev. Lett. 110, 135002 (2013)), the direction of electron spin oscillates in the laser field. Since the electrons move around the laser axis, the net influence of the laser field is nearly zero and the contribution of the wakefield can be ignored. Finally, an ultra-short electron beam with polarization of 
$99\%$ can be obtained using longitudinal self-injection.
Competition among the two-plasmon decay (TPD) of backscattered light of stimulated Raman scattering (SRS), filamentation of the electron-plasma wave (EPW) and forward side SRS is investigated by two-dimensional particle-in-cell simulations. Our previous work [K. Q. Pan et al., Nucl. Fusion 58, 096035 (2018)] showed that in a plasma with the density near 1/10 of the critical density, the backscattered light would excite the TPD, which results in suppression of the backward SRS. However, this work further shows that when the laser intensity is so high (
$>{10}^{16}$ W/cm2) that the backward SRS cannot be totally suppressed, filamentation of the EPW and forward side SRS will be excited. Then the TPD of the backscattered light only occurs in the early stage and is suppressed in the latter stage. Electron distribution functions further show that trapped-particle-modulation instability should be responsible for filamentation of the EPW. This research can promote the understanding of hot-electron generation and SRS saturation in inertial confinement fusion experiments.
There is a growing consensus on brain networks that it is not immutable but rather a dynamic complex system for adapting environment. The neuroimaging research studying how brain regions work collaboratively with dynamic methods had demonstrated its effectiveness in revealing the neural mechanisms of schizophrenia.
To investigate altered dynamic brain functional topology in first-episode untreated schizophrenia patients (SZs) and establish classification models to find objective brain imaging biomarkers.
Resting-state-functional magnetic resonance data for SZs and matched healthy controls were obtained(Table1).
Power-264-template was used to extract nodes and sliding-window approach was carried out to establish functional connectivity matrices. Functional topology was assessed by eigenvector centrality(EC) and node efficiency and its time-fluctuating was evaluated with coefficient of variation(CV). Group differences of dynamic topology and correlation analysis between Positive and Negative Syndrome Scale(PANSS) scores and topology indices showing group differences, which also were used in establishing classification models, was examed.
The CV of node efficiency in angular and paracingulate gyrus was larger in SZs. There are 13 nodes assigned into several brain networks displaying altered CV of EC between groups(Figure1.A). Fluctuation of EC of the node in DMN, which was lower in SZs, showed negative correlation with PANSS total scores(Figure1.B). Dynamic functional topology of above nodes was used to train classification models and demonstrated 80% and 71% accuracy for support vector classification(SVC) and random forest(RF), respectively(Figure2).
Dynamic functional topology illustrated a capability in identifying SZs. Aberrated dynamics of DMN relevant to severity of patient’s symptoms could reveal the reason why it contributed to classification.
No significant relationships.
