Highly collimated, plasma-filled magnetic flux tubes are frequently observed on galactic, stellar and laboratory scales. We propose that a single, universal magnetohydrodynamic pumping process explains why such collimated, plasma-filled magnetic flux tubes are ubiquitous. Experimental evidence from carefully diagnosed laboratory simulations of astrophysical jets confirms this assertion and is reported here. The magnetohydrodynamic process pumps plasma into a magnetic flux tube and the stagnation of the resulting flow causes this flux tube to become collimated.The extreme collimation of astrophysical jets [1,2,3] and the solar corona heating mechanism [4] are two seemingly unrelated astrophysical mysteries, yet both involve collimation of magnetic flux tubes. Astrophysical observations [2,3] and simulations [1,5] indicate that bipolar plasma outflows (jets) are natural [1,6] features of young stellar objects, black holes, active galactic nuclei and even aspherical planetary nebula [7]. Although it has long been presumed [8,9] that astrophysical jets are magnetohydrodynamically driven, the standard models do not agree on a single collimation process. A similar issue exists in solar physics: solar spicules [10], prominences [11,12] and coronal loops [13] are considered to be plasma-filled filamentary magnetic flux tubes; coronal heating models [14,15] then invoke magnetic reconnection and plasma flow within such filamentary loops. However, the models explain neither the origin of the observed flows nor the extreme collimation (filamentary nature) of the observed structures.We propose that the collimation of any, initially flared, current-carrying magnetic flux tube is due to the following process [16]: a magnetohydrodynamic (MHD) force resulting from the flared current profile drives axial plasma flows along the flux tube; the flows convect frozen-in magnetic flux from strong magnetic field regions to weak magnetic field regions; flow stagnation then piles up this embedded magnetic flux, increasing the local magnetic field and collimating the flux tube via the pinch effect. Thus, the flux tube fills with ingested plasma and simultaneously becomes collimated. This paper presents direct experimental evidence for this process. We use ultra-high-speed imaging and Doppler measurements of the fast plasma flows, combined with direct density measurements before and after the filling of the flux tube.Our experimental setup [17] simulates magneticallydriven astrophysical jets at the laboratory scale by imposing boundary conditions analogous to astrophysical jet boundary conditions (Fig. 1): a disk (cathode) representing a central object such as a star, is coaxial and co-planar with an annulus (anode) representing an accretion disk. A vacuum poloidal magnetic field produced by an external coil links these two electrodes, mimicking a poloidal magnetic field threading the accretion disk. A radial electric field applied across the gap between the FIG. 1: (log color) Typical plasma discharge sequence (#6577, 2 million fps, 40 ns/...

The evolution of relative canonical helicity is examined in the two-fluid magnetohydrodynamic formalism. Canonical helicity is defined here as the helicity of the plasma species' canonical momentum. The species' canonical helicity are coupled together and can be converted from one into the other while the total gauge-invariant relative canonical helicity remains globally invariant. The conversion is driven by enthalpy differences at a surface common to ion and electron canonical flux tubes. The model provides an explanation for why the threshold for bifurcation in counter-helicity merging depends on the size parameter. The size parameter determines whether magnetic helicity annihilation channels enthalpy into the magnetic flux tube or into the vorticity flow tube components of the canonical flux tube. The transport of relative canonical helicity constrains the interaction between plasma flows and magnetic fields, and provides a more general framework for driving flows and currents from enthalpy or inductive boundary conditions. I. INTRODUCTIONPrevious treatments of canonical helicity-also known as generalized vorticity 1 , self-helicity 2 , generalized helicity 3 , or fluid helicity 4 -concluded that the canonical helicities of each species were invariant, independent from each other. Assuming closed canonical circulation flux tubes inside singly-connected volumes, and arguing for selective decay arguments in the presence of dissipation, it was shown that canonical helicity is a constant of the system stronger than magnetofluid energy. Generalized relaxation theories could therefore minimize magnetofluid energy for a given canonical helicity and derive stationary relaxed states.It is puzzling that in a multiple-component plasma a species' canonical helicity must be independent from another. Given a general isolated system, the canonical momentum vectors of ion and electron fluid elements trace out different closed helical paths. Both paths are linked, resembling intertwined helical braids directed along a magnetic field line, and define flux tubes of canonical momentum that interpenetrate each other. On scales larger than the ion and electron skin depths, defined as ⁄ where is the speed of light and is the plasma frequency of the species , or when species momentum is negligible, canonical flux tubes are topologically indistinguishable from magnetic flux tubes, so magnetic helicity suffices to describe the quasistatic evolution of the system. But when species momentum is significant or when phenomena involves scales that include ion or electron skin depths, then canonical flux tubes are distinguishable from one another, and because they overlap, any effort to count helicity in the system should consider gauge dependence.

Pulsed-power technology and appropriate boundary conditions have been used to create simulations of magnetically driven astrophysical jets in a laboratory experiment. The experiments are quite reproducible and involve a distinct sequence. Eight initial flux tubes, corresponding to eight gas injection locations, merge to form the jet, which lengthens, collimates, and eventually kinks. A model developed to explain the collimation process predicts that collimation is intimately related to convection and pile-up of frozen-in toroidal flux convected with the jet. The pile-up occurs when there is an axial non-uniformity in the jet velocity so that in the frame of the jet there appears to be a converging flow of plasma carrying frozen-in toroidal magnetic flux. The pile-up of convected flux at this "stagnation region" amplifies the toroidal magnetic field and increases the pinch force, thereby collimating the jet.

A novel 2D ion temperature measurement for toroidal plasmas has been developed by use of cost-effective discrete tomography reconstruction of 2D ion Doppler spectroscopy composed of a polychromator with an ICCD camera and optical fibres. The 2D projection of the line spectrum collected by 35 (7 × 5) optical fibres is transformed into the r-z profile of the local spectrum by means of the Abel inversion at each wavelength and finally into the 2D (r-z) profile of the ion temperature. Numerical tests of its algorithm indicate that the reconstruction error for a peaked temperature profile is smaller than 15% if the chord-integrated signals have noise smaller than 5%. This system successfully measured the peaked ion temperature profile of a torus plasma on the r-z plane under the condition of negligibly small plasma flow.