Earlier a nondynamo theory in teleparallel gravity was developed by Bamba et al-JCAP 2010. Also earlier I have been obtained a dynamo equation generalised to spacetime with torsion - PLB 2012. In this paper we obtained from this equation $10^{-11}G$ instead $10^{-9}G$ obtained by Bamba et al.

Recently Adelberger et al [Phys Rev Lett 98: 010402, (2007)] have placed a limit to photon mass by investigating the primordial magnetic fields. Earlier Bertolami et al [Phys Lett \textbf{B} 455, 96(1999)] showed that massive photons in a spontaneous Lorentz breaking may generate primordial magnetic fields consistent with galactic dynamo seeds. Torsion coupling constant of order $10^{-5}$, much higher than the previously obtained by de Sabbata and Sivaram of $10^{-24}$, leads to strong amplification of magnetic field able to seed galactic dynamo at recombination era contrary to what happens in general relativistic dynamos. This results in $B\sim{10^{-5}{\beta}G}$ where ${\beta}$ is the massive photon-torsion coupling. Thus in order to obtain the observed galaxy field of $B_{G}\sim{{\mu}G}$ one should have a coupling $\beta\sim{10^{-1}}$, never observed in the universe. Thus we may conclude that the weaker couplings for torsion to e.m fields shall only produce magnetic fields without dynamos starting from extremely strong magnetic seeds. From the strongest photon-torsion considered one obtains the best CMB estimate for torsion generated magnetic fields $\frac{{\delta}B}{B}\le{10^{-4}}$. By making use of the strong photon-torsion limits obtained here, photon mass limit of $m_{\gamma}\sim{10^{-24}eV}$, well-within limits found in literature, which allows us to conclude that a stronger massive photon-torsion limit is physically consistent. Actually this last limit is also graviton mass limit. This results differs from Adelberger et al by two orders of magnitude.

Recently Salim et al [JCAP (2007)], have shown that galactic dynamo seeds can be possibly attainable in bouncing cosmological models with QED Lagrangeans. In this paper we generalise their result by include torsion of spacetime in bouncing cosmology. It is shown that by considering a semi-minimal photon-torsion coupling and a Lagrangean of the type $RF^{2}$ it is possible to find a fast decoupling between magnetic and torsion fields in the contracting phases of the universe. Besides torsion field decays as $K\sim{a^{2/3}}$ while the magnetic field grows as $B\sim{a^{-5.5}}$ thus explaining the fast decoupling between the two fields. It is expected that at some point of the contracting phase the amplification of the magnetic field may give rise to a enough strong magnetic field to seed a galactic dynamo.

Dynamo action is shown to be induced from homogeneous non-minimal photon-torsion axial coupling in the quantum electrodynamics (QED) framework in Riemann flat spacetime contortion decays. The geometrical optics in Riemann-Cartan spacetime is considering and a plane wave expansion of the electromagnetic vector potential is considered leading to a set of the equations for the ray congruence. Since we are interested mainly on the torsion effects in this first report we just consider the Riemann-flat case composed of the Minkowskian spacetime with torsion. It is also shown that in torsionic de Sitter background the vacuum polarisation does alter the propagation of individual photons, an effect which is absent in Riemannian spaces. It is shown that the cosmological torsion background inhomogeneities induce Lorentz violation and massive photon modes in this QED. Magnetic dynamos in this torsioned spacetime electrodynamics are simpler obtained in Fourier space than the cosmic ones, previously obtained by Bassett et al Phys Rev D, in Friedmann universe. By deriving plasma dispersion for linear electrodynamics in Riemann Cartan spacetime, dynamo action seems to be possible for plasma frequencies in some polarizations. The important cosmic magnetic field problem of breaking conformal flatness is naturally solved here since the photon torsion coupling breaks conformal flatness.

Chicone et al [Comm Math Phys (1997)] investigated existence of fast dynamos by analyzing the spectrum kinematic magnetic dynamo. In real non-degenerate branch of the spectrum, the kinematic dynamo operator lies on a compact Riemannian 2D space of constant negative curvature. Here, generalization of Marklund and Clarkson [MNRAS (2005)], general relativistic GR-MHD dynamo equation to include mean-field dynamos is obtained. In the absence of kinetic helicity, adiabatic constant $\gamma={1/2}$ and gravitational colapse of negative Riemann curvature of spatial sections enhance dynamo effect $\frac{{\delta}B}{B}=2.6\times 10^{-1}$. Critical time where linear dynamo effects breaks down de to curvature. At recombination time, COBE temperature anisotropies, implies that magnetic field growth rate is ${\lambda}{\approx{10}^{-9}yr^{-1}}$. This places a bound on curvature till the recombination magnetic field is amplified to present value of $B_{0}=10^{-9}G$, by dynamo action. At present epoch, negative curvature becomes constant and the Chicone et al result is shown to be valid in cosmology. Since negative curvature is non-constant, Hilbert theorem which forbiddes negative constant curvature surfaces embeddeding in $\textbf{R}^{3}$ is bypassed.

Magneto-curvature stresses could deform magnetic field lines and this would give rise to back reaction and restoring magnetic stresses [Tsagas, PRL (2001)]. Barrow et al [PRD (2008)] have shown in Friedman universe the expansion to be slow down in spatial section of negative Riemann curvatures. From Chicone et al [CMP (1997)] paper, proved that fast dynamos in compact 2D manifold implies negatively constant Riemannian curvature, here one applies the Barrow-Tsagas ideas to cosmic dynamos. Fast dynamo covariant stretching of Riemann slices of cosmic Lobachevsky plane is given. Inclusion of advection term on dynamo equations [Clarkson et al, MNRAS (2005)] is considered. In absence of advection a fast dynamo is also obtained. Viscous and restoring forces on stretching particles decrease, as magnetic rates increase. From COBE data ($\frac{{\delta}B}{B}\approx{10^{-5}}$), one computes stretching $\frac{{\delta}V^{y}}{V^{y}}=1.5\frac{{\delta}B}{B}\approx{1.5{\times}10^{-5}}$. Zeldovich et al has computed the maximum magnetic growth rate as ${\gamma}_{max}\approx{8.0{\times}10^{-1}t^{-1}}$. From COBE data one computes a lower growth rate for the magnetic field as ${\gamma}_{COBE}\approx{6.0{\times}10^{-6}t^{-1}}$, well-within Zeldovich et al estimate. Instead of the Harrison value $B\approx{t^{{4/3}}}$ one obtains the lower primordial field $B\approx{10^{-6}t}$ which yields the $B\approx{10^{-6}G}$ at the $1s$ Big Bang time.

A plasma loop twisted Riemannian model is applied to torus dynamos twisted flows it leading to a slow dynamo such as in Moebius strip dynamo, recently considered by Shukurov, Stepanov and Sokoloff [Phys. Rev. \textbf{E 78},025301,(2008)] to modelling Perm dynamo torus in liquid sodium. Since diffusion and advection (stretching), are competing effects for dynamo action, plasma resistivity term is shown to be proportional to loops Riemann curvature (folding). Shukurov et al, also showed that based on Ponomarenko dynamo, a broader torus channel produces a better dynamo. These results agree with Schekochihin et al [Phys Rev \textbf{E} (2002)] where random filamentary magnetic fields are strengthen by curvature. Analysis of spectrum of chaotic fast dynamos, shows that Riemann curvature acts as a damping, since growth magnetic field rate is inversely proportional to Riemann curvature. Comparison with general relativistic MHD dynamo equation, shows that the Ricci tensor, which is a contraction of the Riemann tensor also appears in the diffusion term. Curvature of plasma loop is ${{R^{1}}_{212}}|_{\textbf{Plasma}}\approx{5.6{\times}10^{-19}m^{-2}}$, while for Perm torus is certainly higher. Thus slow dynamos are favoured in dynamo laboratories rather than in plasma loops. It is shown that the curvature-stretching flux rope dynamo coupling energy, coincides with the minimum twist energy ${\epsilon}_{\textbf{twist}}\approx{10^{30}TeV}$ stored in flux ropes. Torus flux tubes around black-holes remain in the order of $2MeV$ and GBR are around $10^{52}TeV$. Since the ${R^{1}}_{212}$ is negative, inflexionary flux tubes fast dynamos may be responsible for this CME mechanism in UHF plasma loops.

Previously Liao and Shuryak [\textbf{Phys. Rev C (2008)}] have investigated electrical flux tubes in monopole plasmas, where magnetic fields are non-solenoidal in quark-QCD plasmas. In this paper slow dynamos in diffusive plasma [{\textbf{Phys. Plasmas \textbf{15} (2008)}}] filaments (thin tubes) are obtained in the case of monopole plasmas. In the absence of diffusion the magnetic field decays in the Early Universe. The torsion is highly chaotic in dissipative large scale dynamos in the presence of magnetic monopoles. The magnetic field is given by the Heaviside step function in order to represent the non-uniform stretching of the dynamo filament. These results are obtained outside the junction condition. Stringent limits to the monopole flux were found by Lewis et al [\textbf{Phys Rev D (2000)}] by using the dispute between the dynamo action and monopole flux. Since magnetic monopoles flow dispute the dynamo action, it seems reasonable that their presence leads to a slow dynamo action in the best hypothesis or a decay of the magnetic field. Hindmarsh et al have computed the magnetic energy decay in the early universe as ${E}_{M}\approx{t^{-0.5}}$, while in our slow dynamo case linearization of the growth rate leads to a variation od magnetic energy of ${\delta}{E}_{M}\approx{t}$, due to the presence of magnetic monopoles. Da Rios equations of vortex filaments are used to place constraints on the geometry of monopole plasma filaments.

A new antidynamo theorem for non-stretched twisted magnetic flux tube dynamo is obtained. Though Riemannian curvature cannot be neglected since one considers curved magnetic flux tube axis, the stretch can be neglect since one only considers the limit of thin magnetic flux tubes. The theorem states that marginal or slow dynamos along curved (folded), torsioned (twisted) and non-stretched flux tubes endowed with diffusionless plasma flows, if a constraint is imposed on the relation between poloidal and toroidal magnetic fields in the helical dynamo case. A formula for the stretch of flux tubes is derived. From this formula one shows that the Riemann flux tube is stretched by an interaction between the plasma flow vorticity and torsion, in accordance with our physical intuition. Marginal diffusionless dynamos are shown to exist obtained in the case of flux tube dynamos exponential stretching. Thus slow dynamos can be obtaining on the flux tube under stretching. Filamentary dynamos anti-dynamos are also considered. As flux tubes possess a magnetic axis torsioned filament, it can also be considered as thegerm of a fast dynamo in flux tubes Riemannian curved space. It is shown that for non-stretched filaments only untwist and unfold filaments can provide dynamo action in diffusive case. A condition for exponential stretching and fast dynamos in filaments is given. These results are actually in agreement with Vishik argument that fast dynamo cannot be obtained in non-stretched flows. Actually the flux tube result is the converse of Vishik's lemma.

Stretching solar dynamos from differential rotation in a Riemannian manifold setting is presented. The spherical model follows closely a twisted magnetic flux tube Riemannian geometrical model or flux rope in solar physics, presented previously by Ricca [Solar Physics (1997)]. The spherical model presented here present new and interesting feature concerning its connection with spherical steady solar dynamos. One of this new feature is represented by the fact that the by considering poloidal magnetic field component much weaker than its toroidal counterpart, one obtains a stretch dynamo action where the Riemannian solar spherical line element is proportional to differential rotation. This result is obtained also by using the Vainshtein-Zeldovich stretch, twist and fold (STF) method to generate dynamos. One notes that for high magnetic Reynolds of $R{m}=O({10^{7}})$ the dynamo action is present for a corresponding small stretching factor of $K^{2}=1.6$ where $K^{2}=1$ represents the unstretched dynamo. The constant stretched dynamos considered here are shown to be Riemann-flat, where the Riemann curvature tensor vanishes. Solar cycle dynamos are therefore compatible with the Riemann-flat stretching dynamo model from solar differential rotation presented here.

Riemannian and filamentary geometries are used to investigate the stretched filamentary dynamos against solar data in solar corona.

Compact Riemannian solar twisted magnetic flux tube surfaces model are tested against solar extreme ultraviolet (EUV) lines observations, allowing us to compute the diameter and height of solar plasma loops. The relation between magnetic and torsion energies is found for a nonplanar solar twisted (torsioned) loop to be $10^{9}$, which shows that the contribution of torsion energy to the solar loop is extremely weaker than the magnetic energy contribution. In this case solar loops of up $5000 km$ in diameter can be reached. The height of $220.000 km$ is used to obtain an estimate for torsion based on the Riemannian flux tube surface, which yields ${\tau}_{0}=0.9{\times} 10^{-8} m^{-1}$ which coincides with one of the data of $(0.9{\pm}0.4){\times}10^{-8}m^{-1}$ obtained by Lopez-Fuentes et al (2003). This result tells us that the Riemannian flux tube model for plasma solar loops is consistent with experimental results in solar physics. These results are obtained for a homogeneous twisted solar loop. By making use of Moffatt-Ricca theorem for the bounds on torsional energy of unknotted vortex filaments, applied to magnetic topology, one places bounds on the lengths of EUV solar loops. New results as the vorticity of the plasma flow along the tube is also computed in terms of the flux tube twist.

Analytical solution of first order torsion ${\alpha}$-effect in twisted magnetic flux tubes representing a flux tube dynamo in Riemannian space is presented. Toroidal and poloidal component of the magnetic field decays as $r^{-1}$, while grow exponentially in time. The rate of speed of the helical dynamo depends upon the value of Frenet curvature of the tube. The $\alpha$ factor possesses a fundamental contribution from constant torsion tube approximation. It is also assumed that the curvature of the magnetic axis of the tube is constant. Though ${\alpha}$-effect dynamo equations are rather more complex in Riemann flux tube coordinates, a simple solution assuming force-free magnetic fields is shown to be possible. Dynamo solutions are possible if the dynamo action is able to change the signs of torsion and curvature of the dynamo flux tube simultaneously.

Analytical solution of ${\alpha}$-dynamo equation representing strongly torsioned helical dynamo is obtained in the thin twisted Riemannian flux tubes approximation. The $\alpha$ factor possesses a fundamental contribution from torsion which is however weaken in the thin tubes approximation. It is shown that assuming that the poloidal component of the magnetic field is in principle time-independent, the toroidal magnetic field component grows very fast in time, actually it possesses a linear time dependence, while the poloidal component grows under the influence of torsion or twist of the flux tube. The toroidal component decays spatially with as $r^{-2}$ while vorticity may decay as $r^{-5}$ (poloidal component) where r represents the radial distance from the magnetic axis of flux tube. Toroidal component of vorticity decays as $r^{-1}$. In turbulent dynamos unbounded magnetic fields may decay at least as $r^{-3}$.

Modifications on a recently introduced fast dynamo operator by Chiconne et al [Comm Math Phys 173, 379 (1995)] in compact 3D Riemannian manifolds allows us to shown that slow dynamos are Lagrangean stable, in the sense that the sectional curvature of the Riemann manifold vanishes. The stability of the holonomic filament in this manifold will depend upon the sign of the second derivative of the pressure along the filament and in the non-holonomic case, to the normal pressure of the filament. Lagrangean instability is also investigated in this case and again an dynamo operator can be defined in this case. Negative curvature (Anosov flows) dynamos are also discussed in their stability aspects.

It is shown that the existence of the cosmic kinematic fast dynamos in Bianchi type IX rotating cosmological models, faces severe difficulties, due to the fact that in these models rotation increases without bounds, which is strictly forbidden by CMB astronomical date which imposes strong bounds on its rotation with respect with its rapid expansion. The only way out of this difficulty is to assume that at least one of the expansion directions of this anisotropic universe decays as fast as the amplification rate of primordial magnetic fields. A solution is found where only one direction of the anisotropic universe expands while the other two remain constants. We compute an amplification of the seed magnetic field in the case where Bianchi IX degenerates into de Sitter metric, fields amplify from 10^{-6}G to 10^{-5}G in spiral galaxies for a cosmological constant of the order $|{\Lambda}|<10^{33}s^{-2}$ and considering that the age of universe of the order of 10^{10}yrs. Another example is given by the ABC chaotic flows in the pseudo-Riemannian spacetime representing the Kasner anisotropic nonsingular universe.

The topological mapping between a torus of big radius and a sphere is applied to the Riemannian geometry of a stretched and twisted very thick magnetic flux tube, to obtain spherical dynamos solving the magnetohydrodynamics (MHD) self-induction equation for the magnetic flux tubes undergoing differential (non-uniform) rotation along the tube magnetic axis. Constraints on the shear is also computed. It is shown that when the hypothesis of the convective cyclonic dynamo is used the rotation is constant and a solid rotational body is obtained. As usual toroidal fields are obtained from poloidal magnetic field and these fields may be expressed in terms of the differential rotation ${\Omega}(r,{\theta}(s))$. In the case of non-cyclonic dynamos the torsion in the Frenet frame is compute in terms of the dynamo constant. The flux tube shear $\frac{\partial}{{\partial}r}{\Omega}$ is also computed. The untwisted tube case is shown to be trivial in the sense that does not support any dynamo action. This case is in agreement with Cowling antidynamo theorem, since in the untwisted case the tube becomes axially symmetric which the refereed theorem rules out. We also show that it is consistent with the Zeldovich antidynamo theorem which rules out planar dynamos. Knowledge of the differential rotation of the Earth, for example, allows one to place limits on the curvature and torsion of the flux tube axis and vice-versa, knowledge of the topology permit us to infer differential rotation and other physical parameters of the stars and planets.

A 3D metric conformally related to Arnold cat fast dynamo metric: ${ds_{A}}^{2}=e^{-{\lambda}z}dp^{2}+e^{{\lambda}z}dq^{2}+dz^{2}$ is shown to present a behaviour of non-dynamos where the magnetic field exponentially decay in time. The Riemann-Christoffel connection and Riemann curvature tensor for the Arnold and its conformal counterpart are computed. The curvature decay as z-coordinates increases without bounds. Some of the Riemann curvature components such as $R_{pzpz}$ also undergoes dissipation while component $R_{qzqz}$ increases without bounds. The remaining curvature component $R_{pqpq}$ is constant on the torus surface. The Riemann curvature invariant $K^{2}=R_{ijkl}R^{ijkl}$ is found to be 0.155 for the ${\lambda}=0.75$. A simple solution of Killing equations for Arnold metric yields a stretch Killing vector along one direction and compressed along other direction in order that the modulus of the Killing vector is not constant along the flow. The flow is shown to be untwisted. The stability of the two metrics are found by examining the sign of their curvature tensor components.

Inhomogeneous plasmas filaments instabilities are investigated by using the techniques of classical differential geometry of curves where Frenet torsion and curvature describe completely the motion of curves. In our case the Frenet frame changes in time and also depends upon the other coordinates taking into account the inhomogeneity of the plasma. The exponential perturbation method so commonly used to describe cosmological perturbatons is applied to magnetohydrodynamic (MHD) plasma equations to find longitudinal modes describing Alfven waves propagation modes describing plasma waves in the medium. Stability is investigated in the imaginary axis of the spectra of complex frequencies ${\omega}$ or $Im(\omega)\ne{0}$.

Two examples of the use of differential geometry in plasma physics are given: The first is the computation and solution of the constraint equations obtained from the Riemann metric isometry of the twisted flux tube. In this case a constraint between the Frenet torsion and curvature is obtained for inhomogeneous helical magnetic flux tube axis. In the second one, geometrical and topological constraints on the current-carrying solar loops are obtained by assuming that the plasma filament is stable. This is analogous to early computations by Liley [(Plasma Physics (1964)] in the case of hydromagnetic equilibria of magnetic surfaces. It is shown that exists a relationship between the ratio of the current components along and cross the plasma filament and the Frenet torsion and curvature. The computations are performed for the helical plasma filaments where torsion and curvature are proportional. The constraints imposed on the electric currents by the energy stability condition are used to solve the remaining magnetohydrodynamical (MHD) equations which in turn allows us to compute magnetic helicity and from them the twist and writhe topological numbers. Magnetic energy is also computed from the solutions of MHD equations.