Knots in electromagnetism

Maxwell equations in vacuum allow for solutions with a non-trivial topology in the electric and magnetic field line configurations at any given moment in time. One example is a space filling congruence of electric and magnetic field lines forming circles lying on the surfaces of nested tori. In this example the electric, magnetic and Poynting vector fields are orthogonal everywhere. As time evolves the electric and magnetic fields expand and deform without changing the topology and energy, while the Poynting vector structure remains unchanged while propagating with the speed of light. The topology is characterized by the concept of helicity of the field configuration. Helicity is an important fundamental concept and for massless fields it is a conserved quantity under conformal transformations. We will review several methods by which linked and knotted electromagnetic (spin 1) fields can be derived. A first method, introduced by A. Rafiada, uses the formulation of the Maxwell equations in terms of differential forms combined with the Hopf map from the three-sphere S-3 to the two-sphere S-2. A second method is based on spinor and twistor theory developed by R. Penrose in which elementary twistor functions correspond to the family of electromagnetic torus knots. A third method uses the Bateman construction of generating null solutions from complex Euler potentials. And a fourth method uses special conformal transformations, in particular conformal inversion, to generate new linked and knotted field configurations from existing ones. This fourth method is often accompanied by shifting singularities in the field to complex space time points. Of course the various methods must be closely related to one another although they have been developed largely independently and they suggest different directions in which to expand the study of topologically non-trivial field configurations. It will be shown how the twistor formulation allows for a direct extension to massless fields of other spin values, such as spin-2 fields satisfying the linearized Einstein vacuum equation, and how the formulation by A. Rafiada can be extended to fields for which the electric and magnetic fields are not orthogonal everywhere. Underlying the various methods is the fact that electric and magnetic field lines can be described as the level curves of complex functions. Compactification of R-3 naturally leads to finite energy solutions because the fields at infinity in all directions should all converge towards zero. An intriguing question that is raised by the finite energy is whether there is a connection to the quantization of the classical electromagnetic field. We will review some issues related to this question. Another interesting question is why the general formulation of topologically nontrivial solutions uses the electric and magnetic fields instead of the electromagnetic vector potentials. This leads to a discussion of the Clebsch representation of the electromagnetic field strength 2-form. Finally, a topic of great interest is the possibility of experimentally generating and investigating linked and knotted field configurations. Since the non-trivial topological field solutions exploit the special conformal symmetry of the underlying vacuum wave equations it will only be possible to approximate the solutions in an experiment, which necessarily introduces material objects that will break the special conformal symmetry. We will review the research on plasma configurations in which the magnetic field-line configuration approximates plasma torus knots leading to the prediction of topological solitons in plasma. (C) 2016 Elsevier B.V. All rights reserved.