One view of the cuprate high-transition temperature (high-T c ) superconductors is that they are conventional superconductors where the pairing occurs between weakly interacting quasiparticles, which stand in one-to-one correspondence with the electrons in ordinary metals -although the theory has to be pushed to its limit [1]. An alternative view is that the electrons organize into collective textures (e.g. charge and spin stripes) which cannot be mapped onto the electrons in ordinary metals. The phase diagram, a complex function of various parameters (temperature, doping and magnetic field), should then be approached using quantum field theories of objects such as textures and strings, rather than point-like electrons [2,3,4,5,6]. In an external magnetic field, magnetic flux penetrates type-II superconductors via vortices, each carrying one flux quantum [7]. The vortices form lattices of resistive material embedded in the non-resistive superconductor and can reveal the nature of the ground state -e.g. a conventional metal or an ordered, striped phase -which would have appeared had superconductivity not intervened. Knowledge of this ground state clearly provides the most appropriate starting point for a pairing theory. Here we report that for one high-T c superconductor, the applied field which imposes the vortex lattice, also induces antiferromagnetic order. Ordinary quasiparticle pictures cannot account for the nearly fieldindependent antiferromagnetic transition temperature revealed by our measurements.La 2-x Sr x CuO 4 , is the simplest high-T c superconductor. The undoped compound is an insulating antiferromagnet, where the spin moments on adjacent Cu 2+ ions are antiparallel [8]. Introduction of charge carriers via Sr doping reduces the ordered moment until it vanishes at x<0.13. In addition, for x>0.05 the commensurate antiferromagnetism is replaced by incommensurate order [2,3,9,10], where the repeat distance for the pattern of ordered moments is substantially larger than the spacing between neighbouring copper ions. La 2-x Sr x CuO 4 becomes a 2 superconductor for Sr dopings of 0.06

The excitations responsible for producing high-temperature superconductivity in the copper oxides have yet to be identified. Two promising candidates are collective spin excitations and phonons 1 . A recent argument against spin excitations is based on their inability to explain structures observed in electronic spectroscopies such as photoemission 2-5 and optical conductivity 6,7 . Here, we use inelastic neutron scattering to demonstrate that collective spin excitations in optimally doped La 2−x Sr x CuO 4 are more structured than previously thought. The excitations have a two-component structure with a lowfrequency component strongest around 18 meV and a broader component peaking near 40-70 meV. The second component carries most of the spectral weight and its energy matches structures observed in photoemission 2-5 in the range 50-90 meV. Our results demonstrate that collective spin excitations can explain features of electronic spectroscopies and are therefore likely to be strongly coupled to the electron quasiparticles.Since their discovery, considerable progress has been made in understanding the properties of the high-critical-temperature, T c , cuprate superconductors. We know, for example, that the superconductivity involves Cooper pairs, but with d-wave rather than the s-wave pairing of conventional Bardeen-CooperSchrieffer (BCS) superconductors. One outstanding issue is the pairing mechanism itself. For conventional superconductors, identifying the bosonic excitations that strongly couple to the electron quasiparticles played a pivotal role in confirming the phonon-mediated pairing mechanism 8,9 . In the case of the copper oxide superconductors, electronic spectroscopies such as angle-resolved photoemission (ARPES) and infrared optical conductivity measurements 6,7 have revealed structures in the lowenergy electronic excitations, which may reflect coupling to bosonic excitations. ARPES measurements on Bi 2 Sr 2 CaCu 2 O 8 , Bi 2 Sr 2 CuO 6 and La 2−x Sr x CuO 4 have shown rapid changes or 'kinks' in the quasiparticle dispersion, E(k), for energies in the range 50-80 meV (refs 2-5). These features in ARPES have been interpreted in terms of coupling to phonon modes 5 . However, the ARPES measurements do not distinguish between coupling to lattice and spin excitations. Identifying phonons as the strongly coupled bosons is not without its difficulties: we must explain what is special about the phonons in the cuprates; interactions with phonons do not naturally explain other important properties of the cuprates, such as the large linear temperature dependence of the normal-state resistivity at optimal doping and the origin of d-wave symmetry of the superconducting gap itself.The interpretation of the kinks and other features in electronic spectroscopies 2-7 in terms of coupling to collective spin excitations 10 has been hampered by the lack of magnetic spectroscopy data. Most neutron scattering data refer to YBa 2 Cu 3 O 6+x , a compound for which ARPES data are scarce. Although ARPES kinks have also been re...

High-resolution neutron scattering experiments on optimally doped La2-xSrxCuO4 (x=0.16) reveal that the magnetic excitations are dispersive. The dispersion is the same as in YBa2Cu3O6.85, and is quantitatively related to that observed with charge sensitive probes. The associated velocity in La2-xSrxCuO4 is only weakly dependent on doping with a value close to the spin-wave velocity of the insulating (x=0) parent compound. In contrast with the insulator, the excitations broaden rapidly with increasing energy, forming a continuum at higher energy and bear a remarkable resemblance to multiparticle excitations observed in 1D S=1/2 antiferromagnets. The magnetic correlations are 2D, and so rule out the simplest scenarios where the copper oxide planes are subdivided into weakly interacting 1D magnets.