Refrigeration cycle rates of the order of 15 Hz are desirable for efficient solid state based magnetocaloric cooling, placing an upper bound on the combined magnetic transition and the heat transfer times of the order of tens of msecs. We use microcalorimetry and magnetometry to probe the transition dynamics as a function of magnetic field sweep-rate, sample size, thermal environment, temperature and hydrostatic pressure in LaFe11.74Mn0.06Si1.20. Although second order caloric materials follow the magnetisation or demagnetisation driving field without lag, here we show that the field driven evolution of the first-order phase transition in La(Fe, Si)(13)-based compounds show temporal dynamics on timescales that are significantly longer than tens of msecs, associated with the thermal linkage within the sample and the linkage to the external bath. We observe that features associated with the first nucleation of the transition are field sweep rate independent, and from measurements of the latent heat we infer that the barriers to magnetisation and demagnetisation are of different magnitude. Increasing the temperature or applying hydrostatic pressure reduces the dynamic effects, suggestive of diminishing first-order character of the transition under these conditions.

We examine the character of the itinerant magnetic transition of DyCo2 by different calorimetric methods, thereby separating the heat capacity and latent heat contributions to the entropy-allowing direct comparison to other itinerant electron metamagnetic systems. The heat capacity exhibits a large lambda-like peak at the ferrimagnetic ordering phase transition, a signature that is remarkably similar to La(Fe,Si)(13), where it is attributed to giant spin fluctuations. Using calorimetric measurements, we also determine the point at which the phase transition ceases to be first order: the critical magnetic field, mu H-0(crit) = 0.4 +/- 0.1 T and temperature T-crit = 138.5 +/- 0.5 K, and we compare these values to those obtained from analysis of magnetization by application of the Shimizu inequality for itinerant electron metamagnetism. Good agreement is found between these independent measurements, thus establishing the phase diagram and critical point with some confidence. In addition, we find that the often-used Banerjee criterion may not be suitable for determination of first order behavior in itinerant magnet systems. DOI: 10.1103/PhysRevB.87.134421

The full text of this corrigendum is available in the PDF.

We measure transport ac losses in planar one-layer four-turn parallel coated conductor arrays (1 × 4) and in two-layer (2L) four-turn parallel coated conductor arrays (2 × 4) with a frequency up to 1 kHz. The horizontal separation between the conductors, i.e., g h , and the vertical separation between the neighboring superconducting layers, i.e., g v , were varied to investigate the transport ac loss dependence on g h and g v . In 2L arrays, the tapes in the top layer sit either aligned with the tapes in the bottom layer or aligned with the gaps between the tapes in the bottom layer. We show that the losses differently scale in arrays with low turn numbers to the scaling expected with an infinite array of tapes.

Magnetic refrigeration at around ambient temperatures has become of considerable technical and commercial interest over the last few years. It depends upon the magnetocaloric effect, and suitable working materials are those that undergo a phase transition which can be driven by a modest magnetic field. We focus here on one attractive family of intermetallics based on the compound La(Fe,Si)(13). Its metamagnetic phase transition is accompanied by a peak in the heat capacity that can be several times larger than the background and, for certain compositions and fields, also a well-defined first order transition with associated latent heat. It seems that some key aspects drawn from the bestiary of magnetism are particularly helpful in optimising magnetocaloric performance, namely itinerant electron magnetism and spin fluctuations. They appear to assist in maximising the entropy change at the phase transition without incurring the penalty of unduly large hysteresis. Many of these features are shared by other groups of compounds that have attractive performance.

AC loss measurements in five single pancake coils wound with 4 mm wide commercial high temperature superconductor wires were carried out to investigate the dependence of coil AC loss on separation between the superconductor layers in the neighbouring coil turns (g) and coil turn number (N) for a given number of ampere-turns NI. The highest frequency was set at approximately 1 kHz. The AC losses measured at different frequencies agreed well with each other. AC loss in the coils with the same N increases with decreasing g. When g is increased to 1.5 times the tape width (6 mm), the loss level is similar to that in an isolated wire. Transport AC loss per unit length in the pancake coils increases with increasing turn number. However, when g is increased to 1.5 times the tape width, the loss level in the coils with different turn number is almost the same. This indicates that, at the same NI, a coil with greater N is advantageous, even considering the conductor length difference. Therefore to achieve a given level of NI for a coil and minimize AC loss, we should favour coils with more turns. Coil AC loss can be measured using voltage taps attached to copper blocks outside a coil with reasonable accuracy. This is important for measuring AC loss in a coil with a complicated structure where voltage taps are not able to be attached inside the coil. AC loss in a nine-turn pancake coil wound with a 5 m long 9/2 Roebel cable (i.e. with 9 mm x 2 mm width strands) was measured using voltage loops arranged in each Roebel strand in the central turn of the coil. The AC loss in the coil was compared with two straight 9/2 Roebel cables with and without spacing between the strands. The combination of inter-strand spacing and turn spacing is an effective way to reduce AC loss in a single pancake coil wound with a Roebel cable.

Microcalorimetry has proven to be a versatile tool to investigate first order magnetic phase transitions as it can be used in different experimental modes to separate the latent heat from heat capacity. However, the methodology fails if the latent heat contribution is below instrumental resolution of 10 nJ. If the nucleation size of the new phase is much less than 100 μm, the typical size of the fragment measured, the latent heat could appear to be too distributed in temperature or magnetic field to be detected. Here, we show that for certain classes of magnetic transition, our microcalorimetry technique can be extended to enable an estimate of the latent heat to be obtained from a combination of heat capacity and magnetic measurements. This technique is best suited for material systems with weakly first order phase transitions, or highly distributed due to inhomogeneity.

Magnets fabricated with HTS wire can not be operated in a true persistent mode as superconducting joints of sufficient technological quality have not been achieved to date. In order to maintain a constant magnetic field in a HTS magnet a power supply has to be permanently employed, which then leads to heat losses in the cryo-system through the employment of current leads. By using a flux pump these losses can be minimized. We present a new flux pump based on 2G HTS wire. This device energized at 77 K a 2.7 mH 2G HTS double-pancake coil to its critical current of 49 A within 112 seconds. The operating principle will be described and data of the current ramping is shown. Considering the simplicity of the device and the potential to increase the generated current to 200 A and more, this new flux pump is very promising for many superconducting devices including HTS and LTS magnets and rotating machines.

Transport AC loss in a short length of 9/2 YBCO Roebel cable, (i.e. with 9 x 2 mm width strands), with 0.25 mm spacers between the strands is measured. The frequency varies from 59 to 354 Hz. The result is compared with the loss for a cable without spacers between the strands. Transport AC loss is decreased by the presence of the spacers. The AC loss reduction due to the extra spacing is more significant when the amplitude of the cable current is small compared to the cable I-c. The losses in the cable with spacers normalized by the square of the cable currents plotted against I-t/I-c approximately agree with those in a cable without spacers. Electromagnetic modelling was carried out for 9/2 and 8/2 cables, modelled as parallel stacks, to assist in understanding the above experimental results. The 8/2 configuration allows the greater use of symmetry to speed computations. Reasonable agreement between the numerical results and the measured results was obtained. This supports suggestions made in previous publications that the transport AC loss in a Roebel cable is roughly equivalent to the loss in two parallel stacks carrying the same current in each tape. The electromagnetic analysis in the 8/2 stacks shows the flux lines are more perpendicular to the strand face when the vertical space between strands is smaller, and this leads to a larger induced electrical field and larger AC loss. At small current amplitudes, the modelling shows the spacing has a strong effect on the AC loss in the surfaced part of the strands.

Here we study the influence of Al doping on the magnetization, heat capacity, and entropy change of La(Fe0.88Si0.12-xAlx)(13) where x = 0, 0.048, and 0.081. When x = 0, the system shows a remarkably sharp heat capacity feature associated with spin fluctuations coincident with, but quite distinct from the latent heat spike of the first order paramagnetic to ferromagnetic phase transition. With the addition of Al the magnetic and calorimetric features become more distributed in field, suggesting that Al adds disorder to the system. For both finite x compositions studied here, the latent heat disappears and the transition can be classified as second order. Although the entropy change associated with the transition is reduced once Al is substituted for Si, the adiabatic temperature change, Delta T-ad is still significant. In La(Fe0.88Si0.12-xAlx)(13) the balance between changes in the field dependence of the heat capacity with respect to overall Delta T-ad gain is highlighted, showing that a small amount of Al doping clearly offers some advantage for application.

Transport AC loss in a short length of 5/2 YBCO Roebel cable (five 2 mm wide strands) with weakly magnetic substrate is measured. The AC loss data are compared with those for a single strand. All the strands composing the cable and the single strand are cut from the same stock material. Transport AC loss in the 5/2 Roebel cable is larger than that predicted by both the Norris strip and ellipse models over a large range of the cable current. The difference between normalized AC losses in the Roebel cable and the single strand is larger at low I t / I c than at high I t / I c , where I t is the amplitude of transport current. These AC loss characteristics were not observed in transport AC loss results for YBCO Roebel cables composed of coated conductor with non-magnetic substrate. Results from an electromagnetic analysis of two parallel stacks of wire with magnetic substrate are presented to elucidate the results.

Here we study the influence of sample preparation on the magnetocaloric properties of CoMnSi. Slow cooling from the high temperature hexagonal phase of the melt to the room temperature orthorhombic phase encourages the formation of a homogeneous material with large entropy changes when the system undergoes a coincident first order structural and (meta) magnetic transition. Samples that were quenched directly after annealing show a compressed a axis lattice parameter. Hall probe imaging indicates that the quenched sample has spatially inhomogeneous magnetic properties, which we attribute to strain because within error neither x-ray diffraction nor energy dispersive x-ray analysis indicates a second compositional phase. Calorimetric methods and global magnetization are used to examine the entropy changes of the pure and mixed magnetic phase compounds and we make a direct comparison of these materials in terms of their refrigerant capacity.

Point contact Andreev reflection spectra have been taken as a function of temperature and magnetic field on the polycrystalline form of the newly discovered iron-based superconductor Sr2ScFePO3. A zero bias conductance peak which disappears at the superconducting transition temperature dominates all of the spectra. Data taken in high magnetic fields show that this feature survives until 7 T at 2 K and a flattening of the feature is observed in some contacts. Here we inspect whether these observations can be interpreted within a d-wave, or nodal order parameter, framework which would be consistent with the recent theoretical model where the height of the P in the Fe-P-Fe plane is key to the symmetry of the superconductivity. However, in polycrystalline samples care must be taken when examining Andreev spectra to eliminate or take into account artefacts associated with the possible effects of Josephson junctions and random alignment of grains.

Here we study the calorimetric and magnetic behaviour of melt-spun LaFe11.6Si1.4, a potential magnetic refrigerant material system that exhibits the rare combination of a large entropy change and low thermal and magnetic field hysteresis. We are able to separate the calorimetric contribution from latent heat and changes in equilibrium heat capacity explicitly by using two separate calorimetric probes. The heat capacity of this sample exhibits significant changes of the order of 500–1000 J K−1 kg−1 in response to magnetic field that results in large changes in entropy. The different contributions to entropy change from latent heat and heat capacity are shown to evolve as the material is field driven through its itinerant metamagnetic transition. We demonstrate explicitly that in the melt-spun sample studied here, the majority of the total entropy change comes from the equilibrium change of heat capacity.

Transport AC loss in a short length of 9/2 YBCO Roebel cable (nine 2 mm wide strands) is measured. The AC loss data are compared with those in a 5/2 YBCO Roebel cable (five 2 mm wide strands) as well as that in a single strand. All the strands composing the cables and the single strand are insulated and cut from the same stock material. The validity of the measurement method was reconfirmed by results at a range of frequencies. At a wide range of I-t/I-c, the normalized AC losses in the Roebel cable were around 6.2-6.7 times of those in the single strand. This is less than the nine times predicted for a tight bundle of nine conductors. The normalized transport AC losses in the 5/2 Roebel cable are much smaller than those in the 9/2 Roebel. This should be due to larger superposition of magnetic field in the 9/2 Roebel. The I-c of the 9/2 and 5/2 Roebel cables is determined by serial connection of the strands. This eliminates the effect where differing resistances in the current terminations cause uneven current sharing between strands when the strands are connected in parallel.

Oxypnictide superconductor NdFeAsO0.85 sample was irradiated with 2 GeV Ta ions at a fluence of 5x10^10 ions/cm2. High resolution transmission electron microscopy study revealed that the irradiation produced columnar-like defects. The effect of these defects on the irreversible magnetisation in polycrystalline randomly oriented fragments was studied as a function of field angle and field sweep rate. We find that the critical current density is enhanced at fields below the matching field (~1 Tesla) but only marginally. The pinning enhancement is anisotropic and maximum along the defect direction at high temperatures but the pinning then becomes more isotropic at low temperatures. The creep rate is suppressed at high temperatures and at fields below the matching field, indicating the columnar defects are efficient pinning sites at these H and T conditions.

Here we examine the constituent components that make a magnetocaloric material attractive for application. The field-temperature phase diagram is studied and using calorimetry, the first-order and second-order components of the magnetic field-driven magneto-structural phase transition in CoMnSi0.92Ge0.08 are extracted. It is demonstrated that below 262 K the transition shows a latent heat component associated with first-order behavior when the material changes from antiferromagnetic to ferromagnetic order. Such a transition is known as a metamagnetic transition. We identify 262 K as a tricritical point and above this temperature T-crit the transition shows only continuous, second-order characteristics. Hall-probe imaging that has a five micron pixel resolution is then used to study the striking differences in the spatial evolution of the transition above and below T-crit. We demonstrate that the hysteresis of the transition is linearly related to the magnitude of the latent heat; an observation that has important implications for the use of this and other first-order systems for application as magnetic refrigerants.