A superconducting point contact is used to determine the spin polarization at the Fermi energy of several metals. Because the process of supercurrent conversion at a superconductor-metal interface (Andreev reflection) is limited by the minority spin population near the Fermi surface, the differential conductance of the point contact can reveal the spin polarization of the metal. This technique has been applied to a variety of metals where the spin polarization ranges from 35 to 90 percent: Ni0.8Fe0.2, Ni, Co, Fe, NiMnSb, La0.7Sr0.3MnO3, and CrO2.

Using the point contact Andreev reflection technique, we have carried out a systematic study of the spin polarization in the colossal magnetoresistive manganite, La0.7Sr0.3MnO3 (LSMO). Surprisingly, we observed a significant increase in the current spin polarization with the residual resistivity. This counterintuitive trend can be understood as a transition from ballistic to diffusive transport in the contact. Our results strongly suggest that LSMO does have minority spin states at the Fermi level. However, since its current spin polarization is much higher than that of the density of states, this material can mimic the behavior of a true half-metal in transport experiments. Based on our results we call this material a transport half-metal.A half-metallic ferromagnet is a metal that has an energy gap at the Fermi level, E F , in one of the two spin channels. Only the other channel has states available for transport, and thus the electric current is fully spin-polarized. Finding half-metallic or other highly spin-polarized metals would bring about major advances in magnetoelectronics, since device performance improves dramatically as the spin polarization of the metal approaches 100%.1 Although half-metallicity has been predicted in quite a number of materials, the experimental situation is still controversial, especially for the manganese perovskite, La 0.7 Sr 0.3 MnO 3 (LSMO). Theoretical 2 and experimental values 3-6 of the spin polarization of this fascinating material with highly unusual structural, magnetic and electronic properties, obtained by different techniques vary from 35% to 100%. Not surprisingly, when Park et al. concluded from their spin-resolved photoemission spectroscopy measurement that LSMO is completely spin-polarized 3 it attracted immediate attention. This result was important not only from a practical viewpoint, but also as a potential new insight into the microscopic physics of this system, since the values of the spin polarization are extremely sensitive to the band structure of LSMO. Importantly, the measured value of the spin polarization, P n , depends on the experimental technique. It is often possible 10 to define P n in the following form:where N ↑ (E F ), N ↓ (E F ) and v F ↑ , v F ↓ are the majority and minority spin DOS and the Fermi velocities, respectively. This definition allows a direct comparison between different experiments and the theory, since all the quantities in Eq. 1 can be evaluated from the band structure. The spin polarization P 0 (n =0) measured by spinresolved photoemission measurements is determined only by the DOS at the Fermi level.11 Transport experiments measure a different spin polarization, which includes the Fermi velocities (Eq.1). In the ballistic, or Sharvin, limit (mean free path, L, larger than the contact size, d) the DOS is weighted linearly with v F , and P 1 is measured. 12In the diffusive, or Maxwell regime (L < d), as in the classical Bloch-Bolzmann theory of transport in metals, the weighting is quadratic in v F (n=2) and P 2 is measured (a...