[From my M.Sc. thesis
paper]
Zinc-Antimonide
(ZnSb) thin films have been prepared by using single-source resistive heating
vacuum evaporation technique on to glass substrate at a pressure of about
7.99x10-4 Pa.
Electrical
properties of as-deposited films are studied. The value of resistivity is low
and its value range from 0.02 to 0.05 W-m.
Temperature coefficient of resistance (TCR) studies shows negative values
indicating semiconducting nature of the films. Activation energy and grain size
of the films is also studied. The activation energy of the films is found to
decrease with film thickness. The values of activation energy vary from 0.37 to
0.52 eV. It is also found that the grain size increases with increasing film
thickness. The range of the grain size is 9 to 13 Å. Thickness dependent
resistivity measurements follow the Fuchs-Sondheimer size effect theory.
The
band gap of the films was determined from absorption coefficient data. The band
gap of the films was found to be ~ 0.5 eV. This result is well in agreement with
the band gap of balk ZnSb semiconductor (0.56 eV). From the optical measurement,
it was found that the band is of indirect in nature.
Refractive
index of the films was calculated from optical data and the value of the
refractive index was found to be ~ 2.6. Dielectric constant of the films was
determined and found to be ~ 8. The optical conductivity of 70 nm films was
found to be of the order of 6r1015/sec.
Electrons have charge as well as spins and up to all types of semiconductor electronics uses the movement of electrons (charge) and not concern the electron spins. A new approach in electronics known as magnetoelectronics is emerging very fast that is based on the spin orientation of the high-density electron carriers in magnetic metals rather than on the low-density electrons or holes in traditional semiconductor electronics.
We can generate spin-polarized carriers and this important issue is to know how long these electrons remember their spin orientation. This is especially important for electronic application, because if the spin relax too rapidly, then distance traversed by the spin polarized carriers in a device will be too short to serve any practical purpose. M. Johnson et. al. in 1985 have shown that the spin diffusion length of injected carriers from ferromagnetic to paramagnetic element Al is 0.1 mm at 40K. This result showed that spin-polarized current could travel distance comparable to those in modern electronic device structures without loosing “memory” of their spin orientation.
Spin-polarized transport will occur naturally in any material for which there is an imbalance of spin population at the Fermi level. This imbalance commonly occurs in ferromagnetic metals. Therefore, a ferromagnetic metal may be used as a source of spin-polarized couriers injected into a semiconductor, a superconductor, a metal or can be used to tunnel through an insulating barrier.
The world of magnetoelectronic applications becomes very receptive when Fert’s group in France discovered “giant magnetoresistance” (GMR) effect in 1988. GMR is a quantum mechanical effect observed in layered magnetic thin film structures that are composed of alternating layers of ferromagnetic and non-magnetic layers. When the magnetic moments of the ferromagnetic layers are parallel, the spin-dependent scatterings of the carriers are minimized, and the resistance is very low. On the other hand, when the moments of ferromagnetic layers are anti-parallel, the spin-dependent scattering of the carriers is maximized and layers offer very high resistance. The directions of the magnetic moments are manipulated by external magnetic field applied to the samples.
The spin-relaxation length is much longer than the typical 1-10nm layers in most device structures. This means that an electron can pass through many layers before forgetting its spin-orientation. Within this length, each magnetic layer acts as a spin filter. This explains the increase of GMR effect with number of layers. T. Takahata et al. in 1989 has been found large magnetoresistive effect in Co/Au/Co sandwiches and multi-layers with perpendicular anisotropy. B. Dieny et al. in 1991 has been found that this effect is increases with increasing number of layers.
An important concept (M.Johnson et al. in 1987) involved in spin-polarized transport is the shift of subband chemical potential that accompanies the accumulation of spin-polarized electrons in a normal metal. If the layer thickness is less than the electronic mean free path then interfacial spin dependent scattering is minimized and GMR effect is observed. When spin-polarized current is driven from magnetic film into a non-magnetic film faster than spin-polarization can diffuse away from the interface, a non-equilibrium population of spin polarized electrons builds up in this region. Due to this non-equilibrium population an inequivalent chemical potentials for the spin-up and spin-down subbands of the metal is established. The chemical potential of the ferromagnet is held in equilibrium by the intrinsic ferromagnetic/non-magnetic metal interface associated with the non-equilibrium spin accumulation, that tries to drive the electron back across the interface and into ferromagnet. Because spin and charge both are carried by the electron, a gradient of spin density results in an electric field, which can generate current flow.
Using this effect a three terminal bipolar device known as spin transistor have been demonstrated. This device consists of a normal metal sandwiched between two ferromagnetic layers. Current is driven from the first ferromagnetic film (emitter) into the non-magnetic metal base. Second ferromagnetic film is used as a collector. If the magnetic moments of the two ferromagnetic layers are parallel, spin accumulation in the base will create an electric field that pushes current into the collector. On the other hand, when the magnetic moments are anti-parallel, the spin accumulation electric field at the base-collector has opposite sign, current is pulled from the collector to the base produces a negative current. All ferromagnetic/insulator/ferromagnetic devices carry much lower current than all metal GMR devices, which may be an advantage for portable device that has limited power. However the high impedance of tunneling device may prove to be unattractive in terms of response time or noise. This challenge increases as device sized are reduced because tunneling devices carry currents perpendicular to the plane of the films and as the area of the device shrinks, the resistance increases. Therefore a metal base transistor is more attractive for magnetoelectronic device applications than MTJ’s.
Main disadvantage of a ferromagnetic/nonmagnetic metal/ferromagnetic structure is that the mean free path of electrons is not enough long at room temperature to avoid spin-flip scattering of the spin-polarized carriers. To eliminate this disadvantage researchers need to introduce a base that exhibits long mean free path. Several researcher groups are pursuing the injection of spin-polarized carriers into a two-dimensional electron gas channel that is formed at a compound semiconductor heterostructure interface. The long mean free path of electrons in this channel is expected to yield micrometer length paths to avoid spin-flip scattering of the carriers.
The main objective of the research proposal is to introduce and develop a base for spin-transistor that exhibits a long mean free path for the carriers. Co/Cu/Co is a possible structure because the d-conduction band of Cu closely matches the majority spin d-conduction band of Co that can minimize the spin-flip scattering of the spin-polarized carriers. Multi-layer magnetic metal film is also an attractive material for the base of a spin-transistor because applying a magnetic field can control the mean free path of the electrons within the base.
Home
Page