D. Melville

High gradient magnetic separation of red cells from whole blood

It is demonstrated that red blood cells may be separated from other blood components using a high gradient magnetic separator. The (SI) magnetic susceptibility of red blood cells is estimated to be 3.88×10-6when the haemoglobin is in the completely deoxygenated state. The magnetic separation effects have been studied using a filter of circular stainless steel wire with flow rates between 10-4ms-1and 6×10-4ms-1and magnetic fields in the range 0.6 to 2.4 T. The results indicate that the filter quickly saturates and the variation of filter performance with field and flowrate is discussed in terms of the force balance and the particle trajectory model. Scanning electron microscopy and free haemoglobin tests on the filtered red blood cells show no evidence of serious damage or cell rupture.

First order magnetisation processes in Pr 2 (Co 1-x Fe x ) 17 alloys

The discontinuity in the hard direction magnetisation curve for the alloy Pr 2 (Co 0.6 Fe 0.4 ) 17 has been studied over the temperature rauge 5K-300K using a partially aligned polycrystalline sample. By means of the singular point detection technique we have measured the temperature variation of the anisotropy field and also the position of the magnetisation discontinuity. We conclude that the discontinuity is associated with a first-order rotation of the magnetisation vector, and its appearance as temperature is lowered is due to the rapid decrease of the second anisotrcpy constant K 2 to below a negative value K 2 = -K 1 /6. Analysis of the anisotropy constants obtained by Shanley for a Pr 2 (Co 0.8 Fe 0.2 ) 17 alloy enable a good estimate to be made of the position of the anomalous discontinuity for this alloy.

A bench top magnetic separator for malarial parasite concentration

A bench top magnetic separator is described which is suitable for separating red blood cells infected with malarial parasites from whole blood. The separator is based on a compact 0.7T permanent magnet and uses a high gradient magnetic filter of approximately 2.5 cm3volume. Experiments with malaria parasitised mouse red cells (Plasmodium berghei) show that final parasite concentrations in excess of 90% can be achieved.

Differential Blood Cell Separation using a High Gradient Magnetic Field

A technique for the separation of erythrocytes from whole blood is described which exploits the magnetic property of haemoglobin in the reduced state. The technique is characterized by the use of a filter consisting of a cylinder, containing stainless steel wire mesh, placed between the jaws of an electro magnet. When activated, the electromagnet induces a magnetic field gradient in the vicinity of each of the constituent wires, sufficient to attract and trap erythrocytes in suspension. The number of erythrocytes captured varies with the applied field (0–1.4 Tesla in these experiments) and flow rate (1.9–12.9 × 10−4 m s−1). The capture process does not cause haemolysis or observable surface damage to the erythrocytes and neither leucocytes nor platelets are retained by the filter.

Fractionation of blood components using high gradient magnetic separation

We review a wide range of strategies for using high gradient magnetic separtion to fractionate blood components. The physics bases of the techniques involve: (i) variations in cell haemoglobin and cell geometry; (ii) changes in haemoglobin magnetic susceptibility in various carrier environments; (iii) the use of paramagnetic carrier fluids. Single wire and bulk filter studies are reported which show that these techniques can be successfully applied to: (a) white cell and platelet counting; (b) cell fractionation on haemoglobin type and content; (c) diamagnetic white cell and platelet separation; (d) nucleated red cell studies; (e) sickle cell separation. We show that red blood cells can be magnetically separated both paramagnetically and diamagnetically and offer a basis or the assessment of diamagnetic capture efficiency.

Metal/Langmuir-Blodgett/metal junctions using PbIn superconducting electrodes☆

Current-voltage characteristics are presented for tunnel junctions with insulating layers of 20,21-oxiranheneicosanoic acid. The critical current dependence on the magnetic field indicates conduction through small regions and is consistent with a model based on bridging filaments5,6. Small discrete changes between successive samples of critical current are interpreted in terms of an average area of about 2 × 10−15 m2 for each filamentary superconductor.

Dye permeability at phase transitions in single and binary component phospholipid bilayers

By encapsulating a pH-sensitive dye, phenol red, in multilamellar liposomes of DMPC, DPPC and DMPC/DPPC mixtures, the permeability of these phospholipid bilayers to dye as a function of temperature has been studied. For both DMPC and DPPC liposomes, dye release begins well below the main gel-to-liquid-crystalline phase transition (24 degrees C and 42 degrees C, respectively) at temperatures corresponding to the onset of the pretransition (about 14 degrees C and 36 degrees C, respectively) with DPPC liposomes exhibiting a permeability anomaly at the main phase transition (42 degrees C). The perturbation occurring in the bilayer structure that allows the release of encapsulated phenol red (approx. 5 A diameter) is not sufficient to permit the release of encapsulated haemoglobin (approx. 20 A diameter, negatively charged). In liposomes composed of a range of DMPC/DPPC mixtures, dye release commences at the onset of the pretransition range (determined by optical absorbance measurements) and increases with increasing temperature until the first appearance of liquid crystalline phase after which no further dye release occurs. Interestingly, the dye retaining properties of DMPC and DPPC liposomes well below their respective pretransition temperature regions are very different: DMPC liposomes release much encapsulated dye at incubation temperatures of 5 degrees C whilst DPPC liposomes do not.

Inviscid approximation trajectories in high gradient magnetic separation

The precise conditions corresponding to the inviscid approximation in the axial configuration for high gradient magnetic separation (HGMS) trajectories has been modeled. Using red blood cells in a stopped-flow situation, it has been shown that the trajectories follow the analytical equations developed in the potential flow model. A small improvement is obtained by taking account of creeping flow corrections. The least well-defined parameter for nonspherical particles is the ratio (A/l) of area to characteristic length. By measuring (A/l) independently in terminal velocity experiments under gravity, a good fit is obtained to all the details of the trajectory equations without resort to adjustable parameters. The chief problems that arise are due to the nonsphericity of the red cell and the consequent possibility of (A/l) changing during the course of a trajectory.

Axial particle trajectory measurements in high-gradient magnetic separation

A technique has been developed for studying the trajectories of paramagnetic particles in a high-gradient magnetic separator. Human red blood cells are used as the paramagnetic particle and their trajectories in relation to a single magnetized stainless-steel wire are imaged. The flow velocity is parallel to the wire and has a value of approximately 50 μm . s-1. The magnetic field of 1.4 T is applied perpendicular to the wire. A phase modulation technique is used to image the red blood cells with a resolution of less than 0.5 μm. The measured particle trajectories confirm the main features of current theoretical models.

Transverse particle trajectories in high-gradient magnetic separation

We have used human red blood cells as a model particle system for studying transverse particle trajectories relevant to high gradient magnetic separators. Trajectories measured around a single 50 μm diameter wire are compared with calculations based on potential and laminar flow models of the fluid motion. The magnetic velocity is determined in an independent experiment which precisely models potential flow. At the low Reynolds numbers (Re = 10-3) used here it is found that the laminar flow modeI gives better agreement with experiment.