Magnetization recovery

Media. Recovery Wet Magnetic Drums. In HMS recovery service, the particle size of the feed, particularly the magnetic portion, is quite fine so that dmm wear is not a serious problem. Maximum magnetic recovery is important and the highest magnetic purity and soHds content in the magnetic concentrate are desired. 

A single wet dmm separator is typically used in plants where the feed soHds are in the 10—15% range. Extremely dilute feeds make magnetic recovery more difficult and feed volume should be reduced by a factor of 0.5—0.75 for such dilute feed. Por soHds in the 20—25% range, either single-dmm 914-mm units or double-dmm 762-mm units are recommended. Above 25% soHds, 914 mm dia double dmms are recommended. 

Magnetic Content of Feed Solids. The magnetic content of the soHds is typically >60%. Conventional selection procedures can be used in such cases. When the magnetic content of the soHds falls below 60% the feed volume should be reduced and multidmm separators considered in order to maintain higher magnetic recovery. 

Media Si. Media are suppHed in several size grades and the grade used varies at each plant. The finer grades improve media stabiUty, but finer particles are more difficult to recover and the feed rate of these finer-grade slurries should be reduced by a factor of 0.5—0.75 to maintain magnetic recovery. A typical size analysis as used in various heavy-media separation plants treating coal (qv) is given in Table 4. 

Finally, selective separation and dewatering of one suspended substance in a slurry containing different minerals or precipitates is possible by selectively adsorbing a magnetic material (usually hydrophobic) onto a soHd that is also naturally or chemically conditioned to a hydrophobic state. This process (Murex) was used on both sulfide ores and some oxides (145). More recently, hydrocarbon-based ferrofluids were tested and shown to selectively adsorb on coal from slurries of coal and mineral matter, allowing magnetic recovery (147). Copper and zinc sulfides were similarly recoverable as a dewatered product from waste-rock slurries (148). 

When Wqi / Wq2 the magnetization recovery may appear close to singleexponential, but the time constant thereby obtained is misleading [50]. The measurement of 7) of quadrupolar nuclei under MAS conditions presents additional complications that have been discussed by comparison to static results in GaN [50]. The quadrupolar (two phonon Raman) relaxation mechanism is strongly temperature dependent, varying as T1 well below and T2 well above the Debye temperature [ 119]. It is also effective even in cases where the static NQCC is zero, as in an ideal ZB lattice, since displacements from equilibrium positions produce finite EFGs. 

NMR relaxometry Free induction decay (T2 ) or solid echo Spin echo decay (T2) Magnetization recovery curve (7)) Eads (1998)... 

Magnetic recording techniques, 17 419 Magnetic recovery, in ore concentration, 15 447... 

In a Ti experiment, the magnetization recovery M(f) is measured after a saturation piflse. In our case we have two contributions in Eq. 9 so that M(f) contains two time constants which differ by more than a factor of three from each other so that they can be separated by a corresponding fit procedure. The two time constants ti and t2 correspond to the relaxation rates at cul and 2 l. 

Feed Solids Content. A good HMS plant operation keeps the medium as free of fines as possible by effective screening of the heavy-media separation vessel feed. Reduced fines reduce viscosity problems in the medium and result in sharper separation of sink and float products. It also improves magnetic recovery on the magnetic drum separators and gives a cleaner magnetic concentrate. The use of cyclones in the HMS circuit, either as the heavy-media separation vessel or as a densifier for rinse or wash water, increases the solids content and must be evaluated in selecting the media recovery wet dmm separators for plants in which cyclones are used. 

Desired Magnetic Recovery. In ore concentration, maximum recovery is desired at all times. Rejection of middling particles, although sometimes desired, is difficult to accomplish on wet magnetic drum separators. 

Experimental techniques are available to measure magnetization recovery under different conditions. These are, for instance, the selective and non-selective variants of the inversion recovery experiment described in Section 1.7.4, and will be discussed in more detail in Chapter 9. We anticipate here, from Chapters 7 and 9, in a qualitative way, the kind of information contained in these experiments. 

In the case of T measurements we have mentioned that cross relaxation provides multiexponential magnetization recovery (Sections 1.7.4 and 7.2.2). A far less known analogy may occur in the linewidths, as already discussed (Section 8.8) when two protons are dipole-dipole coupled and cross correlation occurs between Curie relaxation and proton-proton dipolar relaxation. In this case, we are in the presence of two overlapping signal components with different linewidths, i.e. of biexponentiality in T2 [35], Pulse sequences are available to remove the effects of cross correlation [36]. Such effects are common in paramagnetic metalloproteins where Curie relaxation is usually relevant (in principle, such cross correlation effects can be operative also in the case of 7i, although only to the extent that Curie relaxation on T is effective). 

Rb and 1H SLR rate as a function of temperature is a very important parameter which shows the suppression of phase transition and reveals the frustration in the mixed system. Temperature dependence of Ti in any ordered system can be described by the well known Bloembergen-Purcell-Pound (BPP) type expression. However, disordered systems show deviations from BPP behaviour, showing a broad distribution of relaxation times. The magnetization recovery shows a stretched exponential recovery of magnetization following M(t)=Mo(1 — 2 exp (— r/Ti) ) where a is the stretched exponent. 

Sobol et al.8 have measured proton SLR time (7 ) in Rb1 x(NH4)xH2As04 systems in the range 100-4.2 K. Magnetization recovery was found to be non-exponential in the entire range of temperature. The MR data fit to a stretched exponential recovery and the exponent a was found to be temperature dependent implying broadening of the distribution of microscopic correlation times p(r ) with decreasing temperature. 

---