I had separated magnetic fraction of typical construction sand obtained from river. Being a Student from Bangladesh, I can readily infer the riverine sand to be source from eastern Himalayas (Assam-Meghalaya region) upon considering course of the possible rivers from which the sand can be obtained. The sand is possibly collected from Northern Mymensingh or Sylhet Division of Bangladesh. A typical Loudspeaker magnet ( about 8 cm dia and 1.5 cm thick, collected from local Junkyard) and a magnetic stainless steel utensil (about 25 cm dia and 1.5 cm rim height) was used to separate the magnetic fraction.

the strongest magnetic particles attached to the magnetic assembly (magnet below the center of utensil, sand of about 5 g per batch hand-vibrated over the plate being held horizontal) was black colored, presumably magnetite (and possibly very low amount of ilmenite). nevertheless, there was weaker but certain attraction to magnetic field by grains of brown to off-white particles (about .5-1 mm size), which can not be wholly hematite, but possibly other iron (and perhaps titanium?) containing minerals. The yield of magnetic fraction upon 1-time magnetic separation is 2-3%, but repeated magnetic separation of around 5-6 times can generate yield close to 10% or even more.

As far as I know, Magnetic fraction of heavy minerals found in river sand at Bangladesh (except for beach placer deposits) is around 5%. So, what can be these weakly magnetic particles? can it be Titanium containing mineral like Leucoxene? can the mineralogical nature be estimated without XRD and qualitative chemical analysis?