Özden Özdemir's research while affiliated with University of Toronto and other places

... Thus, the linear distribution of magnetic susceptibility maxima ( Fig. 2 and Fig. S2) are most likely due to enhanced concentration of traffic-derived magnetic minerals. Since magnetic characteristics of natural materials are also grain-size dependent (Dunlop and Ö zdemir, 1997), they may be further used for establishing grain size trends in the magnetic data set. Anhysteretic susceptibility (χ ARM ) shows spatial variations which are generally similar to those of χ (Fig. S3 from the SI), except for lower values ((0 -100)*10 -8 m 3 /kg), typical of natural Vertisols obtained for sites from the older part of the garden, covered with denser forest. ...

... The curves resemble data for crushed magnetite sized 3 − 9 µm (Dunlop & Özdemir 2018, Fig. 6 A robust method for the calculation of the Verwey transition temperature using remanence data was described by Liu et al. (2004), however a physically justified technique for doing the same with magnetic susceptibility data is lacking. An empirical approach is to calculate the numerical derivative of the data and select the point of steepest ascent as the transition temperature. ...

... Initially, it must be highlighted that the materials with magnetic properties can be grouped into three categories: ferromagnetic (sensu lato) paramagnetic, and diamagnetic (Dunlop & Özdemir, 1997;Liu et al., 2012;McBride, 1986). Paramagnetic materials, when subjected to an external magnetic field, acquire an induced magnetization that is weak and parallel (positive) to the orientation of that field. ...

... Partly these data could represent a mixture of smaller PSD or even singledomain (SD) particles and larger MD grains. For magnetite, this corresponds to a physical grain size range of >0.05 mm (theoretical transition to SD for spherical particles (Dunlop and € Ozdemir, 1997)), but not bigger than a few mm (when the grain size exceeds ca. 20 mm), particles can be considered as truly MD (Lowrie, 2007). ...

... Hematite and goethite are "hard" (canted antiferromagnetic) to magnetise and demagnetise, and are therefore characterised by relatively high coercivities (Hc, Hcr) and saturation field (Hs), and by low magnetisation values (Ms, Mrs). In contrast, magnetite and maghemite are magnetically "soft" (ferrimagnetic) and both display high magnetisation values relative to the same hematite content and low coercivities and saturation field (e.g., Tarling, 1983, Ö zdemir andDunlop, 2014). Furthermore, hysteresis parameters also vary for the same mineral. ...

... Experimental tests of this relationship are generally positive (e.g. Dunlop &Özdemir 1993Dunlop &Özdemir , 2000Jackson & Worm 2001), though sometimes higher demagnetization temperatures have been observed (Dunlop 1983;Kent 1985;Kent & Miller 1987); these 256 T. Berndt et al. discrepancies are attributed to the response of larger pseudo-SD (PSD) and/or multidomain (MD) grains (Dunlop &Özdemir 2000), but more experimental evidence is needed. ...

... ∼20 min of zero-field heating to 125°C should be sufficient to remove VRM from a lunar sample held within SD kamacite grains, even if VRM was acquired over a period of several decades (Garrick-Bethell & Weiss, 2010). Thermal treatments may be less effective at removing VRM from multidomain grains due to unblocking tails (Dunlop & Ozdemir, 2000). Future research should explore to what extent thermal pre-treatments under different oxygen fugacity conditions (e.g., heating in vacuum, in air, in Ar atmosphere, in a controlled oxygen fugacity oven) thermochemically alter lunar magnetic carriers and how subsequent AF-based paleointensity determinations may be affected. ...

... By normalizing the values of ferrimagnetic susceptibility with the corresponding values of saturation magnetization to obtain xJa, (Fig. 2c), we can remove the effect of concentration variations with depth to see relative changes in the highsusceptibility grains of SP and MD sizes. Because MD grains are two-to-three times less susceptible than SP grains (Dunlop, Ozdemir & Enkin 1987;Hunt, Moskowitz & Banerjee 1995), and because the absolute magnitude of the Verwey transition is approximately equal in all layers (suggesting a roughly uniform coarse-grained content), we can, as a first approximation, attribute all the variations in xf/os to SP grains (Fig.2~). We will discuss this in more detail in the next subsection. ...

... This is consistent with the sharp indication of a Verwey transition at around 120 K in the low-temperature remanence curves of the bulk magnetic measurements 11 , which indicated that stoichiometric magnetite was present 28 . Electron energy-loss spectroscopy (EELS) analysis of the electronic states of iron and oxygen also showed typical spectra of magnetite, with no evidence of oxidation (Fig. 1I), which was also consistent with the low-temperature remanence curves of the bulk measurements 11,29 . The matrix surrounding magnetite was a layered silicate composed mainly of silicon and magnesium with a Mg:Fe:Si:Al ratio of 47.0:6.1:44.0:2.9 in the square region of Supplementary Fig. S4B; this is close to the ratio for saponite, which contains interlayer H 2 O in addition to structural OH sites. ...

... While the physical mechanisms of remanence have been somewhat clouded by this terminology, the basic idea of composite remanences of different ages and origins is sound. What has perhaps not been appreciated is that CRMs and VRMs produced at relatively modest temperatures have surprisingly high resistance to A F cleaning (Figs 3, 6, 9; see also Dunlop & Ozdemir 1990) and that their coercivity spectra are so similar that AF demagnetization is not a good way of resolving the composite remanence. Zero-field storage or thermal demagnetization are likely to be more successful. ...