MOKE is typically applied to thin film materials where this geometry is particularly important regarding the magnetic properties of the specimen. In most cases the films typically exhibit in-plane magnetic anisotropy but in some cases it is possible to have a material that exhibits perpendicular magnetic anisotropy, or even canted magnetic anisotropy. MOKE is very sensitive to the geometric configuration of the setup since there are several configurations that can be used to characterize the material. This is particularly useful to map-out the magnetic anisotropy characteristics of a sample. This can be very rich in the case of epitaxial thin films for example. Such detailed measurements may be very difficult using SQUID or VSM. In any case, a thick and polycrystalline film without rich magnetic anisotropy may lead to similar magnetometry results using MOKE and SQUID or VSM, provided of course that suitable MOKE spectroscopy has also been carried out to establish the optimal wavelength for the characterization of the material in question.

The answer is no. ´Contributions from the non-linear Kerr effect may change the magnetization loop considerably. Shining p-polarized light can in addition result in transversal effects, i.e. part of the signal is proportional to the magnetization component perpendicular to the reflection plane. Therefore a good starting point is to use s-polarized light, where the latter effect is avoided. Still non-linear effects result in a signal, which can be described by s=A*ml+B*(ml^2-mt^2)+C*ml*mt, with longitudinal and transversal magnetization components ml and mt. Constants A,B,C depend on the material and on the geometry. VSM always measures ml. So even if you assume that the magnetization is homogeneous in your sample or you integrate with MOKE over a representative volume, the MOKE signal and the VSM result are not necessarily the same.

So, it is important to also keep in mind that MOKE is a surface technique that probes a thin film up to the light penetration depth while SQUID or VSM probe the entire specimen. There are several articles in the literature where useful comparisons have been carried out for these techniques.

MOKE is more sensitive to surface effects and in fact, sometimes you can get a distinct hysteresis curve with MOKE while it need not be visible with a VSM measurement.

MOKE is a surface technique, based on the use of a polarized laser beam.

To have a good MOKE signal you need a smooth surface with specular reflection, a sufficiently intense laser beam, and an outstanding optical system that controls the polarization of the incident and reflected beams.

The MOKE information originates from a thin surface layer and cannot be always extrapolated to the bulk surface.

Typically, the MOKE signal as a function of external magnetic field can be correlated to the (VSM) hysteresis loop assuming that the polarization is very well controlled. In most cases, MOKE allows an acceptable estimation of the parameters associated with the VSM hysteresis loop. Shape analysis may not be that accurate and if detailed analysis is performed, higher order contributions are required...typically such terms includes deviations between MOKE and VSM hysteresis loops.

I agree with answers of other people that MOKE only detects surface magnetism.

In addition I should point out that MOKE hysteresis curve depends on domain patterns of the magnetic materials.

In multi-domain samples,despite the MOKE signal from each domain is proportional to M/Ms, total MOKE does not necessarily proportional to total magnetization M_total, since this argument ignores the diffraction effect by domains.

Consider a ferromagnetic sample with stripe domain with a period of p and a domain (+Ms) width of w. Then total magnetization is given as M_total=Ms(2w-p)/p.

Light is diffracted by the stripe domain leading to diffracted light beams of n=0,1,2...th order. Only the 0th order light shows theta_K polarization and higher order light show 90 degree polarization. Takihisa Numata calculated the signals for different detection systems (1) fixed analyzer non-differetial, (2) fixed analyzer differential, and (3) extinction method. His conclusion is that only (2) the differential method gives a magnetization-proportional signal and provides correct hysteresis curve.

Reference: Takuhisa Numata et al: J. Magn. Soc. Jpn. 14 (1990) 642 and Hiroyuki Tanahashi et al.: J. Magn. Soc. Jpn. 14 (1990) 648. (in Japanese)

A very sincere thanks to all of you for taking the time and answer my question. Also my sincere apology for not getting back to this (and RG as whole) for last several days. I am trying to take in all the information here but I think I have the crux of the answer, which is, in general a "no" to my question. However, given the depth of the knowledge in the audience I have here, I will add one more question to it. What happens to (MOKE signals from) thin (less than penetration depth of wavelength) weakly magnetic (paramagnetic) disordered systems? I am working on organic semiconducting systems that fits the description.

@Prof. Sato, Thanks for the references, is there any chance you could direct me to english version of those, (or any other paper). We are using a setup very similar to what you describe in 2 although the rationale of using that particular geometry was detection sensitivity. I am also working on MOKE of some new in-plane magnetized Garnets. Compared to the traditional perpendicular ones, these do show interesting strip domain scattering. I am still in early stages of comparing / measuring magnetization in these samples, but I definitely see a change in the hysteresis pattern.

@Dr. Piramanayagam, Thanks for the info and the papers, I am reading them. Interesting read! We are working on organic systems and are not as well controlled as yours; but may be related to some of the multilayer organic inorganic composites we are planning. Would you mind if I shoot a few questions once I finish reading your paper?

@Dr Gangopadhyay, I asked the authors of the paper cited in the previous answer whether they published English version of the papers. His answer is no. However, he suggested that the following papers may help you to consider the effect of domain diffractions:

T.Numata et. al., Nonlinearity of Faraday Loops, IEEE Trans. Magn.,vol.26, pp.1358-1390 (1990).

T.Numata, et. a., Improved Sensitivity in Novel Scheme of Magneto-Optical Field Sensor, IEEE Trans. Magn., vol.27, pp.5393-5395 (1991).

Regards, Katsuaki Sato

@Dr. Gangopadhyay "What happens to (MOKE signals from) thin (less than penetration depth of wavelength) weakly magnetic (paramagnetic) disordered systems?"

In that case, if the system consists, e.g., of a large number of separate magnetic nanoparticles of the same material, I would assume that the MOKE signal gives a result that is virtually identical to that of any other hysteresis measurement.

@Prof. Sato, Thanks for your help locating the references. I have previously read both these two articles. Both of them (along with another one on contrast reversal by the same authors) contributes fundamental information on measurements of domain scattering from garnet type materials. In last 20 years there has been a lot of improvement in commercially available auto-balanced differential detectors that, in favorable configuration, can produce up to 50+ dBm of noise cancellation (improving signal to noise ratio) taking your noise floor virtually close to shot-noise limit. That helps in lot better contrast ratio and (or) domain scattering (also FR) measurements. These detectors makes measurements lot easier as well!

Best regards, Palash

@Dr. Hertel, Thanks for chiming in. That was exactly my thought! But what actually happens is what Dr. Nambakkat mentioned earlier. At room temperature I do not see any VSM / SQUID signal, but I get strong MOKE and FR signal (no hysteresis though at either of them), VSM / SQUID signal becomes stronger at lower temperature and shows hysteresis below 50 K. MOKE picks up the hysteresis but is much weaker in strength. Standard samples are doing just fine. Problem is we know very little about the magnetic properties of the material (an organic semiconductor) in question.

Regards, Palash

attached is the graph of Moke for Nickel (Bulk). I have been trying to get the moke signal for over a year but the result is same. Could anyone please help me or point out what could be the problem.

Ali, how does your phase (of Y) look like? 

sorry for being late. Phase looks fine Palash. I have gone through with Moke bench-marking sample and the result was good. But nickel bulk does not give hysteresis still. I am still trying to work this out.

One thing more palash if you could help me with, how we can change moke signal(Lockin-Volt) to roatation(degree or radian).

Warmest Regards