At first look, data shows that your sample shows paramagnetic behaviour with very weak ferromagnetism.
Also please check that the shift in Y axis is really coming from your sample behaviour or its just due to the measurement error, because this thing really defines some interesting physics.
Refer to the following paper for more details...........
You need to normalize magnetization value. You may divide this by mass of the sample to obtain mass magnetization (emu/g). Also as Prof. Vinay Sharma mentioned you need to repeat this measurement to ensure results, specially asymmetric MH behaviour . Please also refer the attached paper
Dear Nurul Fadzilah, you need to correct your data by to way
1- re run the sample and add the wight value. So, the magnetization at y axis will be (emu/g) as Akther recommended, then symmetric your loop , to see wither the loop set exactly on y axis or shifted.
2- Or, you can do the correction of your data manually using sigma plot or Excel.
dear, the loop indicated that the sample is not saturated and the grans are very small
As Vinay Sharma states, your loops appears to a weak ferro-/ferrimagentic signal with a fairly dominant paramagnetic one.
However, the data are too noisy to do a quantitative analysis.There is clear evidence of moment noise, but also noticeable drift (the divergence of the loop top at negative high fields) and the large moment offset, if not related to a physical factor, may also be drift related.
There are many things that can be done to improve the data, but the first step is to remeasure the loop. Here are some suggestions. If you give us more information about you machine, we can give you some more specific advice.
1) Most VSMs have different forms of field sweep. Typically called "continuous" or "discrete" modes. Continuous mode sweeps the field while performing the measurements and works very well for strong specimens and it is fast. Discrete mode will hold the field for each measurement point. It will typically reach the desired field, wait for a short period of time to stabilize (say 300 to 500 ms) then perform the measurement. This is much slower, but gives much better data for weak specimens.
2) What averaging time are you using for the measurements? Increasing it will give better data, but may result in higher drift.
3) You can average multiple loops. This is a counterbalance to longer averaging times. Many VSMs allow you to do this automatically, whereby it will measure and average several loops for you. Otherwise you can do it manually.
4) Drift can be a problem that is difficult to quantify and properly address. Thermal drift can occur if the specimen gradually warms or cools during measurement. This can happen as the magnets warm or if the room gets hot (if the specimen region is exposed). Make sure that the magnetics are adequately cooled and that the room temperature is held fairly constant.
5) Take great care is centering your specimen in the measurement region and ensuring the shape of the specimen is as symmetric as possible.
I hope all this helps, but let us know if you have any other questions.
Suppose the measurements were taken correctly, including sample centering. Then this is a very interesting result. However, its interpretation should be based on other information: is the material uniform or in grains? If in grains then how large are they? Do they have core-shell structure? What was the earlier magnetic history of this sample? And so on and on. Anyway, at the first sight we deal with non-uniform material consisting in part of hard ferromagnet. On the other hand, signal to noise ratio strongly suggests that the sample is rather small (thin layer?), hard ferromagnet deposited either on paramagnetic substrate or its signal is combined with sample holder signal. One cannot exclude antiferromagnetic component of the material, too. I can also imagine that this result comes from single crystal of weak ("canted") ferromagnet, preferably in mono-domain state.