Hi,

I would say that the kink you see is an effect of the strong magnetic interactions present in the clusters of MNPs, but as you point out there might be other explanations. Were the measurements done with the samples in powder form or in a liquid suspension? I suggest repeating the ZFC-FC procedure under a lower DC field such as 10 Oe, so that the effect of the field on the energy barrier can be neglected. Also TEM or SEM images of your nanospheres could also help you to find what is going on there.

I hope you find this useful,

Irene

In addition with Dr. Irene; TEM should reveal the distribution of particles and they can be in different sizes. Such magnetic behaviors may result from two sublattices. It it important that your sample has single phase and does not have impurity. If some structural analysis like XRD shows you a second phase, this may trigger the formation of magnetic moments at a different temperature.

Thank you Irene and Murat for your answers.

@Irene yeah I did too think that this might be due to the interparticle interaction as a large number of smaller nanoparticles are self assembled to form the nanospheres. I will attach a TEM micrograph below.Moreover I have performed the experiments on powder samples. 

@Murat I have done the X-ray diffraction studies of the sample but a lot of noise is coming. I have got the signature peaks of Magnetite. Also it is very difficult to tell from XRD if the sample is pure magnetite or hematite is also present there. 

Does your saturation magnetization roughly correspond to that of magnetite or maghemite after substracting the diamagnetic contribution of the coating? Hematite has a very low saturation magnetization at room temperature, so if there is a lot of hematite present in your sample, or your particles have very low crystallinity, you should get values of saturation magnetization way below 60 emu/g. What we do in our lab to accurately determine the saturation magnetization when an unknown amount of surfactant is present is measuring the M(H) curve of a sacrificial sample of known total mass (magnetic particles + surfactant) that we digest in aqua regia afterwards, and determine the Fe content by ICP. In this way you can quantify the amount of iron present in your sample, if you know the stoichiometry of your compound you can calculate the amount of magnetic nanoparticles vs the mass of surfactants. You can find the protocols here:

http://pubs.rsc.org/en/content/articlelanding/ra/2014/c4ra02586e#!divAbstract (In this case the detected ion was Co)

http://www.sciencedirect.com/science/article/pii/S030488531401035X

Keep up with the good work!

Well the saturation magnetization of the magnetic nanoparticles without subtracting the diamagnetic contribution of the coating is coming around 80 emu/g which is near to the saturation magnetization of magnetite. However, as mentioned by you I will try to calculate the saturation magnetization after subtracting the diamagnetic coating contribution. 

But, the mystery of this kink is still unknown to me.

Hi Kakoli, I will start with a disclaimer -- I am not an expert in this; but the discussion seems to be still open, I will add my two cents.

• Your system is highly interacting -- so dipolar interaction will dominate whatever you measure.  Have you tried varying the sample quantity? 
• Whereas 1000 Oe and 500 Oe are very reasonable applied field, you probably should check with much less applied field -- down to 10 - 100 Oe. Does the kink disappear and difference between FC and ZFC increases at higher temperature?
• How does hysteresis curve look at zero field? -- at room temperature and at the "kink" temperature? Does it show a similar kink under demagnetization condition at either of the temperatures?

Whereas I am not solving your problem, I still think the kink is manifestation of strong dipolar interaction near around blocking temperature of the individual smaller particles.  Please correct me if I am wrong.

Cheers,

Palash

Dear Kakoli,

which colour is your sample? one possibility is you are dealing with an hydroxide. For instance, you mentioned a "noisy" XRD spectra. Does it mean you have a "bump" in XRD at low angles?

Though, following Ockham's principle the simpler hypotheses are more likely to be true. I mean here your "kink" is probably due to the Verwey transition. Please note it starts about 125K (I suggest you to calculate the first derivative dM/dH to check this). Of course, additional remanence vs temperature dependence and/or Mossbauer studies (just to mention a few additional methods, for instance, titatrion chemical analysis, XAS-XMCD synchrotron and resistance vs T characterization) would -or not- give support to this statement.    

Now, please be careful when measuring zfc-fc, specially if your sample is in the blocked state at RT (as concluded from your figure. Note the collapsing or not of the curves at different fields). Increasing the initial temperature above 300K may help. 

hi,

check 'kink' with diffrent fields  from  20 Oe to 2000 Oe. as field increases find out in which way the kink shifts (if that was really shifting).  with every applied field take the system to above transition and then cool ,apply field. if this was really kink (plot dM/dT) was shifting with field then check  ur compound with relaxation phenomena in magnetism suitable to nano particles and their interaction with different times and fields (A.C suasceptability) .check the broadening of kink with field. it was just my guess it may r may not work.  

Dear Kakoli Bhattacharya,

I think that this might be indeed related to the Verwey transition. We looked into this question in our recent paper (D. Schmitz et al., Sci. Rep. 4, 5760 (2014), also attached). Just have a look. I am happy to answer questions to this publication.

All the best,

Heiko Wende

Article The dipole moment of the spin density as a local indicator f...

Thank you all for your replies.

@Palash Gangopadhyay. Thank you for your suggestions. Actually in my system the nanoparticles are assembled to form spherical clusters which mean that the nanoparticles are in close proximity and hence the dipolar interaction is dominant. However, I have tried to modify the sample quantity but it has little effect as the nanoparticles in the self assembled structures are densely packed. Also, I have performed the M-T at low field i.e. 5,10 and 100 Oe. I am attaching the curves with this answer. At room temperature, we get a ferromagnetic behavior from the hysteresis loop however the coercivity value is very small i.e. around 10 Oe.  I have not done the hysteresis study at the kink temperature which I will have to do in order to get a clear picture. 

@Carlos Martínez-Boubeta and Heiko Wende. Thank you for your suggestions. I was also thinking that the kink might appear due to the Verwey transition but I have found in literature that the Verwey transition happens at around 120 K. But in my case the kink is appearing at around 140K and it has not shifted its position even when the field has been varied. So, is it possible that the Verwey transition temperature might shift to higher temperature ? 

You are right, 140 K is too high for the Verwey transition. There is also a spin reorientation transition at 132 K. But both transitions will not shift to higher temperatures (maybe to lower temperatures for the nanspheres because of finite-size effects). Therefore the kink is probably due to something else.

hi,

first make sure that dynacool working properly r not.  use any single crystal or any other sample which u know exact transition.

second check: if above was o.k. then use two more options in ur dynacool (i.e Heat capacity and electronic transport) to check the transition.

third check: if the transitions are also same then try to increase the nano size and then see. of course most of times magnetic transition will not change with synthesis but magnetic is associated with electronic or structural transition then i cannot think it may be the same.

i think it will take some time but if ur interested u can try this. all three are my guess. if u think these are resonable then follow it if not leave it.

Thank you @Venkat Venki for your answer. I will try these steps mentioned by you.

I have read the answers on this question, and the possibility of phase separation in the sample can be checked by a local effect such as Mossbauer spectroscopy as mentioned previously. Su-bspectra with hyperfine fields characteristic of the different phases (and different sites in each phase) will determine whether the sample is single-phase (magnetite) or multi-phase. Also, careful analysis of XRD pattern should give information on the presence of more than one phase. However, I would like to make a comment on possible explanation of the observed data if the sample is single-phase (purely magnetite):

Small magnetite particles with typical diameters ~ 10 nm or smaller exhibit superparamgnetism at room temperature. Superparamagnetic particles have characteristic blocking temperature which depends on the characteristic diameter of the particles. So, while large particles are blocked at room temperature, exhibiting behavior characteristic of ferromagnetic material (with non-vanishing coercivity), small particles are superparamagnetic with zero coercivity at room temperature. So, look at the hysteresis loop at room temperature to verify that you have superparamgnetic system. Now if the system is superparamagnetic at room temperature, its blocking temperature is then lower, and depends on the particle volume as I mentioned. The blocking temperature can be calculated for a given particle volume and saturation magnetization using the standard Stoner-Wohlfarth model for a random assembly of single-domain spherical particels, and this temperature can be compared with the position of the kink, since a possible explanation of the kink is that below the blocking temperature the moments of the individual particles are blocked along easy directions, and magnetization processes are through magnetization rotation against magnetic anisotropy, which requires (typically) high applied fields. as the temperature increases, some of the blocked particles become un-blocked by thermal agitation, which aids the applied field in aligning the moments and result in an increase in magnetization with increasing temperature. The broadening of the transition region (from bocked to un-blocked) is determined by the width of the size distribution of the particles; a very narrow particle size distribution should then give a step-like transition (kink).

It would be nice to explore these possibilities for explaining your data