Generally we use in our lab DLS measurements to obtain a first idea of the particle size, because as you will know DLS measurements give you the hydrodynamic diameter.Also, DLS is only capable of giving accurate results if the particles are spherical. If this is not the case, than you will get false results (there exists a nice introduction on DLS from Malvern: 'DLS, an introduction in 30 minutes'). So, you normally will always obtain a bigger size in DLS than according to TEM/SEM.

I also find that different manipulations of you sample can contribute to different results in DLS. Like for instance, when you first filter you solution, you will get more likely a higher particle size (we think this has got to do with the pressure you performe on your solution). So make sure that everytime when you measure DLS, always do it in the same way.

Another thing that is very important is the viscosity of your sample. Do you measure it for each sample? Because a small change in viscosity will lead to a large change in particle size (if you always use a 'standard' viscosity during your measurement, but if the viscosity of your solution changes, than this can lead to a difference in the measured value).

Because you already have your FESEM results and you know the actual particle size, you just can mention in your article that you first used DLS to already get a first idea of how big your particles are (knowing that you measure the hydrodynamic diameter) before further processing and than you performed FESEM to obtain the actual particle size.

I hope I did help you.

I imagine if you are seeing changes in the size as measured by DLS over time that the measured size is increasing, which is a consequence of your nanoparticles aggregating in solution. DLS will only give the "true" size of your particles if they are "perfectly" dispersed, which is of course very difficult to obtain for most materials. It may be worth trying to dilute your samples and dispersing them in an attempt to slow aggregation, though it's unavoidable. Noting this aggregation can still be important for publication.

In order to measure a truly individual particle size, something more like TEM or SEM may be necessary.

It is a common phenomenon of particles aggregation wth time...The size must be measured immediately after preparation. This should be kept in mind that the dilution of sample needs to be optimised priorly, i.e. you should make several dilutions of your sample and see the size. the range of dilutions where you get a plateau, should be always used for particles analysis....If your particles are increasing continuously with time, then you should provide the trend of increase in particle size with time. To keep your particles stable for longer time, you need to use stablizer, the choice of stabilizer depends on the type of polymer and technique of preparation you are using...good luck

Thank you....saeed khan...i am getting z-avg almost same with standard deviation of 10 to 15% but the peak which coming is different........u r mail id u give me i will send my data with details.........my sir saying that every time i have to get similar peaks.......is it possible for nanoparticles without stabilizer.

My email id is [email protected]

the shape of the peak is the indication of homogenity of your particles (i.e. polydispersity index)....you should keep in mind that particle size is not the only parameter which you can use for particle characterization. I mean you should also know the PDI of your dispersion (this value is responsible for the peak shape)...the broader the peak the higher the PDI value,and vice versa. I am gonna send my ID in private message...

I think that a morphological characterization could be useful to detect the mechanism of size variation, then you could improve your experiments.

For correct measurement of your particle size, SEM and TEM will be useful.

If the particle size is in nano range .... choose TEM.

If it is well in the micro range ... choose SEM.

hw i have to study morphological characterization........R. Milazzo.

one more small quation is in the publicatiions few reporting DLS peaks of nanoparticles........hw much accurately they got size distribution of particles size if experiments repeated 3 to 4 times is it possible to get similar size distribution if nanoparticles are prepared without any stabilizer.....

Soumitro Mahanty we already done FESEM of samples and i got particles size below 100nm but if the exp. repeated 3, 4 times we will geting different size distribution peaks in DLS but same size in FESEM, one more Q is our samples are organic compounds then is it possible to do TEM.?

If you have already seen your sample in FESEM.

Then I think you can see it by TEM.

only precaution you have to take is ...... use lower kV operating voltage.

While seeing particle size .... do not confused with agglomeration (clusters of particles).

you can get all kind of information of your particle.

in FESEM i got particles size below 100nm but my sir want to put DLS data in publication........wat i do....

If your FESEM data and DLS data there is a little bit variation then report both of these with repeating no. of times your experiment and get standard deviation out of these ... this may be acceptable....

Generally we use in our lab DLS measurements to obtain a first idea of the particle size, because as you will know DLS measurements give you the hydrodynamic diameter.Also, DLS is only capable of giving accurate results if the particles are spherical. If this is not the case, than you will get false results (there exists a nice introduction on DLS from Malvern: 'DLS, an introduction in 30 minutes'). So, you normally will always obtain a bigger size in DLS than according to TEM/SEM.

I also find that different manipulations of you sample can contribute to different results in DLS. Like for instance, when you first filter you solution, you will get more likely a higher particle size (we think this has got to do with the pressure you performe on your solution). So make sure that everytime when you measure DLS, always do it in the same way.

Another thing that is very important is the viscosity of your sample. Do you measure it for each sample? Because a small change in viscosity will lead to a large change in particle size (if you always use a 'standard' viscosity during your measurement, but if the viscosity of your solution changes, than this can lead to a difference in the measured value).

Because you already have your FESEM results and you know the actual particle size, you just can mention in your article that you first used DLS to already get a first idea of how big your particles are (knowing that you measure the hydrodynamic diameter) before further processing and than you performed FESEM to obtain the actual particle size.

I hope I did help you.

yes exactly what u said thats only happening with my results........i have seen many papers they just mention size and 1 2 DLS distribution grphs and they kept FESEM images thats it......i too told with my sir to go forward with FESEM images but he can't......i think to get same DLS atleast we have to use stabilizer ........but in our case even we can't using any stabilizer.............becouse of only DLS results my work in in pending........i can't able to publish and go forward,...............

thank you Katrien De Keukeleere

Thank you to all

I agree with Katrien. If you are not satisfied with DLS data then you may try the following too. First, you can measure you particle diameter (irrespective of shape) by particle analyzer machine. Secondly, as Katrien said you can go for FESEM imaging. Thirdly, if your particle sizes are less than 50 nm, you have to go for HRTEM imaging for better and exact result. Fourthly, if you have crystalline samples , then by measuring XRD spectra and using Scherrer's equation you can get and idea of the size of your particles.

you can do it many time and find average value to get better DLS result..

What we do in our lab that we measure diameter of particle from DLS as well as we use X ray diffractometer and obtain diameter from scherer formula than we compare both and take standard deviation but we run DLS for sample more than three time. sample for DLS must transparent

for obtaining good results

Dear Raj Kumar,

I have seen a review paper for nanoparticles. I think it will serve you purpose.

Thanks.

I work with HRTEM for Iron Oxide and Silica coated Iron Oxide NPs, the size around 20-50nm is ok. First we can use FESEM and DLS for the imagination of particles size. then HRTEM.

Sorry to intrude

I am from the nanoparticles in air, an aerosol-physicist; for us nanoparticles are known as "ultrafines".

We measure nanoparticles with a size resolution of 1%, down to a size of 3 nm.

We use suspensions with monodisperse nanoparticles for calibration.

This method was also used by us to measure the size (distribution) of nanopaprticles prepared by colleagues who wanted a higher resolution than DLS.

The way this is done: the suspensions are nebulized into micron-sized droplets which are then dried in an air-flow. In this way we have the pure particles.

The particle sizer used is the so-called Differential Mobility Particle Sizer (DMPS); the manual version should be used for optimum szie-resolution.

Maybe you have aerosol-scientists as colleagues in your department who can do this. For them it might be the standard procedure for calibration.

N.B. I have seen publications with this method in the journal "Particle Characterisation" in the early nineties.

Actually I found a rrecent eference with up-to-date info on the method I give above

thank Harry ten Brink

I am agree with Katrien and Kumal, FE-SEM or TEM analysis will give to you crutial information about shape factor, grain size and real size distribution, but don't discard extra-information that give to you DLS changes. Stability from nanoparticles in your media will be important depending of your final use. As extra-information, don't discard z-potential measurements.

thank u mr.Dorothée Vinga Szabó

I am currently using a device, called Izon-qNano, to characterize nanoparticles for a biomedical use. This is a relativevly new technology that can measure each individual particle/event at a time, resulting in the ability to obtain data with a histogram representation of measured particle size, rather than that of Gaussian-distribution obtained in DLS.

http://www.izon.com/capabilities/particle-sizing/

please read my papers concerning nanoparticles that we measured by using TEM. The histogram representing the particle size distribution can also be obtained.

Mr Sahar

If you wan tus to read them you have to give the link to WHICH of your publications at Researchgate you mean. There are many on Raman-measurements. You cannot expect that we sort out where you did TEM

mr Sahar

and by the way, you do not have the full-text of your publications on researchgate, so we cannot read them anyhow?!

TEM is the most reliable way to determine the nanoparticle diameter since you will measure it directly from the micrograph. While DLS does give the hydrodynamic radius, if your solvent is not water though you I have seen students get wildly varying values for your measurement. Depending on the material that you are making, it may be possible to determine the size via UV-Vis absorption spectroscopy (this works pretty well for some of the direct band gap semi-conductor materials).

The size of nanoparticles can also be determined by measuring the

photoluminescence spectra. In this case, you can get a histogram of their size. This type of research we have conducted for nanoclusters of layered structure in the paper, which is presented below.

Yuriy

I cannot open your file: can you please indicate where I can find this publication in your list of contributions at your site here in Researchgate, please.

for more clarity and understanding on the use of DLS you can go through this material

Just to add: if your particle size is more than 10 nanometer, SEM can be used as well.

For conformation you must go through SEM / TEM. DLS principle works on Brownian motion of particles and it might be the reason for getting different particle size at different time interval.

Harry,

This file can be opened by Adobe Acrobat 7.0. The paper is in my profile (contribution) in RG.

Harry,

this paper is here :

https://www.researchgate.net/publication/234072899_Time-resolved_photoluminescence_spectroscopy_of_excitons_in_layeredsemiconductor_PbI2_nanoclusters?ev=prf_pub

Article Time-resolved photoluminescence spectroscopy of excitons in ...

TEM is best way to measure the particle size. When you do the synthesis of nanoprticle it do happen that you will get multi dispersed particles. Just try to synthesis the another new particles or you can take the particle size which is more common .

Yes, i do agree that TEM is the best method. However, dealing with the nanoparticle with the size of about 20 nm, I have no problem at all using FESEM. And for particle size distribution analysis, what about using AFM?

If your particules solve in water clearly. you can use particule size analyzer. Your result which you found it almost equal the XRD results.

If your nanoparticles are electrondense enough use the TEM, it will be more accurate for sizes bigger than 1-2 nm, then you can analize and treat the picture in order to measure the size and make a size distribution and an average. The problem is the possible aggregation after some time, before the preparation of the TEM sample you can sonicate the sample a little bit to avoid this problem.

The problem with the DLS is that you have to have spherical-shape particles to get an accurate real size, and you have to take into account other parameters like diffraction index and so on, and maybe you don't know this data.

XRD will give you the crystalline size, but maybe it's not the real size of your nanoparticles. In the case of Quantum Dots you cannot use this technique, the crystallite size is too small.

If yuour nanoparticles have quantum confinement you can also use the absroption to get your nanoparticles size, because the extinction coeficient will change for every wavelenth depending on the size of your nanoparticles.

@ Suman : The size of nanoparticles in semiconductor materials can be determined if a quantum size effect takes place, i.e, when the size of the nanoparticles is near or less than the size of the exciton in the bulk material. In this case the energy of the optical transition between the upper valence band and lower conduction band is larger than in the bulk material. The value of this energy will increase with decreasing size of nanoparticles.

In the photoluminescence spectra of semiconductor materials usually appears the exciton emission bands. In the case of the manifestation of quantum size effect the blue shift of the exciton band should be observed, reflecting the increase in band gap of nanoparticles. If all nanoparticles are the same size you will observe only one exciton band shifted to relative the exciton energy in bulk material. There are some analytical relations to determine the size of the nanoparticles, using the values of the excitonic bands energies observed in the photoluminescence spectra of semiconductor nanoparticles. If you are dealing with nanoparticles of different sizes, a number of exciton emission bands with different energy positions depending on the size of nanoparticles is observed in the photoluminescence spectrum. Therefore, in this case, you can define the different sizes of nanoparticles that appear in the photoluminescence spectrum. The intensity of individual exciton bands will reflect the contribution of different nanoparticles: the greater the intensity of a particular exciton photoluminescence band, the number of the nanoparticles of this size will be greater in the investigated material. Thus, it is possible to construct a histogram of the distribution of nanoparticles in size.

The paper, which I presented above, is an example where the determination of nanoparticles with layered structure based on research excitonic photoluminescence is described. The relation that is used here, is suitable only for highly anisotropic nanoparticles, particularly for materials with layered structure. I see that you are researching the semiconductor CdS nanoparticles. In this case, it is necessary to use another relation obtained for nanoparticles of sphere form. If you're interested, I can send you a paper where this ratio is presented.

to Bilge

You write 2 lines. You do not give any details on the method? What do you want us to know?

apart from previous suggestions, we successfully use (also as an easy routine method) Laser Doppler spectroscopy.

Bernhard

This is already what mr Kumar does, see the text that he wrote under the question

oh, I did not recognize that "DLS" was meant to be "Laser Doppler" ( never used this abbreviation ...). Then I would assume his measurements may not appropriately be done:

- viscosity not well accounted for

- absorption too strong

- concentration too high / too low

- temperature not well controlled

- the colloidal dispersion he is using may not be stable

Bernhard

Apologies

I was maybe too quick DLS indeed is dynamic light scattering and meant for submicron particles

Laser Doppler is for flow and particles over 1 micron, but you might correct me when I am wrong

thanks, Harry, (I even never used "DLS" for "dynamic light scattering" ...). anyway, my comment above is still valied when using Laser Doppler spectroscopy, one has to consider those aspects I mentioned above.

thank you Harry & Bernhard.......for valuable information,

Harry, Laser Doppler is suitable down to a bout 3 nm, I forgot to comment on that part of your recent comment; we have used this since 15 and more years.

Several DLS, SEM and AFM comparisons are in attached publication.

Article Polymerization Model for Hydrogen Peroxide Initiated Synthes...

Raj! 1st time I'm hearing of IIT Mandi. (I'm ANCIENT?) Where is it? Please post a link. I'm still in the process of reading all the prior comments.

What is the typical size range you are interested in monitoring?

Have you considered XRD? Specifically SAXS? (Small Angle X-ray Scattering)

What is the composition of your Nano particles?

With the use of SAXS you should be able to make the solution and watch all the postulations prior in real time. Tell/sell your "Sir" that he needs to consider the advantages of SAXS open-mindedly! You need to "sell" it to him. Tell/sell him to check this link and email me if he needs further "selling or telling". These methods have been tried and tested for over a century now and really do not need "selling or telling" for the perceptive minds. Tell him to stop "groping in the dark" and "get with it"! Just jesting!

This Panalytical webinar may be helpful: "X-ray Diffraction Techniques in Transmission Geometry" - http://www.linkedin.com/groupItem?view=&gid=2683600&type=member&item=257639037&qid=6338245e-dda1-4e21-966d-ca1fc369920e&trk=group_most_recent_rich-0-b-ttl&goback=%2Egmr_2683600%2Egde_2683600_member_257395175%2Egmr_2683600

Note: XRD gives "diffracting domain" size in a poly-crystalline material. We've extracted dislocation density information as well. Please check the works of Sigmund Weissmann et. al. of Rutgers Univ.

[email protected]

Certainly come join us at the " X-ray Diffraction Imaging for Materials Microstructural QC" LinkedIn group and share in the collective wisdom of our membership. We could always use the additional embellishment: http://www.linkedin.com/groupItem?view=&gid=2683600&type=member&item=257395175&qid=3a221713-5808-4ca7-857a-04c39c191f1c&trk=group_most_recent_rich-0-b-ttl&goback=%2Egde_2683600_member_238135655%2Egmr_2683600

Www.iitmandi.ac.in

Beautiful! What a wonderful location!

Another possible technique to achieve morphological information about magnetic nanoparticles in solution (suitable for ferrofluids or biomedical applications as drug delivery or magnetic hyperthermia) is the measure of the complex susceptibility as a function of the frequency of the alternating magnetic field and the fit of the data in the framework of the Debye model (Nèel and Brownian relaxation).

This technique allows to obtain both the size distribution function of functionalised magnetic nanoparticles in solution and the size distribution function of the core of these nanoparticles. See for example the paper by Roseweig on JMMM 2002 or my paper on J Nanopart. Res of 2013

All of us are guilty of neglecting the original question at this stage of the thread.

Mr Kumar asked a very specific question with respect to his problem of measurements with DLS in his system and asked which result ]he should publish. The answer is best contained in the reply by Katrien de Keukeleere IMO.

Dear Suman,

You asked me present more detailed information regarding determination of size nanoparticles by measuring photoluminescence spectra. I did that. Do this information helped you ?

Best regards,

Yuriy

The general answer to the question "Which one should I consider for publication" is, the one with the best precision. How repeatable are your results? Clearly that is the dominant factor.

If the results are changing with dwell time then that has to be quantified and a root cause determined. It seems that there are some kinetic changes occurring after the creation of the solution. From your discussions it is likely that the size is changing due to agglomeration and other mechanisms as a function of time. This issue would have to be resolved and solution stabilized before you can publish the result. However, if you quantify your results precisely and characterize the time dependent changes, then you may be able to understand the kinetics involved. You may consider publishing those results.

I would refrain from publishing results that have uncertainty and ambivalence.

Depends on how much they vary in function of time. Is it some nm or much more? DLS is not at the edge of precision... How long do you integrate? Try to measure much longer and repeat a lot. Use the spectral data of your nanoparticle, too. Try to dilute the sample or/and check for possible precipitation. DLS is not as trivial as people usually think...

thank you to all for valuable suggestions...the new quation arising in mind after this discussion is

if we get accurate particles size and shape from FESEM and STEM then what is the important to include DLS data in publication ?

one more Q is from FESEM images if we calculate particles size by using Image J software......it can give the avarage size of the particles can i able to put in publication. is it good method ?

prasently i decided that to put selective data of DLS which can suitable with FESEM and STEM results.

FESEM and STEM size are highly unrepresentative, because you scan a small amount of NP. It start to make a bit kind of sense with 1000 NPs but you are still somehow biased. Try small-angle X ray scattering as a complementary method...

Generally, DLS is a relatively accurate method to determine the average hydrodynamic diameter as well as the polydispersity of spherical nanoparticles (NPs). To directly observe the morphology of NPs, there are also a couple of methods, including SEM, TEM, AFM, etc. Peticularly, cryo-TEM rises as a powerful technology in structure analysis (includes NP shape) of samples in their native environment and the results are relatively believable.

Excellent point Alex! What is the typical beam size or sampling size in SAXS measurements? How many NP would we typically be averaging over? Billions? Trillions?

Why wouldn't AFM be the choice tool for this application? Availability & cost?

Jun Yue: 3D information from TEM without complex sample holder and without tomography??

For SAXS, the beam size is around 2,5 cm x 1 mm x 1 mm (concentration 0,1 mg/mL Ag) , meaning that for nanoparticle with average size of 20 nm you actually average around 10^14 np for silver... This is representative!

AFM is not precise for measuring lateral resolution, only height. And it would be very very tedious...

Got it for AFM! No automation yet? Thanx Alex!

SAXS - 2.5cm? Which dimension? I'd have thought that beam ("point source") cross-section would have been a lot smaller than 1mm x 1mm. Unless, you mean a "line source".

Right! Classically if you are not interested in 2D experiments (what is NOT always the case) you have "line source" (ie "Kratky geometry"), and these are pretty interesting because you scan very large areas of your sample. The price you have to pay is the deconvolution of your data afterwards, which is yet well-known.

Thank you @ Ravi Ananth and Alexandre Mantion

There should be no big different in Z-average if there is:

Decrease in size possible reasons:

*Temperature not stable (make sure the readings are done at the same temperature)

*Break up of your sample

Increase in size possible reasons:

*Temperature not stable (make sure the readings are done at the same temperature)

*Aggregation of your sample (try to stabilize your sample by better surfactants steric stabilization or ionic “double layer”)

Or you can take at least the average of your readings

Or characterize its size by SEM too to compare

Thanks to All of great minds for very interesting informations

DLS gives the particle size which is average . Calculate the deviation of series of measurements and if it is in acceptable limit the size measured is acceptable. The variation is attributed to the viscosity of the medium. Not to worry if the deviation of measurements  is small.

Just to add: The particle size can be also determined -by theoretical calculation  - from  absorption wavelength of the UV-VIS spectra by using hyperbolic band model approximation Or by using Brus equation …ect