According to me better to consider number distribution. Because in number distribution the impression of unwanted dust particle will be less.

You should measure Zeta potential too in both the solvent in order to have better understanding.

Dear Metwally Ezzat :

 We are now preparing an article on the topic of your question. It will feature experiments SAXS and DLS of micellar solutions of sodium dodecyl sulfate in water. You will be hard to explain their data without our work .” Liquid Polyamorphous Transition and Self-Organization in Aqueous Solutions of Ionic Surfactants.” Yuriy A Mirgorod · Tatiana A Dolenko. See my profile. We could cooperate with you, if you are interested in it. My e-mail on our website http://www.nano.kursk.ru

Best regards.

It really depends on what you are interested in.

the reason for the different values is due to statistics really

Large particles scatter much more light than the equivalent mass of small particles, the intensity distribution is therefore weighted toward the large particles.

Likewise the volume or mass average is going to be larger than the number average Iin the same way as number average mol wt of polymers is always less than the weight average).

Thus as you can see from your data the number average (which is the average size of all the particles divided by the number of particles, is going to produce the smallest size, the weight or volume(its the same thing) will be larger than the number average but smaller than the intensity average which will give the highest average.

The intensity average is what the instrument measures directly and from this it calculates thru software the volume and number average.

So if you are interested in the average size  from a number viewpoint use the number average, if from a weight viewpoint use the weight or volume average. 

The intensity average is not so useful but itr is what the instrument actually measures

Hope this helps

Paul

Hi Metwally,

As previously stated, DLS directly measures an intensity weighted distribution, which is sensitive to large particles given that scattering intensity is proportional to the 6th power of the particle radius. For this reason, some researchers prefer to use this results as it doesn't rely on any conversion or assumptions. 

That said, intensity peaks for large particles can be misleading and therefore the number distribution may be preferred.

The topic is discussed further here:

http://www.materials-talks.com/blog/2014/01/23/intensity-volume-number-which-size-is-correct/

In terms of looking at your sample in different dispersants, it may be useful to measure the zeta potential of your sample as this can inform what the stability of your sample is in each dispersant. It may also be possible to infer some stability information by looking at the shape of the correlation function measured by DLS and noting how reproducible this is. 

Dear Sir,

I have a follow-up question related to this one. How effective is DLS measurement if we want to calculate the Emulsion droplet sizes in an oil-water emulsion. 

You can indeed measure emulsions using DLS, although some care may be needed to optimise the concentration of the emulsion to be able to obtain reliable results (at high concentrations, multiple scattering or internal interactions may occur, leading to inaccurate size results). 

Measurement of size and zeta over a series of dilutions should be able to reveal the optimum emulsion and sufficient concentration for stability.

The following webinar may be of interest

http://www.malvern.com/en/support/events-and-training/webinars/W091123StudyingEmulsionStability.aspx

Re emulsions, it would depend a bit ob the size DLS works by measuring the diffusion of particles so if the particles are large they do not diffuse much and so the technique is not to accurate. I would use DLS if the particles are less than a micron and static (Fraunhofer) light scattering if they are more than 5 microns, in between either would be OK but you are on the upper limit of DLS and lower limit of static light scattering.

re zeta potential measurements everything of course only holds if you use an ionic dispersant

Dear All.

Thank you very much for the useful answers and very interesting discussions

As soon as you try to characterize any polydisperse sample, you encounter your problem. Sometimes you don't see it if you use only one method.

In order to choose the proper answer you have to ask your question:

If you want to know HOW MANY particles of given size you have in your sample - go for number distribution

If you want to know what amount of your material (e.g. SDS molecules) is in particles (micelles) of given size - you go for volume (mass) distribution.

A simple example: you have two kinds of micelles (perfectly monodisperse) 5nm and 10nm size. You make a mixture such that the you have equal numbers of each kind of micelles. If you ask now what is the volume ratio the two micelle species, you will get a ratio of 8, because volume goes as size^3. So almost 90% of the solute is in the form of big micelles. What is the mean size of the micelles in your mixture then? The number average is 7.5nm and the volume average is 9.44nm. Both are correct and describe the same system.  You decide which to choose by asking a question that fits your application. 

Concerning the mean value provided by DLS (intensity average), it is even more biased towards higher molecular volumes. Although in principle it is possible to calculate the volume and number distributions from DLS data, it works properly only for relatively narrow distributions and identical shape of the particles. In complex mixtures you have to make many assumptions, so  the number distribution is usually very questionable.

Technical advice: purify your SDS as much as possible and try to obtain a monodisperse sample. Remember also that interactions in pure water will make the apparent size concentration dependent. Only the value extrapolated to zero concentration represents the real size of the micelle. It can be seen in your data - the smallest component in water is only 6.8nm in diameter which is definitely too small for an SDS micelle.

you did not revealed the nature of what you measure. Beside micelles it could be anything if you do not have a well defined micellar solution. Would be odd to have in equilibrium 3 different kind of micelles of different size.

Further, usually if you mix 2 micellar solutions, in almost all cases mixed micelles are formed and you should not get separate peaks unless you do not measure very fast.

Alexander Malm ,

a problem with dilution is, it might change the size of oil droplets sizes if the concentration of emulsion stabilizers is changed in the solutions