Choosing the Concentration Range for DLS Size Measurement

When nanoparticle size is determined using dynamic light scattering (DLS), one important parameter is particle concentration. When the particle concentration is too low, the scattering from the particles is weak and the measurement results are noisy. When the particle concentration is too high, the measurement results are distorted due to particle-particle interactions. These interactions can be particle-particle collisions at high particle concentrations. Or, the interactions can be long range electrostatic interactions. In either case, these interactions change particle motion compared to the motion assumed when calculating particle size from DLS data. At high particle concentration, the data seems normal, but the size results are distorted. Therefore, when developing methods, it is important to keep particle concentration in mind.

There are two approaches to choosing particle concentration. The first is to inspect suspensions and choose a clear to just barely cloudy suspension. This is the so-called eyeball method. The second approach is to measure the size of a series of suspensions and plot measured size as a function of particle concentration. One then chooses the range of concentration over which the measured size does not change.  Note that concentration is varied exponentially rather than linearly in such a study. That is, the series of concentrations should be of the form 2, 4, 8, 16 … rather than 2, 4, 6, 8 … . When encountering a new nanoparticle system for dynamic light scattering size measurement, the first technique is most common. Once an approximate concentration range is established, the concentration study becomes significantly easier and faster. 

Below are results of studies on two particle systems. One has an average size of about 1,000 nm and the other has an average size of about 36 nanometers (and is a true nanoparticle).  Here, we present photographs of the suspensions in sample cells to show what nice looking suspensions look like. These photos can be used to guide implementation of the eyeball method. We also show the results of the concentration studies to demonstrate how such studies should be implemented.

The first study shown here is for a sample with an approximate particle size of one micron. In the photograph below, we see a series of suspensions of increasing cloudiness. In anticipation of the next figure which shows the measurement results, we note the suspensions that gave nice measurement results. 

In the figure below, a plot of measured size as a function of concentration is shown. The error bars show the standard deviation of repeated measurements. At low concentration, the results are noisy. At high concentration, the size results are distorted. 

The second study is of a nanoparticle with a size of about 36 nm. Due to the smaller particle size, these suspensions are clearer than the suspensions shown above. Here, the samples with the optimum nanoparticle concentration are somewhat clearer than those in the example above.

In the figure below, a plot of measured size as a function of concentration is shown. At high concentration, the size results are distorted. Note that the range of concentrations for measurement is much higher for this nanoparticle than for the one micron particle shown in the first example. In fact, the optimum concentration range depends on the particle size distribution and scattering power of the particle. 

The best approach to characterizing a new system is to find the concentration range where the suspension looks OK. Then, use this knowledge to guide a concentration study. Fortunately, dynamic light scattering measurements are often not particularly sensitive to concentration. In both of the examples above, the range of concentrations for good particle size measurement results spans roughly one decade. For a single measurement, the eyeball method may be sufficiently precise. When planning to use DLS to study a nanoparticle system in depth or when developing QC method, a concentration study will help ensure the best quality data.