When we get a size distribution from dynamic light scattering (DLS), there is always an issue with numbering the peaks. Do you label the peak number from left to right, i.e. small to large size? Or is it better from right to left, that is large size to small size? How do we number the peaks? Do we select the highest one? Let’s look at what we mean by this. Typical bimodal size distribution may have two peaks like below. As an aside, this discussion is valid for the ZSXplorer software.
Does the peak number even matter?
If there was only a single peak, then of course that one would be Peak One. In the above case, the peaks are very different. So you could easily tell from the actual listed values in the Statistics table that one is around 80nm and the other around 400nm. Who cares which one is Peak One and which one is Peak Two? Well, the situation can get more complicated where the difference is not as easily observable. And it can get entirely tricky when overlaying several distributions. Then you need to be very aware of the peak sorting if you look for any statistics on specific peaks like Peak One or Peak Two.
It is the peak by Area !
The Peaks are numbered by their respective area. That is to say, the area under the peak in the graph, the distribution. The peak with the largest total area under the peak is Peak One. The peak with the next largest area is Peak Two and so on. So, after lifting this secret, let’s reiterate that the peak ordering is :
✅ by peak area (largest peak area is number One) ✅
❌ not by size, neither from small to large nor from large to small ❌
❌ not by peak height ❌
Below is a real example to illustrate peak sorting or peak numbering.
What happens when transforming to volume?
If the distribution only has a single peak, then it’s easy. Peak One will remain Peak One for any distribution, be that intensity, or volume, or even number. When there are more peaks, however, this can get tricky. Now the peak, which used to be number Two can out of a sudden become Peak One. Why is that? Because the same ordering principle applies in the transformed distribution: the peak with the largest area under the peak is Peak One and so on. While this may seem confusing at first, there still is the same logic behind it. Just look at each transformation result independently.