SAXS analysis: The analysis pipeline developed in Mathematica (*.nb). To use the scripts, follow these steps:

1. First the notebook "Polydisp fitting functions thickness and radius - number weighing full form factor.nb" should be run. It does the following: 
- It imports all the data from the SAXS data folder (note that you should set the folder location to the relevant storage location on your computer). 
- The notebook contains all prerequisite functions for data processing (background subtraction, merging of SAXS patterns, averaging, binning, slope calculation), form factor fitting and visualization.
- It contains some example form factor as also provided in Figure (S)1. 

2. To analyze a specific SAXS sample, run the notebook in the corresponding folder with the suffix "- number", e.g. "Sample playground LH-ESRF-C22 NEW.nb". These notebooks contain the number-weighted SAXS analysis as detailed in the manuscript. 

3. To compare the SAXS data with DLS data, the SAXS data was analyzed in an intensity-weighted manner rather than number weighted. To reproduce these findings, first run the notebook "Polydisp fitting functions thickness and radius - intensity weighing.nb" and follow with the notebook (suffix: "- intensity") for the sample of interest.


Note: the intensity-weighted fits are only used to compare SAXS with DLS data.

Sample names:

C14: GNP synthesized with 6 mg/mL EC, 500*g centrifugation
C15: GNP synthesized with 6 mg/mL EC, 1000*g centrifugation
C16: GNP synthesized with 6 mg/mL EC, 2000*g centrifugation
C20: GNP synthesized with 3 mg/mL EC, 500*g centrifugation
C21: GNP synthesized with 3 mg/mL EC, 1000*g centrifugation
C22: GNP synthesized with 3 mg/mL EC, 2000*g centrifugation

E27: Gibbsite from Janne-Mieke Meijer & Tom Jenniskens (1.0 wt% in water)
E72: Gibbsite from Mark Vis DB940 (0.18 wt% in water)


version 1 - corrections:
* We realized the number weighing was not fully correct because, by multiplying by V^2 (n^2R^4), we had assumed linear bin sizes. However, the bins were logarithmic. In the new version, we've corrected for this by multiplying with the bin size.
* We have added plots where we plot the diameter distribution instead of the radius distribution.
* We made the form factor plots in Figure (S)1 with tick marks every 3 decades instead of 2. We have written a PowerTicks3 function for this.
