American Association for Aerosol Research - Abstract Submission

AAAR 39th Annual Conference
October 18 - October 22, 2021

Virtual Conference

Abstract View


Retrieving Pre-factor and Fractal Dimension of a Single Soot Aggregate

Divjyot Singh, LAURENCE LU, Alexei Khalizov, New Jersey Institute of Technology

     Abstract Number: 569
     Working Group: Carbonaceous Aerosol

Abstract
Combustion soot aggregates have a complex morphology which influences their radiative forcing, surface chemistry, cloud nucleation efficiency, and atmospheric lifetime. The morphology is commonly described by the fractal law, using the scaling prefactor and fractal dimension. These fractal parameters are usually calculated from the dependence of aggregate monomer counts on the radius of gyration across an ensemble of differently sized aggregates. This approach returns the average fractal parameters for the ensemble rather than for a single soot aggregate.

In some applications, like investigating soot restructuring, it is necessary to retrieve fractal parameters of a single aggregate. Such analysis has been performed on 3D electron tomography images using the common cube-counting method, which discretizes an aggregate into voxels of a certain size and counts the number of voxels forming the aggregate. This process is repeated several times with different voxel sizes and the fractal parameters are then derived from the linear regression of a log-log plot of the voxel counts against the voxel sizes. This cube-counting method is computationally expensive. Therefore, we explore the application of a monomer-based approach, in which a large number of sub-aggregates are randomly chosen in a range of sizes inside the aggregate. We derive the fractal parameters from the linear regression of a log-log plot of the monomer counts of the sub-aggregates against their radii of gyration.

We test both methods on aggregates of varying sizes and morphology, generated via a diffusion-limited cluster-cluster aggregation algorithm, and compare their speed and accuracy. We illustrate how the monomer-based approach is computationally fast and can be used in simulations of soot aggregates where fractal parameters need to be found frequently.