On Composition, Morphology, and Size Distribution of Airborne Mineral Dust - Mineral Dust: A Key Player in the Earth System - page 27

2.3.1 Particle Shape and Morphology

Generally, mineral dust particles are nonspherical. Their shapes 1 vary between
crystals without defects and aggregates of small irregular grains, but also near-
spherical shapes are found for processed dust particles (Laskin et al. 2005a , b ).
The shape of the dust particles has importance for their radiative interaction
(e.g., Kalashnikova and Sokolik 2002 ; Nousiainen 2009 ) and for their settling
behavior (Ginoux 2003 ;LiandOsada 2007a ), though for both influences, high
uncertainties exist (Nousiainen et al. 2011 ). Surface roughness influences active
remote sensing (Zubko et al. 2007 ; Gasteiger et al. 2011 ). Shape and morphology
also affect the effective surface area and therefore might modify heterogeneous
chemical reactions.
Currently, there are mainly three approaches to measure the particle shape:
microscopy, radiation scattering, and measurement of aerodynamic properties. Light
microscopy, electron microscopy, and atomic force microscopy have been used
in many cases to study the shape of single mineral dust particles collected on
substrates. The depolarization of scattered or transmitted radiation has been used
to retrieve information on effective particles shapes (e.g., Dubovik et al. 2006 ;
Müller et al. 2012 ). However, the retrieved results do not necessarily resemble
real particles shapes (Nousiainen et al. 2011 ). Aerodynamic measurements have
also been used to determine the effective particle shape parameter (Kaaden et al.
2009 ). While the latter two approaches lack the close connection to real geometrical
shape, the microscopy approach instead is usually limited by the comparatively
small number of analyzed particles, a limitation which is somewhat overcome by
the use of automated instruments (e.g., Reid et al. 2003 ; Kandler et al. 2007 ;Coz
et al. 2009 ).
From the microscopic techniques, primarily a two-dimensional projected shape
of the particle is retrieved, which then is usually parameterized by several shape
descriptors. Mainly the aspect ratio is used, in principle defined as the ratio
of the longer to the shorter axis of an ellipse fitted to the particles shape but
often also calculated as the ratio of four times the squared particle's longest
projection to pi times the projected area. We will use the latter definition for
the data presented here. It must be emphasized that the difference between these
two definitions are significant for non-elliptic shapes like those of dust particles.
In addition, other shape descriptors like fractal dimension, circularity, roundness,
shape factor, and solidity are sometimes used. The shape descriptors are usually
derived from discretized images; the latter four - depending on shape perimeter
or holes - are more prone to resolution-dependent artifacts (Podczeck et al. 1999 ;
Almeida-Prieto et al. 2007 ). An illustration of several shape descriptors applied

1 The expressions particle shape and particle morphology are used interchangeably throughout the
literature with slightly different meanings. We will use the term particle shape to describe the
general outline of a particle, and the term particle morphology shall refer to small-scale structures
like surface roughness.