1. INTRODUCTION

1
Ice Nucleation by Refractory Nanoparticles
Russell Saunders1, Ottmar Möhler2, Martin
Schnaiter2, Stefan Benz2 and John Plane1
1 School of Chemistry, University of Leeds,
Leeds LS2 9JT UK, 2
Forschungzentrum Karlsruhe, IMK, 76021 Karlsruhe,
Germany. E-mail r.w.saunders_at_leeds.ac.uk
or j.m.c.plane_at_leeds.ac.uk
.
1. INTRODUCTION
5. Contact Angles
4. Data Set for Fe2O3 Nanoparticles

Nucleation rates in the chamber are calculated
from the ratio of the rate of ice crystal growth
to aerosol surface area

• We have studied the role of nanoparticles of
• meteoric composition (i.e. metal and silicon
• oxides) in the heterogeneous formation of ice
• at the AIDA facility in Karlsruhe, Germany.
• The participation of such aerosol in the
  formation of NLC has long been implicated but
• never examined under controlled conditions
  approaching those of the mesosphere.
• Results of experiments on in-situ generated
  amorphous Fe2O3 particles, and crystalline powder
  samples of Fe2O3, SiO2 and MgO are used to
  determine the contact angle (?) for these
  compounds the critical parameter which defines
  the ice nucleating ability of a substance.

(1)
Classical Nucleation Theory
(2)

• rN is the radius of the nucleating particle
  (taken as the modal size in the aerosol -
  surface area distribution),
• A is a pre-exponential factor,
• ?F is the free energy of ice embryo formation.

2. Ice Nucleation Chamber (AIDA)
The main (AIDA) chamber provides 84 m3 of air
which can be adiabatically expanded to set up a
super-saturation with respect to ice at
temperatures down to 180 K. Aerosol samples can
be introduced via a powder disperser or an inlet
port for particles generated from the gas-phase
in a flow cell adjacent to the chamber. A range
of aerosol/ice particle detection instruments
connected to AIDA or a smaller pre-chamber (APC)
are used to monitor the timed-evolution of ice
nucleation. An Ar laser system is also used to
detect forward/backscatter and depolarisation at
488 nm, resulting from the formation and growth
of ice crystals
(3)

• Figure 3. Initial air temperature in chamber is
  200 K (see top panel). Next panel down shows the
  variation in RH (ice) solid line and RH (water)
  dashed line. Third panel down indicates the
  growth in ice crystal number (cm-3) during the
  course of the expansion cycle while the bottom
  panel tracks the size-resolved detection of ice
  crystals. All plots are shown with expansion time
  (s) on the abscissa.
• The detected initial aerosol distribution had a
  modal diameter of 25 nm.

• ag is the critical embryo radius,
• ssv is the surface tension of ice/vapour
  interface,
• f (m,x) is the matching function

where
and
As ? ? 0, f (m,x) ? 0 i.e. a perfect epitaxial
fit between ice embryo and nucleating particle
Fe2O3 nanoparticles
Water droplets detected at RH (ice) lt 100
Figure 5. Calculated ? values for all samples
studied in the AIDA chamber. The cartoon in the
figure depicts the definition of ? with respect
to heterogeneous ice nucleation. The dashed line
at ? 18.2 represents a nominal value assumed
in calculations for meteoric smoke
Ice crystals
Figure 1. Schematic of the AIDA chamber and
accompanying instrumentation used in the ice
nucleation experiments.
3. In-Situ Particle Production
6. CONCLUSIONS
Figure 4. Variation in the ice number fraction
(nf) - the ratio of ice crystal to initial
aerosol particle number, with RH (ice). The
horizontal dashed line represents a defined
threshold nucleation level while the vertical
line indicates the corresponding threshold RH
(ice) level of 118.
Fe2O3 nanoparticle contact angles are small (good
ice nucleators) but show a slight dependence over
the temperature range accessible in the AIDA
chamber. Recent laboratory work with a silicon
wafer (Trainer et al., JPC, 113, 2009) indicates
that such a temperature dependence can be much
stronger at lt 180 K. For T lt 150 K (polar summer
mesopause), electrostatic charging of particles
is likely to remove such a constraint (Gumbel
Megner, JASTP, in press)
The identified threshold conditions at a given
temperature are used to determine the surface
nucleation rate (J cm-2 s-1), which in turn is
used to evaluate the contact angle (?) for the
sample.
Figure 2. Schematic of the aerosol flow system
used to generate iron oxide nanoparticles from
the photo-oxidation of iron pentacarbonyl.