Nucleation effects on the habit of vapour grown ice crystals from −18 to −42^°C

Abstract

SUMMARYIce crystals have been nucleated and grown by a new method where supersaturation is controlled by combining static diffusion chamber and expansion chamber techniques. Crystals were nucleated and grown on 50 µm diameter glass filaments at pressures of 500 to 300 hPa, typical of clouds with temperatures from −18 to −42 • C, respectively. Crystals were also nucleated on particles of finely powdered silver iodide and kaolinite adhering to glass filaments, in order to investigate effects on habit due to different nucleating materials. Critical supersaturations for nucleation were measured for each case and were in good agreement with previous nucleation studies. Crystals were grown at ice supersaturations which ranged from less than 1% to just above water saturation. For ice growth below a critical supersaturation, nucleation was achieved by applying a brief adiabatic expansion to the static diffusion chamber with a filament in place and the chamber evacuated to the target pressure for the temperature of interest. Nucleation occurred at a critical supersaturation, followed by growth at a lower value. A review of past laboratory experiments and recent in situ observations reveals good agreement between laboratory and field observations when the effects of nucleating methods and growth times are taken into account. Contrary to studies which report the habit of ice as 'prism'-like or columnar below approximately −22 • C, the habit is found to be plate-like for temperatures ranging from −18 • C to −40 • C at ice supersaturations relevant to the atmosphere. Additionally, the plate-like-habit distribution is dominated by polycrystalline forms, which is consistent with aspects of past laboratory studies and recent observations obtained with improved aircraft instrumentation. The habits and habit distributions obtained from glass and kaolinite most closely resemble those observed in the atmosphere, while the results from silver iodide show large differences. Below −40 • C, the habit becomes columnar except at low supersaturation, and is still dominated by polycrystals which appear as rosettes for ice supersaturations greater than about 20%. 'Bullet rosettes' were observed to nucleate and grow on clean glass threads below −40 • C but not on kaolinite which predominantly grew single columns and irregular polycrystals with columnar components; silver iodide produced single columns with few polycrystals. This behaviour indicates that rosettes result from ice nuclei which share few or no crystallographic similarities with ice, and is consistent with that expected from homogeneous nucleation. Critical supersaturation measurements for nucleation were extended down to −70 • C, and revealed that below −40 • C ice crystals can nucleate on glass at ice supersaturations of 25%, and at values of approximately 15% on kaolinite or silver iodide.

Freezing of water droplets colliding with kaolinite particles

“…Several earlier studies (Mason and Maybank, 1958;Roberts and Hallett, 1968;Shaller and Fukuta, 1979;Bailey and Hallett, 2002) demonstrated that the threshold temperature for ice nucleation by deposition freezing on kaolinite is about 253 K. For immersion mode freezing, an onset temperature of 259 K has been reported (Pitter and Pruppacher, 1973). The results presented here confirm that contact freezing is the most efficient ice nucleation process involving kaolinite particles, with freezing temperatures being observed up to 267 K. In previous work Pitter and Pruppacher (1973) studied contact freezing of water drops with kaolinite and montmorillonite particles.…”

Some ice nucleation characteristics of Asian and Saharan desert dust

“…Following this, the concentrations fell back to the background levels and then eventually rose again, increasing the activated fraction to just below the 8% threshold while the humidity remained close to or at water saturation. This result seems anomalous but it may indicate that deposition nucleation is occurring at these relatively warm temperatures, as found by Roberts and Hallett (1968) and Bailey and Hallett (2002). This difference may be compositional, size related or a result of low nucleation efficiencies (<0.5% of the dust sample) that mean we are not able to observe the ice produced via deposition at these temperatures with the SID.…”

“…They found that kaolinite particles that had previously formed ice crystals now activated at 10% supersaturation with respect to ice, suggesting that ageing could produce more efficient ice nuclei. More recently, Bailey and Hallett (2002) have nucleated ice on glass filaments coated with kaolinite at temperatures up to −12 • C and 12% supersaturation with respect to ice. In contrast, Salam et al (2006) used commercially available mineral dust particles and found that warmest temperature at which kaolinite particles with a mean diameter of 1 µm formed ice was −22 • C and 23% supersaturation with respect to ice, which is very close to water saturation.…”

Dimensions and aspect ratios of natural ice crystals

“…Since initial work that identified crystal shapes using in situ ob-servations of ice clouds (e.g., Weickmann, 1948, and studies summarized by Dowling and Radke, 1990, and by Heymsfield and McFarquhar, 2002), much effort has been devoted to quantify the microphysical and scattering properties of ice crystals. One important finding is that ambient atmospheric conditions (e.g., temperature and humidity) govern the growth and morphological properties of ice crystals (e.g., aufm Kampe et al, 1951;Lamb and Scott, 1974;Gonda, 1980;Fukuta and Takahashi, 1999;Bacon et al, 2003;Bailey and Hallett, 2002, 2012Korolev et al, 2004) upon which the corresponding scattering properties depend (e.g., Takano and Liou, 1995;Macke et al, 1996;Yang and Liou, 1998;Baran et al, 2001Baran et al, , 2005Yang et al, 2005Yang et al, , 2013Um and McFarquhar, 2007.…”