Snowflakes start out as little seeds --
from the supersaturation vs. temperature graph on the "shapes" page, you
can see that most start out as hexagonal prisms, and that most are plates,
until you get to very low temperatures, where the plates get thick, becoming
more like prisms. The difference is shown in the figure at right.
Modeling the diffusion like we do with our applet is approximating the
plate to be very thin -- two dimensional. Also, the hexagonal plate
structure has to do with the crystal lattice and surface physics of ice,
which we won't discuss on this page -- in fact, not much is known about
it, and it's still a very active area of research.
The reason that snowflake growth is symmetrical
is thought to be this: A seed that starts out as a hexagon has corners
which poke out into the surrounding vapor more than the edges do.
As a result, the corners grow more quickly. So whenever a part of
the growing flake pokes out a bit, it tends to grow even more! That's
why snowflakes are so complicated. The symmetry happens because of
the many different crystal structures seen in the graph on the "shapes"
page -- as the snowflake gets shuffled around in the cloud, it moves into
different regions of the graph, and so the same thing happens to all six
corners at the same time, thus preserving symmetry. Plus, if the
snowflake really does grow because of the random diffusion of particles,
then it helps to have a whole lot of particles -- the more particles, the
longer the flake has to develop symmetries.
Unfortunately, random diffusion is not
all that's going on. All sorts of surface physics happens at the
snowflakes boundary, which includes surface tension and lattice structure
of the crystal.
But! We're physicists, so since when
do we bother with complicating things? Our model will only try to
explore the consequences of random diffusion.
MAIN
PAGE SNOWFLAKE
SHAPES DIFFUSION
OUR
SPIFFY MODEL OF DENDRITIC GROWTH
OUR
HOT APPLET OF DENDRITIC GROWTH
THE
RESULTS OF OUR HOT APPLET OF DENDRITIC GROWTH
LINKS
