Re: What is the chemistry behind borax crystals?

Dear User,

I assume that you mean borax = Na2B4O7x10H2O, hydrated sodium borate. 
Otherwise, you'd have to clarify what kind of boric acid salt are you using 
and what are the conditions of the experiment. Perhaps, it should be noted 
in general that the more information the curious users of the MadSci 
network provide with their question, the more accurate would our answers 
be...

For the decription of borax, please look at 
 http://mineral.galleries.com/minerals/carbonat/borax/borax.htm

Crystallization from solution involves obtaining saturated solution, and 
bringing it to the point of supersaturation and nucleation. Nucleation is 
not directly related to supersaturation, in fact many substances can be 
brought to the supersaturation levels of 200% and more, without 
crystallization occuring until a nucleus is introduced externally. Most 
salts, however, crystallize spontaneously when their concentration reaches 
certain level. Borax can be crystallized in many shapes, but most of these 
crystals would still belong to the same monoclinic crystal form. What makes 
shapes different, in most cases, is preferential development of certain 
faces of the growing crystal and also sporadic nucleation on the surface of 
the crystal. Ideally, borax crystals are near-square prisms, but under 
certain conditions, they can grow into a variety of shapes, some of which 
resemble snowflakes. There is nothing surprising about that, if one 
considers the fact that snowflakes are crystals too. The fact that these 
crystals (or, sometimes, crystalline assemblys) look nice rather than being 
just boring geometrical figures is a subject matter of a branch of physics 
which deals with chaos in materials and interfaces. The rules are more or 
less generic for all substances, not just for ice or borax, but of course, 
the crystal structure of each particular substance would determine whether 
it can or cannot be crystallized as "snowflakes" "feathers" and other 
complicated shapes rather than more conventional-like crystals. As a small 
illustration of how a snowflake may be formed, consider the following: 
assume that you have a growing crystal. At any given moment of time, there 
is a possibility that another crystal would nucleate on the growing face. 
Also, there is a probability distribution governing which faces grow 
fastest. Anf, finally, since growing crystal depletes the solution around 
it, some points of the crystal grow very fast, forming spikes, whilst the 
other points are "left behind" having to grow in the medium which has 
already been depleted by the faster-growing parts. s a result, for 
instance, a fast-growing edge would develop a "smile" in the middle, 
because the middle part would not get as much material deposited as the 
ends. If a hexagonal crystal is growing very fast diffusion does not bring 
the concentrations to equilibrium fast enough and the resulting shape would 
be a six-spiked star, instead of a hexagon. That's your basis for a 
snowflake. Now add some bifurcations along the spikes, and you get the 
second step, then you can add bifurcations on the bifurctions and so forth, 
until the resulting shape is ... a snowflake. 
The shape is governed by an array of parameters, which would include: 
temperature, pH, concentration of the substance, additives, pressure etc. 
a small demo of snowflake shape dependance on temperatures may be found 
here http://www.gfherald.com/news/weather/howm/shapes.htm

Cheers !

A.G.E.