The U.S. Army Corps of Engineers had an extravagant
slogan during World War II: "The difficult we do at once; the impossible
takes a little longer."
This time division into the easy, the difficult,
and the impossible also applies to chemical reactions. Some chemical
reactions are very fast; others will take place eventually if you
have the patience to wait. Yet
a third class of chemical reactions will never go in a desired direction
without outside help, even if you wait forever.
If you want a particular reaction to occur, it is
obviously of interest to be able to predict into which category
the reaction falls. In the next two chapters we will see what governs
how fast a reaction will go. In this chapter we are concerned with
the simpler question of predicting whether a given reaction will
ever occur by itself, given unlimited time. The key step will be
learning how to measure the order or disorder that is produced when
molecules interact, or the entropy of a reaction.
Whether a reaction ever will proceed by itself depends
on two quantities that sometimes co-operate but more often conflict:
heat or energy, and disorder or entropy.
The central theme of this chapter is that a lowering
of energy, and an increase in disorder, both are changes that tend
to occur spontaneously.
In the melting of the icicle, water can lose heat
and go to a state of lower energy by freezing, but at a cost of
increasing its order in the ice crystal.
Conversely, a frozen icicle can go to a more disordered
state by melting, but only if enough heat is supplied to break the
hydrogen bonds in the ice crystal. The energy factors say "freeze,"
and the entropy factors say "melt". For reasons that we will explore
in this chapter, energy is more important at low temperatures, and
entropy, or disorder, dominates at higher temperatures. The temperature
at which these two conflicting tendencies balance is the melting
point of ice.