- #36
bbbeard
- 192
- 4
vibhuav said:Can someone give a physical meaning for imaginary numbers?
As Ken G has indicated, complex numbers are "rotational" numbers. They are handy whenever a physical problem involves a rotation or a phase. Consider, for example, a case where we have two unit vectors which make angles "a" and "b" with the x axis, and want to know how adding the angles changes the projections onto the x and y axes. In other words, we want to know how the sines and cosines of a and b relate to the sines and cosines of the total (a+b). We could look up our trig identities
sin(a+b) = sin(a)*cos(b) + cos(a)*sin(b)
cos(a+b) = cos(a)*cos(b) - sin(a)*sin(b)
or we could just use the identity
exp(i*x) = cos(x) + i*sin(x)
to learn
exp(i*(a+b)) = exp(i*a)*exp(i*b)
=(cos(a) + i*sin(a))*(cos(b) + i*sin(b))
=(cos(a)*cos(b) + i2*sin(a)*sin(b)) + i*(sin(a)*cos(b) + cos(a)*sin(b))
= cos(a+b) + i*sin(a+b)
... which only works because i2 = -1. In other words, the algebra of rotations is embedded within the complex numbers by virtue of the definition i2 = -1.
This has many uses. For example, if we have two arbitrary vectors, if we want to add them to find a resultant, then ordinarily we can just add them component-by-component. So converting them to complex notation (x,y) => x + i*y doesn't buy you much. But if you want to rotate a vector, you can simply multiply (x+i*y) by exp(i*theta) to find a new rotated vector.
Because waves are for the most part oscillatory phenomena, complex numbers help to denote the time evolution, exp(i*omega*t), or the spatial dependence, exp(i*k*x). We could write the same thing in terms of sines and cosines, of course, but the complex exponential is much simpler to manipulate. This "rotational" aspect of complex numbers is why they are so ubiquitous in quantum mechanics.
BBB