I suggested in a comment, to look for colorings that are the same for each <i>a</i>, except for a fixed permutation.

Suppose the colors are given by function c(x), for x &ge; 1. And a permutation function p(i,j), where i,j &isin; [1..n]. Where i is the index of the permutation, and j the index of the permutation of the color.

If n = 4 and the coloring starts with abcd, then p(1,1) = a, p(1,2) = b, p(1,3) = c, p(1,4) = d, p(2,1) = b, p(2,2) = d.

If the coloring is the same for each <i>a</i>, then we get the following equation:

c(an) = p(c(a),c(n))

From this it follows, that the permutation must p(i,j)=p(j,i).

However, it is more important to give a condition when a coloring can be constructed:

<b>Lemma 1:</b> If p(x,y) = p(v,w) for xy = vw, then a coloring can be constructed.

<b>Lemma 2:</b> For each n a permutation exists such that p(x,y) = p(v,w) for xy = vw.

To see how this works, for n = 4, we start the following permutation matrix:<br/>
abcd<br/>
bd..<br/>
c...<br/>
d...<br/>

Fill in the second row:<br/>
abcd<br/>
bdac<br/>
ca..<br/>
dc..<br/>

And complete it:<br/>
abcd<br/>
bdac<br/>
cadb<br/>
dcba<br/>

From this construct the coloring:
abcdxaxcdxxbxxxa

By each prime number > n (the x values in the coloring), you can choose a color, but you have to continue with the permutation of that color.

Lemma 1 is trivial to prove, because it follows that c(xy) is c(yx).

I haven't proven lemma 2 yet. For n is 1,2,3,4, there is only 1 permutation matrix. For n = 5 there are a few. The condition p(x,y) = p(v,w) for xy = vw becomes relative weaker for larger n. So, I don't think it is a problem to create such permutation matrix (also given the number of different Sudokus). Of course, lemma 2 needs to be proven properly.

<b>Edit:</b> For Lema 1, it might be necessary to add an associativity condition. That p(x,p(y,z)) = p(p(x,y),z). With associativity, the proof is trivial. I don't know if it can be done without. 

<b>Edit 2:</b> The associativity condition is indeed necessary. And this is also part of the solution. If you have a permutation that consists of 1 cycle, then the full permutation matrix can be created by applying the permutation multiple times. This ensures the associativity condition.

It is important that after n times applying the permutation, the colors are in original order again. This means that for n = 5, you must have 1 cycle, because one cycle of 3 plus one cycle of 2 does not come in original order after applying 5 times.

However, this does not count for n = 6. The solution of François is not a permutation of 1 cycle:<br/>
abcdef<br/>
bdface<br/>
cfbead<br/>
daebfc<br/>
ecafdb<br/>
fedcba<br/>

As you can see, a brings you to b, from b to d and back to a. You need a second permutation to bring a to c, b to f etc.

A permutation with 1 cycle for n = 6 from which a coloring can be created, also exists.

I think it is easier to create a coloring with a permutation of 1 cycle. You only need to ensure that the few multiplications that are below n fit, and then create the permutation to have 1 cycle. I think for larger n, there is plenty of possibilities for that.