Quantcast
Channel: Proving that no tile can fill both squares and equilateral triangles - Mathematics Stack Exchange
Viewing all articles
Browse latest Browse all 2

Proving that no tile can fill both squares and equilateral triangles

$
0
0

Cut up a square into a finite number of identical tiles.

Here is one possibility:

square with 4 tiles

How do I prove that the tiles could never be rearranged to form an equilateral triangle (with filled interior and no overlapping tiles)? This is easy to see for the cut drawn above (though I intentionally chose 60 degree angles), but how do I prove it in general?

As for my attempts at proving this, I think this becomes a question of listing the ways that a square can be cut into equal pieces. Instead of the cut drawn above, one could cut the square into "rational rectangles" and halve the rectangles into triangles, but all angles then must have rational tangents and therefore can never sum to a 60 degree angle. Also, by equating the area of the square and the area of the equilateral triangle, we know that some tile edge lengths must irrationally divide the square edge length. Beyond these two conclusions, I am stuck, but have a feeling that there is a very easy proof here which will leave me feeling foolish.

To be clear, all tiles must be used in the rearrangement and no tiles can be flipped over, but I think I could relax these two conditions also and still never form an equilateral triangle.


Viewing all articles
Browse latest Browse all 2

Latest Images

Trending Articles





Latest Images