Doubling the bit length makes the problem much much harder than just twice as hard. The n in the sqrt(n) you mentioned is the number being factored, which grows

Doubling the bit length makes the problem much much harder than just twice as hard. The n in the sqrt(n) you mentioned is the number being factored, which grows exponentially with bit length.