We rearrange the initial equation (b2+7(a−b))2=a3b: (b2+7(a−b))2=a3b⇔b4+14b2(a−b)+49(a−b)2=a3b⇔a3b−b4−14b2(a−b)−49(a−b)2=0⇔b(a3−b3)−14b2(a−b)−49(a−b)2=0⇔(a−b)(ba2+ab2+b3−14b2−49(a−b))=0.(1) If a=b, then it is evident that (1) holds. Therefore we have the infinite set of the integer solutions: a=b=t, where t∈Z.
Let a=b, then from (1) it follows that ba2+ab2+b3−14b2−49(a−b)=0.(2) If b=0, then from (2) it follows that a=0, i.e. a=b=0, a contradiction. Thus b=0. Consider (2) as a quadratic equation with respect to a. ba2+(b2−49)a+b3−14b2+49b=0.(3) The discriminant D(b) of this equation is D(b)=(b2−49)2−4b(b3−14b2+49b)=((b−7)(b+7))2−4b2(b−7)2==(b−7)2((b+7)2−4b2)=−(b−7)3(3b+7). Equation (3) has real solutions iff D(b)≥0, i.e. for b∈[−7/3;7]. The segment [−7/3;7] contains exactly ten integers: −2,−1,0,1,2,3,4,5,6 and 7. Since b=0 and a1,2=(49−b2±D)/(2b), we have exactly four integers (b=−2,3,6,7) such that D(b) is a perfect square (we need integer solutions of (3)).
1) If b=−2, then D=272 and either a1=−9/2 or a2=−18, so, in this case (a;b)=(−18;−2) is a required solution.
2) If b=3, then D=210 and either a1=4/3 or a2=12, so (a;b)=(12;3) is a required solution.
3) If b=6, then D=52 and either a1=13/12 or a2=85/12, so there are no solutions in this case.
4) If b=7, then D=0 and a1,2=0, so (a;b)=(0;7).
Want a route through all this instead of an archive? The track
puts 2,000 problems in a working order, from AMC 10 level to the IMO shortlist.
Source: MathNet,
licensed CC-BY-4.0.
Statement and solution reproduced as published; topic and difficulty added by this site.