Number theoryDifficulty 5.6AIME, harderProve itTaiwan
Prove that for all primes p>100 and every integer r, there exist two integers a and b such that p divides a2+b5−r.
Solution
Throughout this entire solution, all congruence relations are modulo p.
Fix p, let P={0,1,…,p−1} be the complete residue system modulo p. For all r∈P, let Sr={(a,b)∈P×P:a2+b5≡r}, and let sr=∣Sr∣. Our goal is to prove that for all r∈P, sr>0.
We will use the known fact that for all congruence classes r∈P and all positive integers k, there exist at most k values x∈P such that xk≡r.
Lemma. Let N be the number of quadruples (a,b,c,d)∈P4 satisfying a2+b5≡c2+d5. Then N=r∈P∑sr2(1) and N≤p(p2+4p−4).(2)
Proof.
a. For all congruence classes r, there exist exactly sr pairs (a,b) satisfying a2+b5≡r and sr pairs (c,d) satisfying c2+d5≡r. So there exist sr2 quadruples satisfying a2+b5≡c2+d5≡r. Summing over all r∈P, the result follows.
b. Choose an arbitrary pair (b,d)∈P and look for possible values of a,c.
1. Suppose b5≡d5, and let k be the number of pairs (b,d). b can be chosen in p different ways. For b≡0, d can only be 0. For nonzero b, d has at most five possible values. So we have k≤1+5(p−1)=5p−4. The values of a and c must satisfy a2≡c2, so a≡±c, and there exist exactly 2p−1 such pairs (a,c).
2. Now suppose b5=d5. In this case, a and c must be distinct. Using (a−c)(a+c)=d5−b5, the value of a−c uniquely determines the value of a+c, and also determines the values of a and c. Therefore, there exist p−1 suitable pairs (a,c).
Thus, for the k pairs (b,d) satisfying b5=d5, there exist 2p−1 pairs (a,c), and for the other p2−k pairs (b,d), there exist p−1 pairs (a,c). Therefore, N=k(2p−1)+(p2−k)(p−1)=p2(p−1)+kp≤p2(p−1)+(5p−4)p=p(p2+4p−4).
Suppose that for some r∈P, Sr=∅. Clearly r=0. Let T={x10:x∈P∖{0}}. Because every congruence class is the 10th power of at most 10 elements in the set P, from p>100 we obtain ∣T∣≥10p−1≥4.
For all t∈T, we have Str=∅. Indeed, if (x,y)∈Str and t≡z10, then (z−5x)2+(z−2y)5≡t−1(x2+y5)≡r, so (z−5x,z−2y)∈Sr. Therefore, among S1,…,Sp−1 there exist at least 10p−1≥4 empty sets and among s0,s1,s2,…,sp−1 there exist at most p−4 nonzero numbers. Then by the AM−QM inequality, we obtain N=r∈P∖rT∑sr2≥p−41r∈P∖rT∑sr2=p−4∣P×P∣2=p−4p4>p(p2+4p−4), which, by the Lemma, cannot be true.
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 translated into English from zh; metadata (topic, difficulty) added by this project.