login

Year-end appeal: Please make a donation to the OEIS Foundation to support ongoing development and maintenance of the OEIS. We are now in our 61st year, we have over 378,000 sequences, and we’ve reached 11,000 citations (which often say “discovered thanks to the OEIS”).

Determinant of the matrix [Jacobi(i^2 + 3*i*j + 2*j^2, 2*n + 1)]_{1 < i, j < 2*n}, where Jacobi(a, m) denotes the Jacobi symbol (a / m).
3

%I #34 Apr 27 2024 12:30:18

%S 1,3,0,125,-1215,0,0,9126441,0,-187590821,0,0,20686753425,0,0,0,

%T 9224101117395305225,0,881852208012283730302080,624391710361368134976,

%U 0,-3428714319207136609529065,0,0,3878246452353765171209988566241,0,0,4308304210666498856284267223158421

%N Determinant of the matrix [Jacobi(i^2 + 3*i*j + 2*j^2, 2*n + 1)]_{1 < i, j < 2*n}, where Jacobi(a, m) denotes the Jacobi symbol (a / m).

%C Conjecture 1: Let n be any positive integer.

%C (i) If a(2*n) is nonzero, then 4*n + 1 is a sum of two squares.

%C (ii) a(2*n + 1) is divisible by phi(4*n + 3)/2, where phi is Euler's totient function. If n is even, then a(2*n + 1)/(phi(4*n + 3)/2) is a square. This has been verified for n = 2..1000.

%C For any odd integer n > 3 and integers c and d, we introduce the notation: {c,d}_n = det[Jacobi(i^2 + c*i*j + d*j^2, n)]_{1 < i, j < n-1}.

%C The following conjecture is similar to Conjecture 1.

%C Conjecture 2: (1) {2, 2}_p = 0 for any prime p == 13,19 (mod 24), and {2, 2}_p == 0 (mod p) for any prime p == 17,23 (mod 24).

%C (2) If n == 5 (mod 8), then {4, 2}_n = 0. If n == 5 (mod 12), then {3, 3}_n = 0.

%C (3) If n == 5 (mod 12) and n is a sum of two squares, then {10, 9}_n = 0. Also, {10, 9}_p == 0 (mod p) for any prime p == 11 (mod 12).

%C (4) {8, 18}_p == 0 (mod p^2) for any prime p == 19 (mod 24), and {8,18}_p == 0 (mod p) for any prime p == 23 (mod 24). If n == 13,17 (mod 24) and n is a sum of two squares, then {8, 18}_n = 0.

%C We have verified Conjecture 2 for p or n smaller than 2000.

%H Zhi-Wei Sun, <a href="/A372314/b372314.txt">Table of n, a(n) for n = 2..73</a>

%H D. Krachun, F. Petrov, Z.-W. Sun and M. Vsemirnov, <a href="https://doi.org/10.1016/j.ffa.2020.101672">On some determinants involving Jacobi symbols</a>, Finite Fields Appl. 64 (2010), Article 101672.

%H Z.-W. Sun, <a href="https://doi.org/10.1016/j.ffa.2018.12.004">On some determinants with Legendre symbol entries</a>, Finite Fields Appl. 56 (2019), 285-307.

%e a(2) = 1 since the determinant of the matrix [Jacobi(i^2 + 3*i*j + 2*j^2, 5)]_{1 < i, j < 2*2} = [1,0; 0,1] is 1.

%t a[n_]:=a[n]=Det[Table[JacobiSymbol[i^2+3*i*j+2*j^2,2n+1],{i,2,2n-1},{j,2,2n-1}]];

%t tab={};Do[tab=Append[tab,a[n]],{n,2,29}];Print[tab]

%o (PARI) f(i,j) = i^2 + 3*i*j + 2*j^2;

%o a(n) = matdet(matrix(2*n-2, 2*n-2, i, j, kronecker(f(i+1,j+1), 2*n+1)));

%o vector(25, n, a(n+1)) \\ _Michel Marcus_, Apr 27 2024

%Y Cf. A000010, A000040, A001481.

%K sign

%O 2,2

%A _Zhi-Wei Sun_, Apr 27 2024