cp's OEIS Frontend

This is a front-end for the Online Encyclopedia of Integer Sequences, made by Christian Perfect. The idea is to provide OEIS entries in non-ancient HTML, and then to think about how they're presented visually. The source code is on GitHub.

A267028 P(n,k) is an array read by rows, with n > 0 and k=1..5, where row n gives the chain of 5 consecutive primes {p(i), p(i+1), p(i+2), p(i+3), p(i+4)} having the symmetrical property p(i) + p(i+4) = p(i+1) + p(i+3) = 2*p(i+2) for some index i.

Original entry on oeis.org

18713, 18719, 18731, 18743, 18749, 25603, 25609, 25621, 25633, 25639, 28051, 28057, 28069, 28081, 28087, 30029, 30047, 30059, 30071, 30089, 31033, 31039, 31051, 31063, 31069, 44711, 44729, 44741, 44753, 44771, 76883, 76907, 76913, 76919, 76943
Offset: 1

Views

Author

Michel Lagneau, Feb 23 2016

Keywords

Comments

a(3 + 5*(n-1)) = A051795(n).
The immediate objective of the sequence is to examine symmetrical properties in the array P(n,k). It is interesting to note that the results with the dimension 5 are generalizable to the dimensions 7, 9, ...
Notation:
We introduce the following function S(i,j) where row i is defined by {P(i,k)} and row j is defined by {P(j,k)}, k = 1..5. Let S(i, j) = 1 if P(i,1) + P(j,5) = P(i,2) + P(j,4) = P(i,3) + P(j,3), otherwise 0.
Conjecture:
For each integer n, there exists an infinite sequence of integers b(n,m), m = 1, 2, ... such that S(n, b(n,m)) = 1.
The following table gives the first values b(n,m).
Notation in the table: "PS" = primitive sequence.
+----+------------------------------------------------+-----------+
| n | sequences b(n,m), m=1,2,... of index |included in|
+----+------------------------------------------------+-----------+
| 1 | 1, 2, 3, 5, 8, 9, 10, 12, 15, 16, 17, 18, ... | PS |
| 2 | 2, 3, 5, 8, 9, 10, 12, 15, 16, 17, 18, 19, ...| {b(1,m)} |
| 3 | 3, 5, 8, 9, 10, 12, 15, 16, 17, 18, 19, ... | {b(1,m)} |
| 4 | 4, 6, 11, 13, 14, 21, 28, 35, 39, 57, 59, ... | PS |
| 5 | 5, 8, 9, 10, 12, 15, 16, 17, 18, 19, 22, ... | {b(1,m)} |
| 6 | 6, 11, 13, 14, 21, 35, 39, 57, 59, 63, 67, ...| {b(4,m)} |
| 7 | 7, 30, 52, 55, 73, 74, 115, 159, 177, 183, ...| PS |
| 8 | 8, 9, 10, 12, 15, 16, 17, 18, 19, 22, 23, ... | {b(1,m)} |
| 9 | 9, 10, 12, 15, 16, 17, 18, 19, 22, 23, 24, ...| {b(1,m)} |
| 10 | 10, 12, 15, 16, 17, 18, 19, 22, 23, 24, 26, ...| {b(1,m)} |
| 11 | 11, 13, 14, 21, 28, 35, 39, 57, 59, 63, 67, ...| {b(4,m)} |
| 12 | 12, 15, 16, 17, 18, 19, 22, 23, 24, 26, 27, ...| {b(1,m)} |
| 13 | 13, 14, 21, 28, 35, 39, 57, 59, 63, 67, 70, ...| {b(4,m)} |
| .. | ... | ... |
| 20 | 20, 43, 56, 96, 113, 131, 135, 156, 196, ... | PS |
| 25 | 21, 33, 37, 38, 40, 47, 48, 65, 76, 79, 83, ...| PS |
...
Example: S(7, 30) = 1.
We observe primitive sequences {b(n,m)} for n = {1, 4, 7, 20, 25, ...}.
(A primitive sequence is a sequence which is not included in another.)
Properties:
(1) S(i, i)= 1 for all i;
(2) S(i, j) = 1 => S(j, i) = 1;
(3) S(i, j) = 1 and S(j, L) = 1 => S(i, L) = 1.
Example:
For n = 1, {P(1,k)} = {18713, 18719, 18731, 18743, 18749};
we choose, for instance, b(1,2) = 3 => for n = 3, {C(3,k)} = {28051, 28057, 28069, 28081, 28087};
S(1,3) = 1 because 18713 + 28087 = 18719 + 28081 = 18731 + 28069 = 18743 + 28057 = 18749 + 28051 = 46800.
In order to find the index L for satisfying the property (3), we choose, for instance, the index b(3,2) = 8 => for n = 8, {P(8,k)} = {97423, 97429, 97441, 97453, 97459} and S(3, 8) = 1 because 28051 + 97459 = 28057 + 97453 = 28069 + 97441 = 28081 + 97429 = 28087 + 97423 = 125510.
Conclusion: S(1, 3) = 1 and S(3, 8) = 1 => S(1, 8) = 1 with 18713 + 97459 = 18719 + 97453 = 18731 + 97441 = 18743 + 97429 = 18749 + 97423 = 116172.

Examples

			The first row is [18713, 18719, 18731, 18743, 18749] because 18713 + 18749 = 18719 + 18743 = 2*18731 = 37462.
The array starts with:
  [18713, 18719, 18731, 18743, 18749]
  [25603, 25609, 25621, 25633, 25639]
  [28051, 28057, 28069, 28081, 28087]
  ...
		

Crossrefs

Programs

  • Maple
    U:=array(1..50,1..5):W:=array(1..2):kk:=0:
    for n from 4 to 10000 do:
       for m from 2 by -1 to 1 do:
          q:=ithprime(n-m)+ithprime(n+m):W[m]:=q:
        od:
        if W[1]=W[2] and W[1]=2*ithprime(n) then
        kk:=kk+1:U[kk,1]:=ithprime(n-2):
        U[kk,2]:=ithprime(n-1):U[kk,3]:=ithprime(n):
        U[kk,4]:=ithprime(n+1):U[kk,5]:=ithprime(n+2):
        else fi:od:print(U):
        for i from 1 to kk do:
         for j from i+1 to kk do:
          s1:=U[i,1]+U[j,5]:
          s2:=U[i,2]+U[j,4]:
          s3:=U[i,3]+U[j,3]:
          s4:=U[i,4]+U[j,2]:
          s5:=U[i,5]+U[j,1]:
         if s1=s2 and s2=s3 and s3=s4 and s4=s5
         then
         printf("%d %d \n",i,j):
         else fi:
         od:
      od: