On a stricter Twin Primes Conjecture and on the Polignacs Conjecture in general Giulio Morpurgo

2025-05-02 0 0 653.91KB 12 页 10玖币
侵权投诉
On a stricter Twin Primes Conjecture, and on
the Polignac’s Conjecture in general
Giulio Morpurgo
March 13, 2023
Abstract
The Polignac’s Conjecture, first formulated by Alphonse de Polignac
in 1849, asserts that, for any even number M, there exist infinitely many
couples of prime numbers P,P+M. When M= 2, this reduces to the
Twin Primes Conjecture. Despite numerical evidence, and many theoret-
ical progresses, the conjecture has resisted a formal proof since.
In the first part of this paper, we investigate a stricter version of the
conjecture, expressed as follows: Let pnbe the n-th prime. Then, there
always exist twin primes between (pn
2)2and p2
n”. To justify this con-
jecture, we formulate a prediction (based on a double-sieve method) for
the number of twin prime pairs in this range, and compare the prediction
with the real results for values of pnup to 6500000. We also analyse what
should happen for higher values of pn.
In the second part, we investigate the validity of the general Polignac’s
Conjecture. We predict the ratio of the number of solutions for any value
of M divided by the number of solutions for M= 2, and explain how this
ratio depends on the factorization of M. We compare the predictions with
the real values for Mup to 3000 (and for the special case 30030) in the
range of from the 1000000-th prime to the 21000000-th prime.
Part 1: The Twin Prime Conjecture
1Introduction
The Twin Prime Conjecture, which is a particular case of the more general
Polignac’s Conjecture [1]) hypothesizes that there exist an infinite number of
pairs k,k+ 2 where both kand k+ 2 are prime (”twin” primes). Despite
numerical evidence (higher and higher twin prime pairs have been found), and
many theoretical progresses [2], [3], the conjecture has resisted a formal proof
since.
In this paper we introduce a stricter version of the conjecture, expressed
as follows: Let pnbe the n-th prime. Then, there always exist twin primes
between (pn2)2and p2
n”. We will also formulate a prediction for the number
of twin prime pairs in this range.
From now on, although many arguments we present apply to the general case,
we will focus on this stricter conjecture, i.e., that there exists at least one pair
of twin primes between (pn2)2and p2
n. One of the reasons to choose such a
restricted range is that, as the range starts with a value higher than p2
n1, all
1
arXiv:2210.15487v2 [math.GM] 10 Mar 2023
the prime numbers up to pn1 can eliminate ”candidates” (defined in the next
section) in the entire range. This would not be true, for instance, if we used a
range from 1 to p2
n; in such a case, piwould only be significant in the portion
of the range above p2
i(introducing complexity in a prediction’s formulation).
We proceed along the following path:
count the number of potential candidates for twin prime pairs between
(pn2)2and p2
n.
use a double-sieve to estimate the probability for a candidate to actually
be a twin prime pair.
derive a prediction for the number of twin primes in the selected range.
compare the prediction with the real result (obtained from a computer
program).
investigate what should happen for larger pnvalues, to show that when
pngrows, also the predicted number of twin pairs grows.
Finally, note that for pn= 3 and pn2 = 1 the solutions 3,5 and 5,7, and for
pn= 5 and pn2 = 3, the solutions 11,13 and 17,19 satisfy the conjecture. In
what follows we will therefore assume pn7.
2Looking for twin primes between (pn2)2and p2
n
2.1 Counting the candidates
Any pair of twin primes can always be expressed as N1, N+ 1, where Nis a
multiple of 6 (3 must divide N, otherwise it would divide one of N1, N+ 1).
To search for twin pairs satisfying our version of the conjecture for a given pn,
we have to examine all the integer numbers Nm= 6m, contained in the range
from (pn2)2to p2
n. These numbers will be our ”candidates”, and for each of
them we will check if both N1 and N+ 1 are prime. We list these candidates
in increasing order, from Nm1(the smallest value above (pn2)2to be divisible
by 6), to Nm last (which is equal to p2
n7, as explained in the next paragraph).
For all pn>3, pnmodulo 3 is either 1 or 2, hence p2
nmodulo 3 is 1. N=p2
n1
is an even number, and its modulo 3 is 0, hence it is divisible by 6. But we
can always exclude this candidate, because N+1 is p2
n, which is not a prime.
Therefore the largest candidate is p2
n- 7.
For a given pn, the size of the range within which we look for twin primes is
p2
n(pn2)2, i.e. 4(pn1). If pn2 is also prime, or if 3 does not di-
vide it, Nm1is equal to (pn2)2+ 5, because also (pn2)2modulo 6 is 1.
So, between (pn2)2and p2
nthere is an ”excluded” region of size 12 (7+5)
that cannot contain useful candidates. The number of candidates will then be
(4(pn1) 12)/6 + 1.
If pn2 is divisible by 3, then (pn2)2modulo 6 = 3; this makes Nm1=
(pn2)2+ 3, and reduces the size of the excluded region to 10. In this case the
number of candidates is (4(pn1) 10)/6 + 1.
In both cases, when pnis large, a good approximation for the number of candi-
dates is
2
n of candidates
=2
3pn(2.1)
2.2 How the values of Nmodulo pidetermine if N1,N+ 1 is a
twin prime pair
For any candidate Nm,Nm–1 and Nm+1 are always smaller than p2
n. Therefore,
for any value of m, we know that Nm–1 and Nm+ 1 are prime if they are not
divisible by any prime from 5 to pn1(we already excluded 2 and 3, by imposing
the condition that Nmis a multiple of 6).
Without loss of generality, let’s call None of these candidates.
We look at the residuals of Nwith respects to each prime from 5 to pn1.
For 5, if Nmodulo 5 is 1, 5 divides N1. If Nmodulo 5 is 4, 5 divides
N+ 1. If Nmodulo 5 is 0, 2, or 3, N1 and N+ 1 are not divisible by
5. So we discard all the Nmvalues for which Nmmodulo 5 is 1 or 4 (two
cases), and we keep the others (three cases).
For 7, we discard all the Nmvalues for which Nmmodulo 7 is 1 or 6, (two
cases) and we keep those for which modulo 7 is 0, 2,3,4, or 5 (five cases).
...
For pn1, we discard all the Nmvalues for which Nmmodulo pn1is 1 or
pn1–1 (two cases), and we keep all the others (pn1–2 cases).
In fact, we are applying a double-sieve to the set of Nmvalues (which form a
finite arithmetic progression).
For pn, the number of primes for which we have to check the set of Nmvalues
is n-3. For example, for p5(= 11), we have to check p4(=7) and p3(=5). For
each of these primes pj, with jgoing from 3 to n1, we discard Nmif the
value of Nmmodulo pjis either 1 or pj1.
2.3 Predicting the number of twin primes
To predict the number of prime twins for a given pn, we assess the effect of
applying our double-sieve on the set of Nmvalues. In particular, we try to
estimate the probability that a candidate is spared by the double sieve.
We start by evaluating the product of terms (pi2)/pi, where the index i
goes from 3 (p3being 5) to n1. We recall that, by choosing only Nmvalues
divisible by 6, we already took into account the effect of p1(=2) and p2(=3).
For a given pn, this product, which we call product pminus2 over p, consists
of n-3 terms.
product pminus2over p =
pn1
Y
primesp3
p2
p=3
5
5
7
9
11
11
13 ··· pn12
pn1
(2.2)
3
摘要:

OnastricterTwinPrimesConjecture,andonthePolignac'sConjectureingeneralGiulioMorpurgoMarch13,2023AbstractThePolignac'sConjecture, rstformulatedbyAlphonsedePolignacin1849,assertsthat,foranyevennumberM,thereexistin nitelymanycouplesofprimenumbersP,P+M.WhenM=2,thisreducestotheTwinPrimesConjecture.Despite...

展开>> 收起<<
On a stricter Twin Primes Conjecture and on the Polignacs Conjecture in general Giulio Morpurgo.pdf

共12页,预览3页

还剩页未读, 继续阅读

声明:本站为文档C2C交易模式,即用户上传的文档直接被用户下载,本站只是中间服务平台,本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间,仅对用户上传内容的表现方式做保护处理,对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私,请立即通知玖贝云文库,我们立即给予删除!
分类:图书资源 价格:10玖币 属性:12 页 大小:653.91KB 格式:PDF 时间:2025-05-02

开通VIP享超值会员特权

  • 多端同步记录
  • 高速下载文档
  • 免费文档工具
  • 分享文档赚钱
  • 每日登录抽奖
  • 优质衍生服务
/ 12
客服
关注