betaFeedback

WorksMathematics

On Carmichael Numbers Constructed from Consecutive Jacobsthal Numbers

Abstract

We prove that no Carmichael number has the form N=(2mJn+1)(2mJn+1+1)(2mJn+2+1), N=(2^mJ_n+1)(2^mJ_{n+1}+1)(2^mJ_{n+2}+1), where Jn=2n−(−1)n3 J_n=\frac{2^n-(-1)^n}{3} is the Jacobsthal sequence, m,n≥1m,n\ge1, and all three factors are prime. Assuming that such an NN is Carmichael, Korselt's criterion yields divisibility conditions relating consecutive Jacobsthal numbers to powers of two. Combining these conditions with the Binet formula and a finite case analysis for n≤4n\le4, we obtain a contradiction. Consequently, this family contains no Carmichael numbers.

Introduction

Carmichael numbers are composite integers NN satisfying

aN−1≡1(modN)a^{N-1} \equiv1 \pmod{N}

for every integer aa such that gcd⁡(a,N)=1\gcd(a,N)=1. By Korselt’s criterion, a composite integer NN is a Carmichael number if and only if NN is squarefree and

p−1∣N−1p-1 \mid N-1

for every prime divisor pp of $N.

As a particular case of the Lucas sequences of the first kind, the Jacobsthal sequence {Jm}m≥0\{J_m\}_{m\geq0} is defined by J0=0J_0=0, J1=1J_1=1, and the recurrence relation

Jm=Jm−1+2Jm−2,m≥2.J_m=J_{m-1}+2J_{m-2}, \qquad m\geq2.

Its Binet formula is given by

Jm=2m−(−1)m3.J_m=\frac{2^m-(-1)^m}{3}.

where 2 and −1-1 are the roots of the characteristic equation

P(x)=x2−x−2.P(x)=x^2-x-2.

For further information on Jacobsthal numbers and their properties, see [7].

The study of Carmichael numbers has a long history. Alford, Granville, and Pomerance proved that there are infinitely many Carmichael numbers [2]. Subsequent research has investigated their distribution and their occurrence in special arithmetic forms. Banks, Finch, Luca, Pomerance, and Stănică studied Sierpiński and Carmichael numbers [3], while Cilleruelo, Luca, and Pizarro examined Carmichael numbers in sequences of the form (2nk+1)n≥1(2^n k+1)_{n\geq1} [4]. Other contributions have addressed the density of related arithmetic sets [5], the distribution of primes in short intervals [8], and the impossibility of certain types of Carmichael numbers [9]. More recently, Alahmadi and Luca proved that no Carmichael number can be of the form 2np+12^n p+1, where pp is prime [1].

Motivated by these investigations, we consider integers constructed from three consecutive terms of the Jacobsthal sequence.

We study integers of the form

N=(2mJn+1)(2mJn+1+1)(2mJn+2+1),N=(2^mJ_n+1)(2^mJ_{n+1}+1)(2^mJ_{n+2}+1),

where m,n≥1m,n\geq1.

Our main result establishes that, whenever all three factors are prime, their product cannot be a Carmichael number.

Theorem 1. Let

N=(2mJn+1)(2mJn+1+1)(2mJn+2+1),N=(2^mJ_n+1)(2^mJ_{n+1}+1)(2^mJ_{n+2}+1),

where

Jk=2k−(−1)k3,m,n≥1.J_k = \frac{2^k - (-1)^k}{3}, \qquad m,n \ge1.

If all three factors are prime, then NN is not a Carmichael number.

Preliminaries

The Jacobsthal numbers satisfy

J1=1,J2=1,Jk+2=Jk+1+2Jk.J_1 = 1,\qquad J_2 = 1,\qquad J_{k+2} = J_{k+1} + 2J_k.

Lemma 1. For every k≥1k \ge1, JkJ_k is odd.

Proof. Since 2k2^k is even and (−1)k=±1(-1)^k = \pm1, the numerator 2k−(−1)k2^k - (-1)^k is odd. As 3 is odd, JkJ_k is odd. □

Lemma 2. For every k≥1k \ge1,

gcd⁡(Jk,Jk+1)=1andgcd⁡(Jk,Jk+2)=1.\gcd(J_k,J_{k+1}) = 1 \qquad\text{and} \qquad\gcd(J_k,J_{k+2}) = 1.

Proof. The recurrence Jk+2=Jk+1+2JkJ_{k+2} = J_{k+1} + 2J_k gives

gcd⁡(Jk,Jk+1)=gcd⁡(Jk,Jk+1+2Jk)=gcd⁡(Jk,Jk+2).\gcd(J_k,J_{k+1}) = \gcd(J_k,J_{k+1} + 2J_k) = \gcd(J_k,J_{k+2}).

Also,

Jk+1=Jk+2Jk−1,J_{k+1} = J_k + 2J_{k-1},

so

gcd⁡(Jk,Jk+1)=gcd⁡(Jk,2Jk−1).\gcd(J_k,J_{k+1}) = \gcd(J_k,2J_{k-1}).

By Lemma 1, JkJ_k is odd, hence

gcd⁡(Jk,Jk+1)=gcd⁡(Jk,Jk−1).\gcd(J_k,J_{k+1}) = \gcd(J_k,J_{k-1}).

Iterating and using J1=1J_1 = 1, we obtain gcd⁡(Jk,Jk+1)=1\gcd(J_k,J_{k+1}) = 1. Therefore also gcd⁡(Jk,Jk+2)=1\gcd(J_k,J_{k+2}) = 1. □

Theorem 2 (Korselt’s criterion). [6] A composite integer NN is a Carmichael number if and only if NN is squarefree and

p−1∣N−1p - 1 \mid N - 1

for every prime divisor pp of $N.

Proof of Theorem 1

Applying Korselt’s criterion to the three prime factors of NN, we obtain the following necessary divisibility conditions.

Lemma 3. Assume that

N=(2mJt+1)(2mJt+1+1)(2mJt+2+1)N = (2^{m}J_t + 1)(2^{m}J_{t+1} + 1)(2^{m}J_{t+2} + 1)

is a Carmichael number, where m,t≥1m,t \ge1 and the three factors are prime. Then

Jt∣(−1)t2m+2,J_t \mid(-1)^t2^m + 2,
Jt+1∣3−(−1)t2m,J_{t+1} \mid3 - (-1)^t2^m,
Jt+2∣1−(−1)t2m−1.J_{t+2} \mid1 - (-1)^t2^{m-1}.

Proof. Let

p1=2mJt+1,p2=2mJt+1+1,p3=2mJt+2+1.p_1 = 2^mJ_t + 1,\qquad p_2 = 2^mJ_{t+1} + 1,\qquad p_3 = 2^mJ_{t+2} + 1.

Since NN is Carmichael, it is squarefree, and Korselt’s criterion gives

pi−1∣N−1(i=1,2,3).p_i - 1 \mid N - 1 \qquad(i = 1,2,3).

Set

k1=Jt,k2=Jt+1,k3=Jt+2.k_1 = J_t,\qquad k_2 = J_{t+1},\qquad k_3 = J_{t+2}.

Then

N−1=2m(k1+k2+k3)+22m(k1k2+k1k3+k2k3)+23mk1k2k3.N - 1 = 2^m(k_1 + k_2 + k_3) + 2^{2m}(k_1k_2 + k_1k_3 + k_2k_3) + 2^{3m}k_1k_2k_3.

By Lemma 1, each kik_i is odd. Hence gcd⁡(2m,ki)=1\gcd(2^m,k_i) = 1. Therefore ki∣N−1k_i \mid N - 1 is equivalent to the condition obtained after dividing by 2m2^m:

k1∣(k2+k3)+2mk2k3,k2∣(k1+k3)+2mk1k3,k3∣(k1+k2)+2mk1k2.\begin{aligned} k_1 &\mid(k_2 + k_3) + 2^mk_2k_3,\\ k_2 &\mid(k_1 + k_3) + 2^mk_1k_3,\\ k_3 &\mid(k_1 + k_2) + 2^mk_1k_2. \end{aligned}

We use the recurrence

Jt+2=Jt+1+2Jt.J_{t+2} = J_{t+1} + 2J_t.

First condition:

Modulo JtJ_t, we have Jt+2≡Jt+1J_{t+2} \equiv J_{t+1}. Hence

(Jt+1+Jt+2)+2mJt+1Jt+2≡2Jt+1+2mJt+12=Jt+1(2mJt+1+2)(modJt).(J_{t+1} + J_{t+2}) + 2^mJ_{t+1}J_{t+2} \equiv2J_{t+1} + 2^mJ_{t+1}^2 = J_{t+1}(2^mJ_{t+1} + 2) \pmod{J_t}.

Thus

Jt∣Jt+1(2mJt+1+2).J_t \mid J_{t+1}(2^mJ_{t+1} + 2).

By Lemma 2, gcd⁡(Jt,Jt+1)=1\gcd(J_t,J_{t+1}) = 1, so

Jt∣2mJt+1+2.(1)J_t \mid2^mJ_{t+1} + 2. \tag*{(1)}

Second condition:

Modulo Jt+1J_{t+1}, the recurrence gives

2Jt=Jt+2−Jt+1≡Jt+2.2J_t = J_{t+2} - J_{t+1} \equiv J_{t+2}.

Multiplying the second divisibility condition by 22, which preserves divisibility, gives

Jt+1∣2(Jt+Jt+2)+2m+1JtJt+2.J_{t+1} \mid2(J_t + J_{t+2}) + 2^{m+1}J_tJ_{t+2}.

Using 2Jt≡Jt+2(modJt+1)2J_t \equiv J_{t+2} \pmod{J_{t+1}}, we obtain

2(Jt+Jt+2)+2m+1JtJt+2=3Jt+2+2mJt+22=Jt+2(2mJt+2+3)(modJt+1).2(J_t + J_{t+2}) + 2^{m+1}J_tJ_{t+2} = 3J_{t+2} + 2^mJ_{t+2}^2 = J_{t+2}(2^mJ_{t+2} + 3) \pmod{J_{t+1}}.

Hence

Jt+1∣Jt+2(2mJt+2+3).J_{t+1} \mid J_{t+2}(2^mJ_{t+2} + 3).

By Lemma 2, gcd⁡(Jt+1,Jt+2)=1\gcd(J_{t+1},J_{t+2}) = 1, so

Jt+1∣2mJt+2+3.(2)J_{t+1} \mid2^mJ_{t+2} + 3. \tag*{(2)}

Third condition:

Modulo Jt+2J_{t+2}, the recurrence gives

Jt+1=Jt+2−2Jt≡−2Jt.J_{t+1} = J_{t+2} - 2J_t \equiv-2J_t.

Therefore

(Jt+Jt+1)+2mJtJt+1≡Jt−2Jt+2mJt(−2Jt)=−Jt(1+2m+1Jt)(modJt+2).(J_t + J_{t+1}) + 2^mJ_tJ_{t+1} \equiv J_t - 2J_t + 2^mJ_t(-2J_t) = -J_t(1 + 2^{m+1}J_t) \pmod{J_{t+2}}.

Thus

Jt+2∣Jt(2m+1Jt+1).J_{t+2} \mid J_t(2^{m+1}J_t + 1).

By Lemma 2, gcd⁡(Jt,Jt+2)=1\gcd(J_t,J_{t+2}) = 1, so

Jt+2∣2m+1Jt+1.(3)J_{t+2} \mid2^{m+1}J_t + 1. \tag*{(3)}

Binet translation:

Let ε=(−1)t\varepsilon= (-1)^t. Then

3Jt=2t−ε,3Jt+1=2t+1+ε,3Jt+2=2t+2−ε.3J_t = 2^t - \varepsilon,\qquad3J_{t+1} = 2^{t+1} + \varepsilon,\qquad3J_{t+2} = 2^{t+2} - \varepsilon.

From (1),

Jt∣2mJt+1+2.J_t \mid2^mJ_{t+1} + 2.

Multiplying by 3,

2t−ε∣2m+t+1+ε2m+6.2^t - \varepsilon\mid2^{m+t+1} + \varepsilon2^m + 6.

Since 2t≡ε(mod2t−ε)2^t \equiv\varepsilon\pmod{2^t-\varepsilon},

2m+t+1+ε2m+6≡ε2m+1+ε2m+6=3(ε2m+2).2^{m+t+1} + \varepsilon2^m + 6 \equiv\varepsilon2^{m+1} + \varepsilon2^m + 6 = 3(\varepsilon2^m + 2).

Hence

2t−ε∣3(ε2m+2).2^t - \varepsilon\mid3(\varepsilon2^m + 2).

Since 2t−ε=3Jt2^t - \varepsilon= 3J_t, dividing by 3 gives

Jt∣ε2m+2.(4)J_t \mid\varepsilon2^m + 2. \tag*{(4)}

From (2),

Jt+1∣2mJt+2+3.J_{t+1} \mid2^mJ_{t+2} + 3.

Multiplying by 3,

2t+1+ε∣2m+t+2−ε2m+9.2^{t+1} + \varepsilon\mid2^{m+t+2} - \varepsilon2^m + 9.

Since 2t+1≡−ε2^{t+1} \equiv-\varepsilon, we have 2t+2≡−2ε2^{t+2} \equiv-2\varepsilon. Therefore

2m+t+2−ε2m+9≡−3ε2m+9=3(3−ε2m).2^{m+t+2} - \varepsilon2^m + 9 \equiv-3\varepsilon2^m + 9 = 3(3 - \varepsilon2^m).

Thus

2t+1+ε∣3(3−ε2m).2^{t+1} + \varepsilon\mid3(3 - \varepsilon2^m).

Since 2t+1+ε=3Jt+12^{t+1}+\varepsilon=3J_{t+1}, dividing by 3 gives

Jt+1∣3−ε2m.(5)J_{t+1}\mid3-\varepsilon2^m. \tag*{(5)}

From (3),

Jt+2∣2m+1Jt+1.J_{t+2}\mid2^{m+1}J_t+1.

Multiplying by 3,

2t+2−ε∣2m+t+1−ε2m+1+3.2^{t+2}-\varepsilon\mid2^{m+t+1}-\varepsilon2^{m+1}+3.

Since 2t+2≡ε2^{t+2}\equiv\varepsilon, we have

2m+t+1−ε2m+1+3≡ε2m−1−ε2m+1+3=3(1−ε2m−1).2^{m+t+1}-\varepsilon2^{m+1}+3\equiv\varepsilon2^{m-1}-\varepsilon2^{m+1}+3=3(1-\varepsilon2^{m-1}).

Thus

2t+2−ε∣3(1−ε2m−1).2^{t+2}-\varepsilon\mid3(1-\varepsilon2^{m-1}).

Since 2t+2−ε=3Jt+22^{t+2}-\varepsilon=3J_{t+2}, dividing by 3 gives

Jt+2∣1−ε2m−1.(6)J_{t+2}\mid1-\varepsilon2^{m-1}. \tag*{(6)}

This proves the lemma. □

We are now ready to prove Theorem 1.

Suppose, for contradiction, that NN is a Carmichael number. By Lemma 3, with t=nt=n and ε=(−1)n\varepsilon=(-1)^n, we have

Jn∣ε2m+2,(7)J_n\mid\varepsilon2^m+2, \tag*{(7)}
Jn+1∣3−ε2m,(8)J_{n+1}\mid3-\varepsilon2^m, \tag*{(8)}
Jn+2∣1−ε2m−1.(9)J_{n+2}\mid1-\varepsilon2^{m-1}. \tag*{(9)}

We first show that n≤4n\le4. Assume n≥5n\ge5, and write

m=qn+r,0≤r<n.m=qn+r,\qquad0\le r<n.

First divisibility

By the first divisibility condition (7),

Jn∣ε2m+2,ε=(−1)n.J_n\mid\varepsilon2^m+2,\qquad\varepsilon=(-1)^n.

Since

3Jn=2n−ε,3J_n=2^n-\varepsilon,

we obtain

2n−ε∣3(ε2m+2).2^n-\varepsilon\mid3(\varepsilon2^m+2).

Write

m=qn+r,q≥0,0≤r<n,m=qn+r,\qquad q\ge0,\qquad0\le r<n,

and set

σ=εq+1∈{−1,1}.\sigma=\varepsilon^{q+1}\in\{-1,1\}.

As 2n≡ε(mod2n−ε)2^n\equiv\varepsilon\pmod{2^n-\varepsilon}, we have

2m≡εq2r(mod2n−ε).2^m\equiv\varepsilon^q2^r\pmod{2^n-\varepsilon}.

Consequently,

2n−ε∣3(σ2r+2).2^n-\varepsilon\mid3(\sigma2^r+2).

Furthermore,

∣3(σ2r+2)∣≤3(2r+2)≤3(2n−1+2)<2(2n−ε),\begin{aligned} \lvert3(\sigma2^r+2)\rvert&\le3(2^r+2) \\ &\le3(2^{n-1}+2) \\ &< 2(2^n-\varepsilon), \end{aligned}

for n≥5n \ge5. Hence there exists δ∈{−1,0,1}\delta\in\{-1,0,1\} such that

3(σ2r+2)=δ(2n−ε).(10)3(\sigma2^r+2)=\delta(2^n-\varepsilon). \tag*{(10)}

If r=0r=0, then

3(σ+2)∈{3,9},3(\sigma+2)\in\{3,9\},

whereas 2n−ε≥312^n-\varepsilon\ge31. This contradicts (1). Therefore r≥1r\ge1.

We next prove that δ≠±1\delta\ne\pm1. Suppose first that δ=1\delta=1. Then

3(σ2r+2)=2n−ε.3(\sigma2^r+2)=2^n-\varepsilon.

If σ=1\sigma=1, we obtain

3⋅2r+6=2n−ε,3\cdot2^r+6=2^n-\varepsilon,

and thus

2r(2n−r−3)=6+ε.2^r(2^{n-r}-3)=6+\varepsilon.

This is impossible because the left-hand side is even, whereas 6+ε∈{5,7}6+\varepsilon\in\{5,7\} is odd.

If σ=−1\sigma=-1, then

6−3⋅2r=2n−ε,6-3\cdot2^r=2^n-\varepsilon,

which gives

2r(3+2n−r)=6+ε.2^r(3+2^{n-r})=6+\varepsilon.

Again, the left-hand side is even and the right-hand side is odd, a contradiction.

Suppose now that δ=−1\delta=-1. Equation (1) becomes

3(σ2r+2)=−(2n−ε).3(\sigma2^r+2)=-(2^n-\varepsilon).

If σ=1\sigma=1, the left-hand side is positive and the right-hand side is negative, which is impossible.

If σ=−1\sigma=-1, we have

6−3⋅2r=−2n+ε,6-3\cdot2^r=-2^n+\varepsilon,

so

2r(3−2n−r)=6−ε.2^r(3-2^{n-r})=6-\varepsilon.

This is impossible because the left-hand side is even, whereas 6−ε∈{5,7}6-\varepsilon\in\{5,7\} is odd.

Therefore δ=0\delta=0. Equation (1) reduces to

σ2r+2=0.\sigma2^r+2=0.

Since r≥1r \ge1 and σ∈{−1,1}\sigma\in\{-1,1\}, it follows that

r=1,σ=−1.r = 1, \qquad\sigma= -1.

Finally, σ=εq+1=−1\sigma= \varepsilon^{q+1} = -1 implies ε=−1\varepsilon= -1. Hence nn is odd.

Second divisibility

Because nn is odd, ε=−1\varepsilon= -1. Thus (8) becomes

Jn+1∣3+2m.J_{n+1} \mid3 + 2^m.

Since nn is odd,

3Jn+1=2n+1−1.3J_{n+1} = 2^{n+1} - 1.

Multiplying by 3, we get

2n+1−1∣3(3+2m).2^{n+1} - 1 \mid3(3 + 2^m).

Write

m=q′(n+1)+s,0≤s≤n.m = q'(n+1) + s, \qquad0 \le s \le n.

Since 2n+1≡1(mod2n+1−1)2^{n+1} \equiv1 \pmod{2^{n+1}-1}, we have 2m≡2s2^m \equiv2^s. Hence

2n+1−1∣3(3+2s).(11)2^{n+1} - 1 \mid3(3 + 2^s). \tag*{(11)}

Now

0<3(3+2s)≤3(3+2n)<2(2n+1−1)0 < 3(3 + 2^s) \le3(3 + 2^n) < 2(2^{n+1} - 1)

for n≥5n \ge5. Therefore the quotient in (11) is a positive integer strictly smaller than 2, hence equal to 1. Thus

3(3+2s)=2n+1−1,3(3 + 2^s) = 2^{n+1} - 1,

or equivalently,

2n+1=3⋅2s+10.(12)2^{n+1} = 3 \cdot2^s + 10. \tag*{(12)}

We show that (12) is impossible.

  • If s=0s = 0, then 2n+1=132^{n+1} = 13, impossible.

  • If s=1s = 1, then 2n+1=162^{n+1} = 16, so n=3n = 3, contradicting n≥5n \ge5.

  • If s≥2s \ge2, then

    3⋅2s+10≡2(mod4),3 \cdot2^s + 10 \equiv2 \pmod{4},

    whereas

    2n+1≡0(mod4),2^{n+1} \equiv0 \pmod{4},

    impossible.

Therefore n≤4n \le4.

Small cases

It remains to check n=1,2,3,4n = 1, 2, 3, 4.

Case n=1n = 1

We have

J1=1,J2=1.J_1 = 1,\qquad J_2 = 1.

Hence the first two factors of NN are equal:

2mJ1+1=2mJ2+1.2^m J_1 + 1 = 2^m J_2 + 1.

Thus NN is not squarefree, contradicting the fact that every Carmichael number is squarefree.

Case n=2n = 2

Here ε=1\varepsilon= 1, and (8) gives

J3=3∣3−2m.J_3 = 3 \mid3 - 2^m.

But

3−2m=−2m≠0(mod3),3 - 2^m = -2^m \ne0 \pmod{3},

since 3∤2m3 \nmid2^m. Contradiction.

Case n=3n = 3

Here ε=−1\varepsilon= -1. From (7),

J3=3∣−2m+2,J_3 = 3 \mid-2^m + 2,

so

2m≡2(mod3).2^m \equiv2 \pmod{3}.

Hence mm is odd.

From (9),

J5=11∣1+2m−1.J_5 = 11 \mid1 + 2^{m-1}.

Thus

2m−1≡−1(mod11).2^{m-1} \equiv-1 \pmod{11}.

Since ord⁡11(2)=10\operatorname{ord}_{11}(2) = 10 and 25≡−1(mod11)2^5 \equiv-1 \pmod{11}, we get

m−1≡5(mod10).m - 1 \equiv5 \pmod{10}.

Therefore

m≡6(mod10),m \equiv6 \pmod{10},

so mm is even, contradicting that mm is odd.

Case n=4n = 4

Here ε=1\varepsilon= 1. From (7),

J4=5∣2m+2,J_4 = 5 \mid2^m + 2,

so

2m≡3(mod5).2^m \equiv3 \pmod{5}.

Since ord⁡5(2)=4\operatorname{ord}_5(2) = 4 and 23≡3(mod5)2^3 \equiv3 \pmod{5}, we get

m≡3(mod4).m \equiv3 \pmod{4}.

In particular, mm is odd.

From (8),

J5=11∣3−2m,J_5 = 11 \mid3 - 2^m,

so

2m≡3(mod11).2^m \equiv3 \pmod{11}.

Since ord⁡11(2)=10\operatorname{ord}_{11}(2) = 10 and 28≡3(mod11)2^8 \equiv3 \pmod{11}, we get

m≡8(mod10).m \equiv8 \pmod{10}.

In particular, mm is even, contradicting that mm is odd.

All cases lead to a contradiction. Therefore no integer NN of the stated form, with m,n≥1m,n \ge1 and all three factors prime, is a Carmichael number.

Declarations

The author declares that there are no conflicts of interest.

References

  1. [1]A. Alahmadi and F. Luca, “There are no Carmichael numbers of the form 2^n p + 1 with p prime,” Comptes Rendus Mathématique, vol. 360, pp. 1177–1181, 2022.DOI
  2. [2]W. R. Alford, A. Granville, and C. Pomerance, “There are infinitely many Carmichael numbers,” Annals of Mathematics, vol. 139, no. 3, pp. 703–722, 1994.DOI
  3. [3]W. D. Banks, C. Finch, F. Luca, C. Pomerance, and P. Stănică, “Sierpiński and Carmichael numbers,” Transactions of the American Mathematical Society, vol. 367, no. 1, pp. 355–376, 2015.DOI
  4. [4]J. Cilleruelo, F. Luca, and A. Pizarro, “Carmichael numbers in the sequence (2^n k + 1)_{n≥1},” Mathematics of Computation, vol. 85, no. 297, pp. 357–377, 2016.
  5. [5]P. Erdős and A. M. Odlyzko, “On the density of odd integers of the form (p − 1)2^n and related questions,” Journal of Number Theory, vol. 11, pp. 257–263, 1979.DOI
  6. [6]A. Korselt, “Problème chinois,” L’Intermédiaire des Mathématiciens, vol. 6, pp. 142–143, 1899.
  7. [7]Horadam, A.F.: Jacobsthal Representation Numbers. Fib. Quart. 34, 40–54 (1996).DOI
  8. [8]A. Granville, “Primes in intervals of bounded length,” Bulletin of the American Mathematical Society, vol. 52, no. 2, pp. 171–222, 2015.arxiv.org/abs/1410.8400
  9. [9]T. Wright, “The impossibility of certain types of Carmichael numbers,” Integers, vol. 12, no. 5, pp. 951–964, 2012.DOI

Paper details

vlnc:W_6j89wvmn
Version
v1 · latest
Published
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

Not recorded for this paper

View statement
Looking for related papers…
Contents