Citation: Kritkhajohn Onphaeng, Prapanpong Pongsriiam. Exact divisibility by powers of the integers in the Lucas sequence of the first kind[J]. AIMS Mathematics, 2020, 5(6): 6739-6748. doi: 10.3934/math.2020433
[1] | Kritkhajohn Onphaeng, Prapanpong Pongsriiam . Exact divisibility by powers of the integers in the Lucas sequences of the first and second kinds. AIMS Mathematics, 2021, 6(11): 11733-11748. doi: 10.3934/math.2021682 |
[2] | Tingting Du, Li Wang . On the power sums problem of bi-periodic Fibonacci and Lucas polynomials. AIMS Mathematics, 2024, 9(4): 7810-7818. doi: 10.3934/math.2024379 |
[3] | Yulu Feng, Min Qiu . Some results on p-adic valuations of Stirling numbers of the second kind. AIMS Mathematics, 2020, 5(5): 4168-4196. doi: 10.3934/math.2020267 |
[4] | Hong Kang . The power sum of balancing polynomials and their divisible properties. AIMS Mathematics, 2024, 9(2): 2684-2694. doi: 10.3934/math.2024133 |
[5] | Adikanda Behera, Prasanta Kumar Ray . Determinantal and permanental representations of convolved (u, v)-Lucas first kind p-polynomials. AIMS Mathematics, 2020, 5(3): 1843-1855. doi: 10.3934/math.2020123 |
[6] | Faik Babadağ, Ali Atasoy . On hyper-dual vectors and angles with Pell, Pell-Lucas numbers. AIMS Mathematics, 2024, 9(11): 30655-30666. doi: 10.3934/math.20241480 |
[7] | Waleed Mohamed Abd-Elhameed, Amr Kamel Amin, Nasr Anwer Zeyada . Some new identities of a type of generalized numbers involving four parameters. AIMS Mathematics, 2022, 7(7): 12962-12980. doi: 10.3934/math.2022718 |
[8] | Waleed Mohamed Abd-Elhameed, Omar Mazen Alqubori . New expressions for certain polynomials combining Fibonacci and Lucas polynomials. AIMS Mathematics, 2025, 10(2): 2930-2957. doi: 10.3934/math.2025136 |
[9] | Ümit Tokeşer, Tuğba Mert, Yakup Dündar . Some properties and Vajda theorems of split dual Fibonacci and split dual Lucas octonions. AIMS Mathematics, 2022, 7(5): 8645-8653. doi: 10.3934/math.2022483 |
[10] | Man Chen, Huaifeng Chen . On ideal matrices whose entries are the generalized $ k- $Horadam numbers. AIMS Mathematics, 2025, 10(2): 1981-1997. doi: 10.3934/math.2025093 |
Throughout this article, let a and b be relatively prime integers and let (Un)n≥0 be the Lucas sequence of the first kind which is defined by the recurrence relation U0=0, U1=1, Un=aUn−1+bUn−2 for n≥2. To avoid triviality, we also assume that b≠0 and α/β is not a root of unity where α and β are the roots of the characteristic polynomial x2−ax−b. In particular, this implies that α≠β and the discriminant D=a2+4b≠0. If a=b=1, then (Un)n≥0 reduces to the sequence of Fibonacci numbers Fn; if a=6 and b=−1, then (Un)n≥0 becomes the sequence of balancing numbers; if a=2 and b=1, then (Un)n≥0 is the sequence of Pell numbers; and many other famous integer sequences are just special cases of the Lucas sequence of the first kind.
The divisibility by powers of the Fibonacci numbers has attracted some attentions because it is used in Matijasevich's solution to Hilbert's 10th problem [7,8,9]. More precisely, Matijasevich shows that
F2n∣Fnmif and only ifFn∣m. | (1.1) |
Hoggatt and Bicknell-Johnson [3] give a generalization of (1.1) by replacing F2n by F3n, and for a general k, they prove that
if Fkn∣m, then Fk+1n∣Fnm. | (1.2) |
Benjamin and Rouse [1], and Seibert and Trojovský [27] also provide a different proof of (1.2). Then the investigation on the exact divisibility results for a subsequence of (Fn)n≥1 begin with the work of Tangboonduangjit et. al [12,29] and is generalized by Onphaeng and Pongsriiam [10]. The most general results in this direction are obtained by Pongsriiam [18] where (1.2) is extended to include the divisibility and exact divisibility for both the Fibonacci and Lucas numbers. Finally, Onphaeng and Pongsriiam [11] have recently given the converse of the results in [18] which completely answers this kind of questions for the Fibonacci and Lucas numbers. Then Panraksa and Tangboonduangjit [13] initiate the investigation on a special subsequence of (Un)n≥0. Patra, Panda, and Khemaratchatakumthorn [14] also obtain the analogue of those results for the balancing and Lucas-balancing numbers. For other related and recent results on Fibonacci, Lucas, balancing, and Lucas-balancing numbers, see for example in [2,4,5,6,15,16,17,19,20,21,22,23,24,25,28] and references there in.
In this article, we extend all results in the literature to the Lucas sequence of the first kind. We organize this article as follows. In Section 2, we give some auxiliary results which are needed later. In Section 3, we give main theorems and some related examples. Remark that the corresponding results for other generalizations of the Fibonacci sequence have not been discovered. For example, the question on exact divisibility by powers of the Tribonacci numbers Tn is wide open, where Tn is given by T0=0, T1=T2=1, and Tk=Tk−1+Tk−2+Tk−3 for k≥3. We leave this problem to the interested readers.
In this section, we recall some well-known results and give some useful lemmas for the reader's convenience. The order (or the rank) of appearance of n∈N in the Lucas sequence (Un)n≥0 is defined as the smallest positive integer m such that n∣Um and is denoted by τ(n). The exact divisibility mk∥n means that mk∣n and mk+1∤n. For a prime p and n∈N, the p-adic valuation of n, denoted by vp(n) is the power of p in the prime factorization of n. We sometimes write the expression such as a∣b∣c=d to mean that a∣b, b∣c, and c=d. We let D=a2+4b be the discriminant and let α and β be the roots of the characteristic polynomial x2−ax−b. It is well known that if D≠0, then the Binet formula Un=αn−βnα−β holds for all n≥0. Next, we recall Sanna's result [26] on the p-adic valuation of Lucas sequence of the first kind.
Lemma 1. [26,Theorem 1.5] Let p be a prime number such that p∤b. Then, for each positive integer n,
vp(Un)={vp(n)+vp(Up)−1if p∣D and p∣n,0if p∣D and p∤n,vp(n)+vp(Upτ(p))−1if p∤D, τ(p)∣n, and p∣n,vp(Uτ(p))if p∤D, τ(p)∣n, and p∤n,0if p∤D and τ(p)∤n. |
In fact, we use Lemma 1 only for p=2, because there is a more suitable version of Lemma 1 when p is odd as given by Panraksa and Tangboonduangjit [13] in their calculation concerning a special subsequence of (Un)n≥0. We state it in the next lemma.
Lemma 2. [13,Lemma 2.3] Let m,n≥1 and p a prime factor of Un such that p∤b. Then, if (i) p is odd, or (ii) p=2 and n is even, or (iii) p=2 and m is odd, we have
vp(Unm)=vp(m)+vp(Un). |
Lemma 3. Let a and b be odd integers. Then, for each positive integer n,
v2(Un)={v2(n)+v2(U6)−1if n≡0(mod6),v2(U3)if n≡3(mod6),0if n≡1,2,4,5(mod6). |
Proof. Since U1=1, U2=a are odd and U3=a2+b is even, we have τ(2)=3. Applying Lemma 1 for p=2, we obtain the desired result.
The next two lemmas are also important tools in proving exact divisibility by Ukn for all n,k∈N.
Lemma 4. [10,Lemma 2.3] Let k, ℓ, m be positive integers, s nonzero integer, and sk∣m. Then sk+ℓ∣(mj)sj for all 1≤j≤m satisfying 2j−ℓ+1>j. In particular, sk+1∣(mj)sj for all 1≤j≤m, and sk+2∣(mj)sj for all 3≤j≤m.
Proof. The statement in [10,Lemma 2.3] is given for s≥1 but it is easy to see that if s≤−1, then we can replace s by −s and every divisibility relation still holds. Therefore this is true for all s≠0.
Lemma 5. Let m,n≥1 and r≥0 be integers. Then
(i) Umn+r=∑mj=0(mj)Ujn(bUn−1)m−jUj+r,
(ii) Umn=∑mj=1(mj)Ujn(bUn−1)m−jUj.
Proof. By Binet's formula, we obtain αn=αUn+bUn−1, βn=βUn+bUn−1, and
Umn+r=αmn+r−βmn+rα−β=1α−β((αUn+bUn−1)mαr−(βUn+bUn−1)mβr)=1α−β(m∑j=0(mj)(αUn)j(bUn−1)m−jαr−m∑j=0(mj)(βUn)j(bUn−1)m−jβr)=1α−βm∑j=o((mj)Ujn(bUn−1)m−j(αj+r−βj+r))=m∑j=0(mj)Ujn(bUn−1)m−jUj+r. |
This proves (ⅰ). Since U0=0, (ⅱ) follows immediately from (ⅰ) by substituting r=0.
Recall that we assume throughout this article that (a,b)=1. This is necessary for the proof of the following lemmas.
Lemma 6. Suppose (a,b)=1. Then (Um,Un)=U(m,n) and in particular (Un,Un+1)=1 for each m,n∈N.
Proof. This is well known.
Lemma 7. Let n≥1 and (a,b)=1. If p is a prime factor of Un, then p∤b. Consequently, (Un,b)=1 for all n≥1.
Proof. Suppose for a contradiction that (a,b)=1, n≥1, p∣Un, and p∣b. We can choose n to be the smallest such integer. Since U1=1, U2=a, we see that n≥3. Since p∣Un=aUn−1+bUn−2 and p∣b, we have p∣aUn−1. By the choice of n, p∤Un−1. So p∣a. Therefore p∣(a,b)=1, a contradiction.
Lemma 8. Let a and b be odd, (a,b)=1, and v2(U6)≥v2(U3)+2. Then v2(U3)=1.
Proof. Since U3=a2+b is even and U6=a(a2+3b)U3, we obtain v2(U3)≥1 and
v2(U6)=v2(U3)+v2(a2+3b). | (2.1) |
If v2(U3)≥2, then 4∣a2+b, and so b≡3(mod4) and (2.1) implies v2(U6)=v2(U3)+1 contradicting v2(U6)≥v2(U3)+2. Thus v2(U3)=1.
We begin with the simplest main result of this paper.
Theorem 9. Let k, m, and n be positive integers. If Ukn∣m, then Uk+1n∣Unm.
Proof. If Ukn∣m, then we obtain by Lemma 4 that, Uk+1n∣(mj)Ujn for all 1≤j≤m, which implies Uk+1n∣Unm, by Lemma 5.
Next, we extend Theorem 9 to include exact divisibility. The proof of Theorem 10 is much longer than that of Theorem 9 since we would like to cover all possible cases. Although many cases can be combined, it is more convenient to state them separately. Recall that for x∈R, the largest integer which is less than or equal to x is denoted by ⌊x⌋.
Theorem 10. Let k,m,n∈N, a,b∈Z, (a,b)=1, n≥2, and Ukn∥m. Then
(i) if a is odd and b is even, then Uk+1n∥Unm;
(ii) if a is even and b is odd, then Uk+1n∥Unm;
(iii) if a and b are odd and n≢3(mod6), then Uk+1n∥Unm;
(iv) if a and b are odd, n≡3(mod6), and Uk+1n2∤m, then Uk+1n∥Unm;
(v) if a and b are odd, n≡3(mod6), Uk+1n2∣m, and 2∥a2+3b, then Uk+1n∥Unm;
(vi) if a and b are odd, n≡3(mod6), Uk+1n2∣m, and 4∣a2+3b, then Uk+t+1n∥Unm, where
t=min({v2(U6)−2}∪{yp−k∣pis an odd prime factor ofUn}) |
and yp=⌊vp(m)vp(Un)⌋ for each odd prime p dividing Un.
Proof. By Theorem 9, we obtain Uk+1n∣Unm. So for (ⅰ) to (ⅴ), it is enough to show that Uk+2n∤Unm. We divide the calculation into several cases.
Case 1. a is odd and b is even. Since U1 and U2 are odd and Ur=aUr−1+bUr−2≡Ur−1(mod2) for r≥3, it follows by induction that Un is odd. From the assumption Ukn∥m, we have Uk+1n∤m, and so there exists a prime p dividing Un such that vp(Uk+1n)>vp(m). Since Un is odd, p is also odd. In addition, p∤b by Lemma 7. So we can apply Lemma 2(ⅰ) to obtain
vp(Unm)=vp(m)+vp(Un)<vp(Uk+1n)+vp(Un)=vp(Uk+2n), |
which implies Uk+2n∤Unm, as required. This proves (ⅰ).
Case 2. a is even and b is odd. Similar to Case 1, we have U1 is odd, U2 is even, Ur≡Ur−2(mod2) for r≥3, and so Un is even if and only if n is even. In addition, there exists a prime p such that p∣Un, vp(Uk+1n)>vp(m), and p∤b. So if 2∤n, then Un is odd, p is odd, and we obtain by Lemma 2(ⅰ) that
vp(Unm)=vp(m)+vp(Un)<vp(Uk+1n)+vp(Un)=vp(Uk+2n), | (3.1) |
which implies Uk+2n∤Unm. If 2∣n, then we can still use either Lemma 2(ⅰ) or Lemma 2(ⅱ) to obtain (3.1), which leads to the same conclusion Uk+2n∤Unm. This proves (ⅱ).
Case 3. a and b are odd. Similar to Case 1, there is a prime p such that p∣Un, vp(Uk+1n)>vp(m), and p∤b.
Case 3.1 n≢3(mod6). If n≡1,2,4,5(mod6), then we obtain by Lemmas 3 and 2, respectively that p is odd and
vp(Unm)=vp(Un)+vp(m)<vp(Un)+vp(Uk+1n)=vp(Uk+2n). | (3.2) |
If n≡0(mod6), then n is even and Lemma 2(ⅰ) or Lemma 2(ⅱ) can still be used to obtain (3.2). In any case, Uk+2n∤Unm. This proves (ⅲ).
Case 3.2 n≡3(mod6) and Uk+1n2∤m. Since Ukn∥m, we can write m=cUkn where c≥1 and Un∤c. By Lemma 4, Uk+2n∣(mj)Ujn for 3≤j≤m. Then we obtain by Lemma 5 that
Unm=Umn≡mUn(bUn−1)m−1+m(m−1)2U2n(bUn−1)m−2a(modUk+2n). |
By Lemma 3, we know that v2(Un)=v2(U3)≥1. Since Uk+1n2∤m and m=cUkn, we see that Un2 does not c. Let d=bUn−1+Un2(m−1)a. By Lemmas 6 and 7, we obtain (Un2,d)=(Un2,bUn−1)=1. Then
Unm≡mUnbm−2Um−2n−1(bUn−1+Un2(m−1)a)≡cUk+1nbm−2Um−2n−1d(modUk+2n). |
By Lemmas 6 and 7, we obtain Uk+2n∣Unm if and only if Un∣cd. But if Un∣cd, then Un2∣cd which implies Un2∣c, a contradiction. So Un∤cd and therefore Uk+2n∤Unm. This proves (ⅳ). To prove (ⅴ) and (ⅵ), we first assume that a and b are odd, n≡3(mod6), and Uk+1n2∣m. (The other condition will be assumed later). Then vp(Uk+1n)≤vp(m) for all odd primes p and v2(Uk+1n)−1≤v2(m). If v2(Uk+1n)−1<v2(m), then v2(Uk+1n)≤v2(m), and so vp(Uk+1n)≤vp(m) for all primes p, which implies Uk+1n∣m contradicting the assumption Ukn∥m. Hence
v2(Uk+1n)−1=v2(m) and vp(Uk+1n)≤vp(m) for every odd prime p | (3.3) |
We now separate the consideration into two cases according to the additional conditions in (ⅴ) and (ⅵ). Observe that v2(a2+3b)=1 is equivalent to 2∥a2+3b.
Case 4. v2(a2+3b)=1. Since U6=a(a2+3b)U3, we obtain v2(U6)=v2(U3)+1. Recall that n≡3(mod6) and Ukn∣m. So n is odd, m is even, and nm≡0(mod6). If Uk+2n∣Unm, then we obtain by Lemma 3 and (3.3) that
v2(Uk+1n)+v2(Un)=v2(Uk+2n)≤v2(Unm)=v2(n)+v2(m)+v2(U6)−1=v2(Uk+1n)−1+v2(U3)=v2(Uk+1n)+v2(Un)−1, |
which is a contradiction. Therefore Uk+2n∤Unm. This proves (v).
Case 5. v2(a2+3b)≥2. Then v2(U6)=v2(U3)+v2(a2+3b)≥v2(U3)+2. By Lemma 8, v2(U3)=1 and so v2(U6)=x+2 where x=v2(a2+3b)−1∈N. For each odd prime p dividing Un, let yp=⌊vp(m)vp(Un)⌋ be the largest integer which is less than or equal to vp(m)vp(Un). Since Ukn∣m, we have yp≥k for all odd p∣Un. Let
t=min({x}∪{yp−k∣p is an odd prime factor of Un}). |
Then t≥0. By Lemma 3 and (3.3), v2(m)=(k+1)v2(U3)−1=k and
v2(Unm)=v2(m)+v2(U6)−1=k+x+1≥k+t+1=v2(Uk+t+1n). | (3.4) |
By the definition of yp, we have vp(m)≥ypvp(Un). So by Lemma 2, if p is an odd prime dividing Un, then
vp(Unm)=vp(m)+vp(Un)≥(yp+1)vp(Un)≥(k+t+1)vp(Un)=vp(Uk+t+1n). | (3.5) |
By (3.4) and (3.5), vp(Unm)≥vp(Uk+t+1n) for all primes p dividing Un. This show that Uk+t+1n∣Unm. It remains to show that Uk+t+2n∤Unm. If t=yp−k for some odd prime p dividing Un, then we recall the definition of yp and apply Lemma 2 to obtain
vp(Unm)=vp(m)+vp(Un)<(yp+2)vp(Un)=(k+t+2)vp(Un)=vp(Uk+t+2n). |
If t=x=v2(U6)−2, then we use Lemma 3 to get
v2(Unm)=v2(m)+v2(U6)−1=k+t+1<v2(Uk+t+2n). |
In any case, Uk+t+2n∤Unm. This completes the proof.
The next example shows that the integer t in Theorem 10(ⅵ) can be any odd positive integer.
Example 11. Let M∈N be given. We show that there are positive integers k, m, n, a, b satisfying the conditions in Theorem 10(ⅵ) with t=M. Choose a=1 and b=(24M−1)/3. Then a and b are odd integers, (a,b)=1, and v2(a2+3b)=4M>2. Next choose any k,n∈N such that n≡3(mod6). Since v2(U6)=v2(U3)+v2(a2+3b)≥v2(U3)+2, we obtain by Lemmas 3 and 8 that v2(Un)=v2(U3)=1 and v2(U6)=4M+1. Since Un≥U3=a2+b>2 and v2(Un)=1, we can write Un=2pa11pa22⋯pass where s≥1, p1,p2,…,ps are distinct odd primes, and a1,a2,…,as are positive integers. Next, choose m=2kpa1(k+M)1pa2(k+M)2⋯pas(k+M)s. Then Ukn∥m and Uk+1n2∣m. Therefore k, m, n, a, b satisfy all the conditions in Theorem 10(ⅵ). Finally, we have
v2(U6)−2=v2(a2+3b)−1=4M−1 |
and yp−k=M for all p∈{p1,p2,…,ps}, and therefore t=min{4M−1,M}=M, as desired.
Next, we prove the converse of Theorem 10.
Theorem 12. k,m,n∈N, a,b∈Z, (a,b)=1, n≥2, and Uk+1n∥Unm. Then
(i) if a is odd and b is even, then Ukn∥m;
(ii) if a is even and b is odd, then Ukn∥m;
(iii) if a and b are odd and n≢3(mod6), then Ukn∥m;
(iv) if a and b are odd, n≡3(mod6), and 2∥a2+3b, then Ukn∥m;
(v) if a and b are odd, n≡3(mod6), 4∣a2+3b, and v2(m)≥k, then Ukn∥m;
(vi) if a and b are odd, n≡3(mod6), 4∣a2+3b, and v2(m)<k, then
m is even, v2(m)≥k+1−v2(a2+3b), and Uv2(m)n∥m. |
Proof. Some parts of the proof are similar to those of Theorem 10, so we skip some details.
Case 1. a is odd and b is even. Similar to Case 1 of Theorem 10, we have Un is odd. For any prime p∣Un, we obtain by Lemma 2 that
vp(Ukn)+vp(Un)=vp(Uk+1n)≤vp(Unm)=vp(Un)+vp(m), | (3.6) |
which implies Ukn∣m. If Uk+1n∣m, then by Theorem 9, we have Uk+2n∣Unm which contradicts Uk+1n∥Unm. Therefore Uk+1n∤m, and thus Ukn∥m.
Case 2. a is even and b is odd. Then Un is even if and only if n is even. So if 2∤n, then for any prime p∣Un, we have p is odd, (3.6) holds, and so Ukn∣m. If 2∣n, then we can still apply Lemma 2(ⅰ) or Lemma 2(ⅱ) to obtain (3.6) and conclude that Ukn∣m. If Uk+1n∣m, then by Theorem 9, we have Uk+2n∣Unm which contradicts Uk+1n∥Unm. So Uk+1n∤m and therefore Ukn∥m.
We now assume throughout that a and b are odd and divide the consideration into four cases according to the additional conditions in (ⅲ) to (ⅵ).
Case 3. n≢3(mod6). If n≡1,2,4,5(mod6), then we apply Lemma 3 to obtain v2(Ukn)=0≤v2(m), and use Lemma 2 to show that for any odd prime p∣Un,
vp(Un)+vp(Ukn)=vp(Uk+1n)≤vp(Unm)=vp(m)+vp(Un). | (3.7) |
If n≡0(mod6), then n is even and we can apply Lemma 2(ⅰ) or Lemma 2(ⅱ) to obtain (3.7) for any prime p∣Un. In any case, we have Ukn∣m. Again, by Theorem 9, we have Uk+1n∤m, and so Ukn∥m. This proves (ⅲ).
Case 4. n≡3(mod6) and 2∥a2+3b. Similar to Case 4 in the proof of Theorem 10 we have v2(U6)=v2(U3)+1. If m is odd, then nm≡3(mod6) and we obtain by Lemma 3 that v2(Unm)=v2(U3)<(k+1)v2(U3)=v2(Uk+1n), which contradicts the assumption Uk+1n∣Unm. So m is even, and thus nm≡0(mod6). By Lemma 3 and the fact that n≡3(mod6) is odd, we obtain v2(m)+v2(U6)−1=v2(Unm)≥v2(Uk+1n)=v2(Ukn)+v2(Un)=v2(Ukn)+v2(U3)=v2(Ukn)+v2(U6)−1, which implies v2(m)≥v2(Ukn). If p is odd and p∣Un, then we apply Lemma 2 to obtain (3.7) Therefore vp(Ukn)≤vp(m) for every prime p dividing Un. Thus Ukn∣m. By Theorem 9, Uk+1n∤m. Hence Ukn∥m. This proves (ⅳ).
Case 5. n≡3(mod6), 4∣a2+3b, and v2(m)≥k. Then U3=a2+b=(a2+3b)−2b≡2(mod4), and so v2(U3)=1. By Lemma 3, we obtain v2(m)≥kv2(U3)=kv2(Un)=v2(Ukn). By Lemma 2, if p is an odd prime dividing Un, then (3.6) holds, and so we conclude that vp(Ukn)≤vp(m) for every prime p dividing Un. Therefore Ukn∣m. By Theorem 9, Uk+1n∤m and so Ukn∥m. This proves (v).
Case 6. n≡3(mod6), 4∣a2+3b, and v2(m)<k. For convenience, let t=v2(m). Similar to Case 4, we have m is even. In addition, v2(U6)=v2(U3)+v2(a2+3b)=1+v2(a2+3b). So k>t≥1 and v2(m)=tv2(U3)=tv2(Un)=v2(Utn). By Lemma 2, if p is odd and p∣Un, then
vp(Un)+vp(Utn)≤vp(Un)+vp(Ukn)=vp(Uk+1n)≤vp(Unm)=vp(m)+vp(Un). |
From the above inequalities, we obtain that vp(Utn)≤vp(m) for every prime p dividing Un. Therefore Utn∣m. If Ut+1n∣m, then we obtain by Lemma 3 that t=v2(m)≥v2(Ut+1n)=t+1, which is false. So Ut+1n∤m. Therefore Utn∥m. From Uk+1n∥Unm, we also obtain k+1=v2(Uk+1n)≤v2(Unm)=v2(m)+v2(U6)−1=v2(m)+v2(a2+3b), which implies v2(m)≥k+1−v2(a2+3b). This completes the proof.
The next example shows that v2(m) in Theorem 12(ⅵ) can be any positive integer in [1,k).
Example 13. Let k≥1 and 1≤M<k be integers. We show that there are m,n,a,b satisfying the conditions in Theorem 12(ⅵ) with v2(m)=M. Choose n∈N and n≡3(mod6).
Case 1. k−M is odd. Choose a=1, b=2k−M+1−13, and m=Ukn2k−M. Then a and b are odd integers, (a,b)=1, and v2(a2+3b)=k−M+1≥2. Since v2(U6)=v2(U3)+v2(a2+3b)≥v2(U3)+2, we obtain by Lemmas 3 and 8 that v2(Un)=v2(U3)=1 and v2(U6)=k−M+2. By Lemma 2, for p>2 and p∣Un we obtain
vp(Unm)=vp(m)+vp(Un)=vp(Ukn)+vp(Un)=vp(Uk+1n). |
By Lemma 3, we have
v2(m)=v2(Ukn)−v2(2k−M)=k−k+M=M |
and
v2(Unm)=v2(m)+v2(U6)−1=M+k−M+2−1=v2(Uk+1n). |
From these, we obtain Uk+1n∥Unm and UMn∥m. Therefore k,m,n,a,b satisfy all the conditions in Theorem 12(ⅵ).
Case 2. k−M is even. Choose a=1, b=5⋅2k−M+1−13, and m=Ukn2k−M. The verification is the same as that in Case 1. So we leave the details to the reader.
Substituting a=b=1 in Theorems 10 and 12, (Un) becomes the Fibonacci sequence (Fn)n≥0 and we obtain our previous results [11,18] as a corollary.
Corollary 14. [18,Theorem 2] and [11,Theorem 3.2] Let n≥3. Then the following statements hold:
(i) if Fkn∥m and n≢3 (mod6), then Fk+1n∥Fnm;
(ii)if Fkn∥m, n≡3 (mod6) and Fk+1n2∤m, then Fk+1n∥Fnm;
(iii)if Fkn∥m, n≡3 (mod6) and Fk+1n2∣m, then Fk+2n∥Fnm;
(iv) if Fk+1n∥Fnm and n≢3(mod6), then Fkn∥m;
(v) if Fk+1n∥Fnm, n≡3(mod6), and 2k∣m, then Fkn∥m;
(vi) if Fk+1n∥Fnm, n≡3(mod6), and 2k∤m, then Fk−1n∥m.
Substituting a=6 and b=−1, in our theorems, (Un) reduces to the sequence (Bn) of balancing numbers and we obtain the results of Patra, Panda, and Khemaratchatakumthorn.
Corollary 15. [14,Theorem 9] For all k≥1 and m,n≥2, we obtain Bkn∥m if and only if Bk+1n∥Bnm.
Similarly by, substituting a=2 and b=1 in our theorems, we obtain the exact divisibility results for the Pell sequence (Pn)n≥0 as follows.
Corollary 16. For all k≥1 and m,n≥2, we obtain Pkn∥m if and only if Pk+1n∥Pnm.
We also plan to solve this problem for the Lucas sequence of the second kind in the future. The answers will appear in our next article.
The authors would like to thank the referees for their comments and suggestions which improve the quality of this article. Kritkhajohn Onphaeng receives a scholarship from Development and Promotion for Science and Technology Talents Project (DPST). Prapanpong Pongsriiam received financial support jointly from the Thailand Research Fund and Faculty of Science Silpakorn University, grant number: RSA5980040.
The authors declare that there is no conflict of interests regarding the publication of this article.
[1] | A. Benjamin and J. Rouse, When does FLm divide Fn? A combinatorial solution, Proceedings of the Eleventh International Conference on Fibonacci Numbers and Their Applications, 194 (2003), 53-58. |
[2] |
P. Cubre and J. Rouse, Divisibility properties of the Fibonacci entry point, Proc. Amer. Math. Soc., 142 (2014), 3771-3785. doi: 10.1090/S0002-9939-2014-12269-6
![]() |
[3] | V. E. Hoggatt, Jr. and M. Bicknell-Johnson, Divisibility by Fibonacci and Lucas squares, Fibonacci Quart., 15 (1977), 3-8. |
[4] | M. Jaidee and P. Pongsriiam, Arithmetic functions of Fibonacci and Lucas numbers, Fibonacci Quart., 57 (2019), 246-254. |
[5] | N. Khaochim and P. Pongsriiam, The general case on the order of appearance of product of consecutive Lucas numbers, Acta Math. Univ. Comenian., 87 (2018), 277-289. |
[6] | N. Khaochim and P. Pongsriiam, On the order of appearance of product of Fibonacci numbers, Contrib. Discrete Math., 13 (2018), 45-62. |
[7] | J. V. Matijasevich, Enumerable Sets are Diophantine, Soviet Math. Dokl., 11 (1970), 354-358. |
[8] |
Y. Matijasevich, My collaboration with Julia Robison, Math. Intelligencer, 14 (1992), 38-45. doi: 10.1007/BF03024472
![]() |
[9] | Y. Matijasevich, Hilbert's Tenth Problem, MIT Press, 1996. |
[10] | K. Onphaeng and P. Pongsriiam, Subsequences and divisibility by powers of the Fibonacci numbers, Fibonacci Quart., 52 (2014), 163-171. |
[11] | K. Onphaeng and P. Pongsriiam, The converse of exact divisibility by powers of the Fibonacci and Lucas numbers, Fibonacci Quart., 56 (2018), 296-302. |
[12] | C. Panraksa, A. Tangboonduangjit, K. Wiboonton, Exact divisibility properties of some subsequence of Fibonacci numbers, Fibonacci Quart., 51 (2013), 307-318. |
[13] | C. Panraksa and A. Tangboonduangjit, p-adic valuation of Lucas iteration sequences, Fibonacci Quart., 56 (2018), 348-353. |
[14] | A. Patra, G. K. Panda, T. Khemaratchatakumthorn, Exact divisibility by powers of the balancing and Lucas-balancing numbers, Fibonacci Quart., accepted. |
[15] | P. Phunphayap and P. Pongsriiam, Explicit formulas for the p-adic valuations of Fibonacci coefficients, J. Integer Seq., 21 (2018), Article 18.3.1. |
[16] |
P. Phunphayap and P. Pongsriiam, Explicit formulas for the p-adic valuations of Fibonomial coefficients II, AIMS Mathematics, 5 (2020), 5685-5699. doi: 10.3934/math.2020364
![]() |
[17] |
P. Pongsriiam, A complete formula for the order of appearance of the powers of Lucas numbers, Commun. Korean Math. Soc., 31 (2016), 447-450. doi: 10.4134/CKMS.c150161
![]() |
[18] | P. Pongsriiam, Exact divisibility by powers of the Fibonacci and Lucas numbers, J. Integer Seq., 17 (2014), Article 14.11.2. |
[19] |
P. Pongsriiam, Factorization of Fibonacci numbers into products of Lucas numbers and related results, JP Journal of Algebra, Number Theory and Applications, 38 (2016), 363-372. doi: 10.17654/NT038040363
![]() |
[20] | P. Pongsriiam, Fibonacci and Lucas numbers associated with Brocard-Ramanujan equation, Commun. Korean Math. Soc., 32 (2017), 511-522. |
[21] | P. Pongsriiam, Fibonacci and Lucas Numbers which are one away from their products, Fibonacci Quart., 55 (2017), 29-40. |
[22] | P. Pongsriiam, Fibonacci and Lucas numbers which have exactly three prime factors and some unique properties of F18 and L18, Fibonacci Quart., 57 (2019), 130-144. |
[23] |
P. Pongsriiam, Integral values of the generating functions of Fibonacci and Lucas numbers, College Math. J., 48 (2017), 97-101. doi: 10.4169/college.math.j.48.2.97
![]() |
[24] | M. K. Sahukar, G. K. Panda, Arithmetic functions of balancing numbers, Fibonacci Quart., 56 (2018), 246-251. |
[25] |
M. K. Sahukar, G. K. Panda, Diophantine equations with balancing-like sequences associated to Brocard-Ramanujan-type problem, Glas Mat., 54 (2019), 255-270. doi: 10.3336/gm.54.2.01
![]() |
[26] | C. Sanna, The p-adic valuation of Lucas sequences, Fibonacci Quart., 54 (2016), 118-124. |
[27] | J. Seibert, P. Trojovský, On divisibility of a relation of the Fibonacci numbers, Int. J. Pure Appl. Math., 46 (2008), 443-448. |
[28] |
C. L. Stewart, On divisors of Lucas and Lehmer numbers, Acta Math., 211 (2013), 291-314. doi: 10.1007/s11511-013-0105-y
![]() |
[29] | A. Tangboonduangjit and K. Wiboonton, Divisibility properties of some subsequences of Fibonacci numbers, East-West J. Math., (2012), 331-336. |
1. | Phakhinkon Napp Phunphayap, Prapanpong Pongsriiam, Divisibility of Fibonomial coefficients in terms of their digital representations and applications, 2022, 7, 2473-6988, 5314, 10.3934/math.2022296 | |
2. | Kritkhajohn Onphaeng, Prapanpong Pongsriiam, Exact divisibility by powers of the integers in the Lucas sequences of the first and second kinds, 2021, 6, 2473-6988, 11733, 10.3934/math.2021682 | |
3. | Prapanpong Pongsriiam, Sums of divisors on arithmetic progressions, 2024, 88, 0031-5303, 443, 10.1007/s10998-023-00566-x | |
4. | Lejla Smajlović, Zenan Šabanac, Lamija Šćeta, On the Hurwitz-Type Zeta Function Associated to the Lucas Sequence, 2022, 60, 0015-0517, 355, 10.1080/00150517.2022.12427451 |