Finite order elements in the integral symplectic group
Kumar Balasubramanian, M. Ram Murty, Karam Deo Shankhadhar

TL;DR
This paper investigates the orders of elements in the integral symplectic group, proving that the set of possible orders is bounded for each dimension and demonstrating that the number and maximum order grow at least exponentially with the dimension.
Contribution
It establishes the boundedness of element orders in symplectic groups and analyzes the exponential growth of their count and maximum order as the dimension increases.
Findings
Set of element orders is bounded for each g
Number of possible orders grows at least exponentially with g
Maximum order of elements also grows at least exponentially
Abstract
For , let be the integral symplectic group and be the set of all positive integers which can occur as the order of an element in . In this paper, we show that is a bounded subset of for all positive integers . We also study the growth of the functions , and and show that they have at least exponential growth.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsFinite Group Theory Research · Graph theory and applications · 14-3-3 protein interactions
Finite order elements in the integral symplectic group
Kumar Balasubramanian
Kumar Balasubramanian
Department of Mathematics
IISER Bhopal
Bhopal, Madhya Pradesh 462066, India
,
M. Ram Murty
M. Ram Murty
Department of Mathematics and Statistics
Queen s University
Kingston, Ontario K7L 3N6, Canada
and
Karam Deo Shankhadhar
Karam Deo Shankhadhar
Department of Mathematics
IISER Bhopal
Bhopal, Madhya Pradesh 462066, India
Research of Kumar Balasubramanian was supported by DST-SERB Grant: YSS/2014/000806.
Research of M. Ram Murty was partially supported by an NSERC Discovery grant.
Abstract
For , let be the integral symplectic group and be the set of all positive integers which can occur as the order of an element in . In this paper, we show that is a bounded subset of for all positive integers . We also study the growth of the functions , and and show that they have at least exponential growth.
1. Introduction
Given a group and a positive integer , it is natural to ask if there exists such that , where denotes the order of the element . In this paper, we make some observations about the collection of positive integers which can occur as orders of elements in . Before we proceed further we set up some notation and briefly mention the problems studied in this paper.
Let be the group of all matrices with integral entries satisfying
[TABLE]
where is the transpose of the matrix and .
Throughout we write , where is a prime and for all . We also assume that the primes are such that for . We write for the number of primes less than or equal to . Also for we let denote the order of . We let denote the Euler’s phi function. It is a well known fact that the function is multiplicative, i.e., if are relatively prime and satisfies for all primes and positive integer (see [2] for a proof). Let
[TABLE]
In this paper we show that is always a bounded subset of for all positive integers . Once we know that is a bounded set, it makes sense to consider the functions , where is the cardinality of and , i.e., is the maximal possible (finite) order in . We show that the functions and have at least exponential growth.
The above problem derives its motivation from analogous problems from the theory of mapping class groups of a surface of genus . We know that given a surface of genus , there is a surjective homomorphism , where is the mapping class group of . It is a well known fact that for () of finite order, we have . Let . The set is a finite set and it makes sense to consider the functions and . It is a well known fact that both these functions and are bounded above by . We refer the reader to [5] for an excellent introduction to the mapping class group and the proofs of some of these facts.
2. Some results we need
In this section we mention a few results that we need in order to prove the main results in this paper.
Proposition 2.1** (Bürgisser).**
*Let , where the primes satisfy for and where for . There exists a matrix of order if and only if
- a)
, if . 2. b)
, if .
Proof.
See corollary 2 in [1] for a proof. ∎
Proposition 2.2** (Dusart).**
*Let be the first primes. For , we have
[TABLE]
Proof.
See theorem 1.14 in [3] for a proof. ∎
Proposition 2.3** (Dusart).**
For , \pi(x)\leq\frac{x}{\log x}\bigg{(}1+\frac{1.2762}{\log x}\bigg{)}. For , \pi(x)\geq\frac{x}{\log x}\bigg{(}1+\frac{1}{\log x}\bigg{)}.
Proof.
See theorem 6.9 in [4] for a proof. ∎
Proposition 2.4** (Dusart).**
*For ,
[TABLE]
*where is the Euler’s constant.
Proof.
See theorem 6.12 in [4] for a proof. ∎
Proposition 2.5** (Rosser).**
For , we have .
Proof.
See theorem 29 in [6] for a proof. ∎
3. Main Results
In this section we prove the main results of this paper. To be more precise, we prove the following.
- a)
is a bounded subset of . 2. b)
has at least exponential growth. 3. c)
has at least exponential growth.
3.0.1. Boundedness of
For each , let . In this section we show that is a bounded subset of .
Let . Suppose for some . This would imply that , which contradicts proposition 2.1. It follows that all primes in the factorization of should be and hence .
Theorem 3.1**.**
For , is a bounded subset of .
Proof.
For , fix and be the set of first primes arranged in increasing order. The prime factorization of any involves primes only from the set . The total number of non-empty subsets of is . Let us denote the collection of these subsets of as . For , let denote the subset of , where is the number of primes in the subset . For a fixed (and hence fixed ), define
[TABLE]
where . The key idea of the proof is to maximize the function considered as a function of the real variables with respect to the inequality constraint . We let denote this maximum. Using the Lagrange multiplier method we see that the function attains the maximum precisely when for all . Under the above condition, the constraint gives us , for any . Now
[TABLE]
From this it follows that for ,
[TABLE]
Therefore, for , we have
[TABLE]
In the above computation, we have used the fact that for , \bigg{(}\frac{2g+1}{x}\bigg{)}^{x} attains the maximum when .
Observing that \displaystyle\prod_{i=1}^{k}\bigg{(}1-\frac{1}{p_{i}}\bigg{)}\geq\frac{1}{2}\frac{2}{3}\bigg{(}\frac{4}{5}\bigg{)}^{\pi(2g+1)-2}, we have
[TABLE]
∎
Corollary 3.2**.**
For , .
Proof.
For , we have . The result follows. ∎
Remark 3.3*.*
Upper bound for for : The bound obtained in theorem 3.1 is an absolute upper bound for . For , we can improve the above upper bound as follows: Using proposition 2.4, we get
[TABLE]
Therefore it follows that for , we have
[TABLE]
3.0.2. Growth of and
In the previous section, we computed an upper bound for the functions and . In this section we show that and have at least exponential growth.
Lemma 3.4**.**
For , we have
[TABLE]
where the sum is over all primes .
Proof.
Let be such that , where denotes the prime number. It follows from proposition 2.2, that for , we have
[TABLE]
∎
Before we proceed further, we set up some notation which we need in the following results.
Let be such that for .
Lemma 3.5**.**
For ,
Proof.
For , we have \pi(y)<\frac{y}{\log(y)}\bigg{(}1+\frac{3}{2\log(y)}\bigg{)} (see proposition 2.3). Using this estimate we get,
[TABLE]
∎
Lemma 3.6**.**
Let and . Then for , we have .
Proof.
By proposition 2.1, it is enough to show that . Using lemma 3.4 and lemma 3.5 , we have
[TABLE]
∎
For , let and be as above. If is any divisor of , then it is easy to see that . Also it is clear that the divisors of are in bijection with the number of subsets of . Since any divisor of is an element in and the number of divisors correspond bijectively with subsets of , it follows that (since number of subsets of ).
We will now show that from which it follows that the function has at least exponential growth.
Theorem 3.7**.**
Let such that . Then for all .
Proof.
From proposition 2.5, we have for all ,
[TABLE]
From this it follows that for all , we have
[TABLE]
∎
Corollary 3.8**.**
Let be as in the above theorem. Then for all .
Proof.
Since , the result follows. ∎
Remark 3.9*.*
For , we can improve the above lower bound to by using proposition 2.3.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1[1] B. Bürgisser, Elements of finite order in symplectic groups , Arch. Math. (Basel) 39 (1982), no. 6, 501–509. MR 690470 (85b:20062)
- 2[2] David M. Burton, Elementary number theory , Allyn and Bacon Inc., Boston, Mass., 1980, Revised printing. MR 567137 (81c:10001 b)
- 3[3] Pierre Dusart, Autour de la fonction qui compte le nombre de nombres premiers , Thesis (1998).
- 4[4] by same author, Estimates of some functions over primes without R.H. , arxiv:1002.0442 v 1.
- 5[5] Benson Farb and Dan Margalit, A primer on mapping class groups , Princeton Mathematical Series, vol. 49, Princeton University Press, Princeton, NJ, 2012. MR 2850125
- 6[6] Barkley Rosser, Explicit bounds for some functions of prime numbers , Amer. J. Math. 63 (1941), 211–232. MR 0003018
